Preventing Illness and Injury in the Workplace

April 1985

NTIS order #PB86-115334 Recommended Citation: Preventing Mness and Injury in the Workplace (Washington, DC: U.S. Congress, Office of Technology Assessment, OTA-H-256, April 1985).

Library of Congress Catalog Card Number 84-601152

For sale by the Superintendent of Documents U.S. Government Printing Office, Washington, DC 20402 Foreword

Congressional interest in work-related disease and injury led to the passage of the Occupational Safety and Health Act in 1970. The Occupational Safety and Health Admin- istration and the National Institute for Occupational Safety and Health were created to administer that Act, which stated an ambitious goal: “to assure so far as possible every working man and woman in the Nation safe and healthful working conditions. ” This report responds to a request from the Chairman of the House Committee on Energy and Commerce and a supporting letter from the Chairman of the Senate Com- mittee on Labor and Human Resources. In this report, OTA examines three main topics: identification of occupational hazards, including the available data on injuries and illnesses; development of control technologies for reducing or eliminating workplace hazards; and the incentives and imperatives that influence decisions to control hazards. Workers, employers, health and safety professionals, and government officials have all contributed to progress in this field. But improvements can still be made. More con- certed effort and better use of existing methods would enhance hazard identification. Further research could improve health and safety control technologies and contribute to their incorporation in U.S. workplaces. Employers’ decisions to control hazards might be fostered by changing the incentives and imperatives that affect those decisions. In preparing this report, OTA staff drew upon the expertise of members of the assessment advisory panel and contractors, as well as other experts in the field of occupational health and safety. Contractors’ reports are available from the National Techni- cal Information Service. In some cases, contractors’ opinions and viewpoints differ from those in this report. Drafts of the final report were reviewed by the advisory panel, chaired by Dr. Morton Corn; executive branch agencies, congressional staff, and other knowledgeable individuals and groups. We are grateful for their assistance. Key OTA staff involved in this assessment were Michael Gough, Karl Kronebusch, Hellen Gel- band, Gwenn Sewell, and Beth Bergman. Denny Dobbin worked on this report while on detail from the National Institute for Occupational Safety and Health.

Director

Ill ADVISORY PANEL FOR PREVENTING ILLNESS AND INJURY IN THE WORKPLACE

Morton Corn, Chair Department of Environmental Health Sciences The Johns Hopkins School of Hygiene and Public Health Baltimore, MD Duane L. Block Samuel Milham, Jr. Ford Motor Co. Washington State Department of Social Dearborne, Ml and Health Services Olympia, WA Richard F. Boggs Organization Resources Counselors, Inc. Kenneth B. Miller Washington, DC Workers Institute for Safety and Health Mark R. Cullen Washington, DC Yale Occupational Medicine Program Ted E. Potter Yale University School of Medicine Shepherd Chemical Co. New Haven, CT Cincinnati, OH Philip E. Enterline Graduate School of Public Health Milan Racic Pittsburgh, PA Allied Industrial Workers Union Milwaukee, WI Melvin W. First Department of Environmental Health Sciences Mark A. Rothstein Harvard School of Public Health West Virginia University College of Law Boston, MA Morgantown, WV Matthew Gillen Marilyn Schule Amalgamated Clothing and Centaur Associates Textile Workers Union Washington, DC New York, NY Melvin Glasser Michael O. Varner Committee for National Health Insurance ASARCO, Inc. Washington, DC Salt Lake City, UT

William J. McCarville James L, Weeks Monsanto Co. United Mineworkers of America St. Louis, MO Washington, DC Wilbur L. Meier, Jr. Roger H. Wingate Dean, School of Engineering Liberty Mutual Insurance Co. (retired) Pennsylvania State University Mirror Lake Post Office, NH University Park, PA

John Mendeloff, Special Consultant University of California—San Diego La Jolla, CA OTA PROJECT STAFF–PREVENTING ILLNESS AND INJURY IN THE WORKPLACE

Roger C. Herdman1 and H. David Banta,2 Assistant Director, OTA Health and Life Sciences Division

Clyde J. Behney, Health Program Manager

Michael Gough, Project Director

Karl Kronebusch, Analyst Denny Dobbin, Analyst3 Hellen Gelband, Analyst Gwenn Sewell, Research Assistant’ Beth Bergman, Research Assistant’

Virginia Cwalina, Administrative Assistant Rebecca Erickson, Secretary/Word Processing Specialist Carol Guntow, Clerical Assistant Brenda Miller, Word Processor/P.C. Specialist

Contractors Linda Starke (Editor), Washington, DC Robert Arndt and Larry Chapman, University of Wisconsin Margit L. Bleecker, The Johns Hopkins Medical Institutes Jacqueline C. Corn, The Johns Hopkins University School of Hygiene and Public Health Robert Goble, Dale Hattis, Mary Ballew, and Deborah Thurston, Clark University and Massachusetts Institute of Technology Philip J. Harter, Washington, DC John L. S. Hickey, Carol H. Rice, and Brian A. Boehlecke, University of North Carolina INFORM, New York, NY Marthe Beckett Kent, Kensington, MD Judith R. Lave and Lester B. Lave, Pittsburgh, PA Douglas MacLean and Mark Sagoff, University of Maryland W. Curtiss Priest, Massachusetts Institute of Technology Jerry L. Purswell and Roger Stephens, University of Oklahoma Springborn Management Services, Inc., Enfield, CT Ruth Ruttenberg, Washington, DC

.— ‘From December 1983. 2Untll August 1983 ‘Detallee from National Inst]tute for Occupational Safety and Health ‘June 1983 -Ju]y 1984 ‘January-June 1983 Contents

Chapter Page l. Summary and Options.....,,.....,...... 3 2. Data on Occupational Injuries and Illnesses ...... 29 3. Health Hazard Identification ...... 41 4. Safety Hazard Identification ...... 67 5. Technologies for Controlling Work-Related Illness ...... 77 6. Technologies for Controlling Work-Related Injury ...... 103 7. Ergonomics and Human Factors...... 123 8. Personal Protective Equipment...... 143 9. Hierarchy of Controls ., ...... 175 10. Training and Education for Preventing Work-Related Injury and illness...... 189 11. A Short History of Private and Public Activities...... 205 12. Governmental Activities Concerning Worker Health and Safety ...... 219 13. Assessment of OSHA and NIOSH Activities ...... ,...... 257 14, Decisionmaking for Occupational Safety and Health: The Uses and Limits of Analysis ...... 275 15. Incentives, Imperatives, and the Decision to Control ...... 297 16. Economic Incentives, Reindustrialization, and Federal Assistance for Occupational Safety and Health ...... 327 17. Preventing Work-Related Injury and Illness in the Future ...... 345 Appendix A.–Supplemental Information on OSHA and NOSH...... 359 Appendix B.—Working Papers...... 381 Appendix C.—Acknowledgments and Health Program Advisory Committee ...... 382 Appendix D.—Glossary of Acronyms and Terms...... 386 References ...... 393 Index ...... 423

vii . . .

Summary and Options

. Contents

...... 4 Associated Deaths, Injuries, and Illnesses ...... 4 Identification of Occupat ional Hazards ...... 6 Technologies to Control Hazards ...... 7 Training and Education ...... 10 Dissemination of Health and Safety Information ...... 11 Incentives and Imperatives That Influence the Decision to Control Hazards . 11 Reindustrialization and Workplace Health and Safety...... 15 options for Controlling Workplace Hazards...... 16 Data and Hazard Identification ...... 16 Improved Control Technologies ...... 18 Education,Training, and Information Dissemination ...... 19 Incentives and Imperatives ...... 21 The Role of OSHA...... 22 Creation of an Occupational Safety and Health Fund...... 24 The Needs of Small Businesses . . .: ...... 25 Assessing Health and Safety Programs...... 26 —

1 ■ Summary and Options

Occupational hazards are not spread evenly: Because the relationship between these hazards Some workplaces, such as banks and offices, have and injuries is usually immediate and direct, rec- few hazards; manufacturing is more dangerous; ognition of the hazards is relatively straightfor- and mining and construction are comparatively ward. The connection between occupational haz- the most hazardous, Certain hazards—some chem- ards and illness is more difficult to pin down. icals and forms of radiation—are concentrated in Although a number of skin and respiratory dis- particular places of work; others—powerful ma- eases and some kinds of poisoning caused by chines and fast-moving machinery—are found metals are definitely associated with work, predominantly in manufacturing and construc- deciding whether other illnesses stem from tion. Each uncontrolled hazard is an opportunity workplace exposures is difficult. for preventing illness or injury. This century has seen some examination of the The exact numbers of workplace-related deaths role of the workplace in injury, illness, and death. and injuries are disputed, but OTA estimates that Motor vehicles used in work are involved in thou- there are about 6,000 deaths annually—about 25 sands of accidents, resulting in many injuries and per working day—due to injuries. Depending on deaths. Construction remains a relatively danger- how injuries are counted, between 2.5 million and ous trade: Powerful earth-moving and erection 11.3 million nonfatal occupational injuries occur machines, high scaffolding, and falling objects are each year. Each working day there are about hazards continually faced by construction work- 10,000 injuries that result in lost work time and ers. Painful and sometimes incapacitating re- about 45,000 that result in restricted activity or petitive-motion disorders are associated with require medical attention. There is so little agree- assembly-line work. Chronic diseases, respiratory ment about the number of workplace-related ill- conditions, and cancers have been linked with ex- nesses that OTA does not take a position on the posures to hazards in a variety of workplaces. controversy about the “correct” number. Most The control of workplace health and safety deaths and injuries occur one at a time or in small hazards can be divided into three steps: hazard numbers in the Nation’s more than 4..5 million identification, development of controls, and the workplaces. decision to control, The first two steps are largely OTA finds that controls for health and safety technical and require specialists. The third step are often developed for specific workplaces and involves generalists, managers, and employers, not disseminated to others. This results in duplica- and may actually occur before hazards are fully tion of effort as employers faced with the same identified and controls are developed. or similar problems are unaware of successful con- The control of illnesses and injuries is not the trols and thus do their own research and devel- sole purview of any particular sector of society. opment. As well as being economically inefficient, Employers and employers’ associations, workers the unshared knowledge about controls may con- and trade unions, universities, and the Federal and tribute to injuries and deaths. State governments have initiated research directed Occupational hazards accompanied the indus- at identifying and controlling hazards, and all trial development of the Nation. In the 19th cen- have participated in decisions to control dangers tury, for instance, advances in manufacturing and that have been identified. Federal involvement has transportation exposed workers to new hazards, increased over the years, and in 1970 Congress including boilers, train couplings, and powered mandated a direct Federal role in all aspects of saws, Scaldings, burns, missing fingers, hands, occupational safety and health, including the set- and arms, and other injuries were the unplanned ting of mandatory nationwide standards for safe consequences of work. and healthful workplaces. 3 4 . Preventing Illness and Injury in the Workplace

The Chairman of the House Committee on occupational health and safety that have been Energy and Commerce requested that the Office developed during this assessment. of Technology Assessment undertake a study of ● Chapters 2, 3, and 4 describe the data avail- technologies to control occupational illnesses and able on workplace deaths, injuries, and ill- injuries. Both his letter and a supporting letter nesses and discuss methods for identifying from the Chairman of the Senate Committee on health and safety hazards. Labor and Human Resources, called for a broad- ● Chapters 5 through 10 consider various con- based study. In addition to requesting examina- trol technologies and current efforts to train tion of the general subject of control technologies, and educate employers, managers, employ- the chairmen asked for evaluation of the avail- ees, and health and safety professionals. able data and systems for collecting data about ● Chapters 11 through 16 discuss the factors work-related deaths, injuries, and illnesses; anal- that are involved in the decision to control ysis of incentives and imperatives that influence hazards. They review the activities of Fed- the decision to control hazards; and a discussion eral agencies, the role of economic analysis of the opportunities for bettering occupational in decisionmaking, the influence of various health and safety as the country enters a period incentives and imperatives on decisionmak- of reindustrialization. Because of the many roles ing, and the opportunities for installing con- given to Federal agencies by the Occupational trols during a period of reindustrialization. Safety and Health (OSH) Act, the activities of the ● Chapter 17 looks at opportunities for pre- Federal Government are important to understand- venting occupational injury and illness in the ing developments and problems in designing, de- future. veloping, and disseminating control technologies, (The contractors’ reports and OTA working in collecting and analyzing data concerning oc- papers prepared for this assessment are available cupational health and safety, and in providing through the National Technical Information Serv- incentives and imperatives for the adoption of ice of the U.S. Department of Commerce. ) controls. The report is organized in five parts: ● This chapter summarizes the findings of the report and presents the options for improving

FINDINGS Occupationally Associated Deaths, data about occupationally related deaths. The Injuries, and Illnesses most reliable estimates are derived from the BLS Annual Survey, although the survey data do not Currently available data are sufficiently ac- include the Nation’s entire work force. OTAs ad- curate and comprehensive to describe the approx- justment of the BLS figures yields an estimate of imate number of occupational injuries and deaths about 6,000 deaths annually from occupational due to such injuries, although these data are still injuries, or about 25 deaths each working day. limited and, in particular, offer little guidance for Occupational fatalities usually occur as isolated prevention. Data about occupational illnesses are events that kill only one or, at most, a few work- far less accurate and comprehensive. ers and attract little publicity. Currently collected data can be used to iden- Deaths tify the most hazardous industries and the types The National Safety Council (NSC, a private of accidents that most commonly result in death. organization) and the U.S. Bureau of Labor Sta- The most dangerous industry is mining, which tistics (BLS) of the Department of Labor compile had 44 fatalities per 100,000 full-time workers in Ch. 1—Summary and Options ● 5

1982. It was followed by the construction, agri- NSC has estimated that for 1980 the total costs culture, and transportation industries (29, 28, and of work injuries amounted to $30.2 billion. 22 fatalities per 100,000 workers, respectively). The leading types of disabling, nonfatal injuries Falling below the all-industry average of 7.4 per are overexertions (largely injuries to the back), 100,000 are manufacturing, wholesale and retail which occur in many industries. Injuries in man- trade, and the service industries, all of which had ufacturing and construction often involve mov- about 4 fatalities per 100,000 workers. The fi- ing machinery and falls. nance, insurance, and real estate industries had the lowest rate of about 2 fatalities per 100,000 Illnesses workers. Three factors generally contribute to incomplete BLS data show that about half of the fatal oc- recording of occupational illnesses: 1 ) many occupational injuries involve motor vehicles, off- cupational diseases are indistinguishable from the-road industrial vehicles, and falls. Comple- nonoccupational illnesses, 2) the occupational menting those findings, an examination of every causes of diseases are often not recognized by on-the-job fatality in the State of Maryland dur- employers and employees, and 3) diseases with ing 1 year found that transportation vehicles, non- long latencies often occur after employment or ex- road vehicles, and gunshots were the leading posure has ceased. Thus the BLS Annual Survey causes of fatal injuries. Truck drivers were the estimate of 106,000 such illnesses in 1983 consists most frequent victims of transportation vehicle mostly of diseases, such as acute dermatitis, that accidents; most gunshot deaths occurred during are easily diagnosed and readily connected with holdups. workplace exposures. Serious diseases—respiratory and necrologic disorders and cancers—are Injuries not generally captured in the BLS records of The OSH Act requires employers to keep rec- workplace-related illnesses. ords of: 1) injuries that caused 1 day or more’s Arguments about the number of occupationally absence from work or “restricted activity” at related diseases may obscure the important fact work, and 2) injuries that required medical atten- that occupational illness is preventable. For in- tion but caused less than a day of missed work. stance, a decade-long debate about the number BLS estimates that in 1983 there were 2.1 million of occupational cancers has been resolved to most “lost-workday” injuries and 2.6 million “medical people’s satisfaction, and it is generally accepted treatment” injuries in the private sector, which that occupational cancers represent something like covers about three-fourths of the work force. In- 5 percent (20,000 annual deaths) or less can- juries to Federal, State, and local government em- of all cer deaths. The more important considerations are ployees may add another 0.4 million lost work- that workers in some industries have borne and day injuries and 0.5 million medical treatment still bear a disproportionate amount of risk and cases. Adding those numbers, there were approx- that, once causes of occupational disease are iden- imately 5.6 million occupationally related injuries. tified, controls can be adopted to reduce risks. The National Center for Health Statistics (NCHS), which uses different definitions for injuries and Some Caveats on Available Data prepares estimates for the entire work force, esti- Accurate data are necessary to know the mag- mates a total of 11.3 million occupationally re- nitude of the workplace health and safety prob- lated injuries in 1981. lem, to target prevention programs, and to assess BLS and NCHS have separately estimated that the progress in controlling illnesses and injuries. workplace-related injuries lead to the loss of, re- Many factors other than control programs, how- spectively, 36.4 million and 60 million to 70 mil- ever, can influence the number of illnesses and lion days of work annually. Projections from injuries. For instance, it has been known for some NCHS data are that workers spend about 44 mil- time that injury rates fall during periods of high lion days in bed because of disability and have unemployment because younger, less skilled work- over 200 million days of restricted activity. The ers are laid off first and there is more time for 6 ● Preventing Illness and Injury in the Workplace maintenance of plant and machinery. OTA finds ing; and that epidemiologic evidence linking ex- that the slowdown in business activity between posures and disease is most convincing. 1980 and 1983 was the most important factor in The traditional role of toxicology has been to the decrease in injury rates during that period. provide information about the mechanisms of dis- Moreover, national injury rates are related to the ease causation. Especially since the late 1960s, level of business activity, going up as business ex- however, toxicology has been used to investigate pands, down as it contracts. The Occupational chemicals in an attempt to predict their effects in Safety and Health Administration (OSHA), on humans. The bulk of the effort has been directed the other hand, points to these declines as a meas- toward identifying chemical carcinogens, but ure of the success of its new programs that empha- some attention is now being directed toward size a cooperative approach between the agency necrologic and reproductive health hazards. and employers. Physicians, both those who specialize in oc- Over the last decade the identification and con- cupational medicine and those in private practice, trol of health hazards, especially of substances sus- have identified many health hazards. As an expected of causing cancer, has received much at- ample, reports of asbestos-associated lung can- tention. Yet the available data about workplace cer cases in the 1930s were an early clue about diseases, even if accurate, would not yet reveal that occupational hazard. More recently, a phy- any effects from a recent reduction in exposures sician noticed an excess of liver cancers in vinyl to carcinogens, Given the long time between ex- chloride workers. His observation led to a very posure to cancer-causing substances and devel- successful effort to reduce exposures to that sub- opment of the disease, years or decades may pass stance. Importantly, physicians speak to work- before cancer rates are affected. An even greater ers, and it is workers who are often the first to problem in relying on figures in this area, how- be aware of hazards. ever, is the inaccuracy of occupational illness data. In 1981, only 234 occupational cancers were Epidemiology, the systematic investigation of reported to workers’ compensation systems in 29 possible associations between exposures and dis- States, which contain about half the Nation’s eases, has confirmed important suspicions about work force. That number can be compared to the work-related illnesses. The now universally ac- 4,000 to 12,000 cancers that are estimated to oc- knowledged case against asbestos, for example, cur from asbestos exposure alone. rests on epidemiologic studies. Positive epidemiologic results showing that an exposure is associ- Identification of Occupational Hazards ated with a disease are the most convincing evidence of a substances toxic effect. Unfortunately, Health the power of epidemiology to detect small risks is limited, and evidence obtained from toxicology Toxicology, occupational medicine, and epi- that a substance is toxic can often be neither con- demiology provide the means for identifying the firmed nor denied by epidemiologic studies. chemical, physical, or biological causes of occupational illnesses. Identifying an association or pos- The National Institute for Occupational Safety sible association between an exposure and disease and Health (NIOSH) conducts various epidemio- often ignites a dispute. Employers, who have in- logic investigations and also makes Health Haz- vestments to protect and are perhaps reluctant to ard Evaluations (HHEs), short-term studies con- accept the idea that employees have been harmed, ducted in response to private and public sector will require more evidence than workers seeking employee or employer requests. HHEs are dean explanation for disease among them. It is clear signed to “determine the toxic effects of chemical, from these controversies that the results of toxico- biological, or physical agents . . . in the work- logic texts often lead to further study rather than place through medical, epidemiologic, and indus- efforts to control a hazard; that physicians’ reports trial hygiene investigations.” HHEs, which be- of associations, depending on the disease and ex- come public reports, have identified and verified posure, may or may not be accepted as convinc- the workplace origins of some illnesses. Ch. l—Summary and Options . 7

Information that could be useful for generat- The first two types of controls, controlling at the ing and examining hypotheses about relations be- source and controlling transmission, are com- tween exposures and health effects is currently col- monly called “engineering controls. ” lected by some industries, but it is not clear that the data are often analyzed and the conclusions Controlling Health Hazards used to decide upon controls. Even data collected by the Federal Government are not used as much Control at the source can be accomplished by as they could be. Useful data are collected by dif- design or modification of process or equipment ferent agencies, but concerns about individual or by substitution of less hazardous materials. privacy have restricted linking data from different This approach offers the greatest opportunist y for sources. Although several committees of govern- prevention, especially when incorporated in the ment scientists have explored ways to remove the initial installation of equipment into a plant. For restrictions, little has been done. example, redesigned dry-cleaning equipment eliminates the need for someone to transfer chemically Safety treated clothes from one machine to another and The idea that “unsafe workers” are a major con- thus prevents worker exposure to that particular tributor to injuries has hampered efforts in injury chemical. Similarly, the very successful control prevention. In the 1920s, a researcher concluded of vinyl chloride exposures involved process that nearly 90 percent of injuries were due to changes that reduced the number of times work- workers’ “unsafe acts” and 10 percent to “unsafe ers’ had to clean reaction vessels, thus lowering conditions. ” Although this ratio of “unsafe acts” exposures during maintenance operations. An il- to “unsafe conditions” is often referred to, it is lustration of control by substitution is the use of not supported by other research. Unfortunately, steel shot instead of sand in abrasive blasting oper- efforts are still made to separate injury causes into ations. This eliminates worker exposure to silica “unsafe acts” and “unsafe conditions, ” while dust, which can lead to silicosis. neglecting the often complex interactions between Ventilation is the method most often used to workers and machines that can lead to injuries. control transmission of health hazards. Local ex- Additional efforts to apply epidemiologic tech- haust ventilation uses an air stream to remove niques to injury analysis should be encouraged. contaminants from work areas. Familiar examples of this include laboratory fume hoods and the Technologies to Control Hazards local exhausts above many kitchen ranges. Simi- lar devices are installed for many types of work- A generalized model of occupational injury and place hazards. General dilution ventilation re- illness is derived from the public health model of duces worker exposure by supplying “fresh” air infectious disease transmission. The model has to the workplace and usually involves the heat- three parts: sources of hazard, transmission of the ing/air-conditioning systems of a plant. These sys- hazard, and workers. Methods for controlling tems can be modified to increase the amount of workplace illnesses and injuries are intuitively airflow and thus dilute airborne hazards. Recent simple. Health and safety professionals generally changes in building ventilation aimed at con- follow a “hierarchy of controls” approach that is serving energy use air recirculation techniques. related to this general model: If not done properly, some of these altered systems may increase worker exposures. first, containing the hazard—whether it is a substance or some physical, electrical, or me- Other ways to control transmission include iso- chanical energy —at its source; lating the source and preventing toxic materials second, interfering with transmission of the from becoming airborne. Worker exposures to as- hazard to the worker; and bestos and cotton textile dust were reduced by third, providing the worker with protective enclosing dusty carding machines. More gener- clothing and equipment. ally, automating processes and locating equipment 8 Preventing Illness and Injury in the Workplace

in rooms or buildings away from workers reduces falling objects, the collapse of buildings and exposures. A common technique for preventing trenchs, and workers falling or being crushed by dust from becoming airborne is to spray water machinery and equipment. on the material. Manufacturing involves the application of Finally, control at the worker may include ad- energy to materials to shape them into usable ministrative procedures, work practices, and the products. Woodworking, hot metalworking, and use of personal protective equipment. Adminis- cold metalworking are three processes with sig- trative procedures include worker rotation among nificant hazards. A number of traditional control jobs to reduce the number of people exposed full- techniques are available to reduce these hazards. time, as well as the scheduling of jobs and proc- These include the installation of guards to pre- esses that generate hazards at times when few vent hands and fingers from getting caught in ma- workers are present. Work practices are simply chinery and material from flying out and strik- job procedures and methods that are designed to ing workers. Machinery and processes can also reduce hazards. Personal protective equipment, be redesigned to minimize the need for workers such as hard hats and respirators, are described to place their arms or legs near moving machin- in more detail below. ery parts. Interlocks and two-hand controls are available to prevent machine operation when Controlling Injury Hazards guards have been removed or when a worker’s hands are inside the machine. Finally, personal Workplace injuries generally involve transfers protective equipment, such as face shields and of energy, and thus controlling them could be ap- goggles, are available to reduce the risk of injury proached as a task of preventing the transmission from flying objects. of energy. For example, mechanical energy can be transmitted to stationary workers by falling Fires and explosions cause deaths and injuries objects, such as bricks on a construction site. Con- as well as large economic losses. For both those trols could involve securing the bricks so they do reasons, efforts to prevent them have resulted in not fall, setting up overhead barriers to prevent careful attention to good plant design, control of any falling bricks from striking workers, and the ignition sources of fires, installation of warn- issuing hard hats to the workers. ing alarms and systems to extinguish fires at early stages, and plans for quick evacuation of burn- However, the terminology, analytical methods, ing buildings. and procedures of safety professionals have usually differed from those used in controlling health Finally, employers often set up formal injury hazards. Safety engineers have tended to use prevention programs. Because management has codes, standards, and models of “good practice” the primary responsibility for prevention of work- that are oriented around particular topics: fire pre- related injury and illness, a successful program vention, electrical safety, machinery design, plant must start with a strong commitment from man- layout, etc. agement. The stronger the commitment at the top, the greater the likelihood of success. Typical man- Recommended “good practice” often involves agement efforts to prevent work-related injury in- common sense and the personal experience of clude establishing company policies, incorporat- safety engineers, with relatively little scientific ing injury prevention into plant design, carefully analysis, systematic data collection, epidemiol- investigating reported injuries to identify hazards, ogy, or experimental research. In addition, as keeping accurate and comprehensive records, mentioned, the view that many or most injuries placing workers in appropriate jobs, and conduct- are due to so-called unsafe acts has interfered with ing safet training for workers and supervisors. the incorporation of injury controls into the de- y sign of plant and equipment. Personal Protective Equipment Nevertheless, injury prevention can be incorporated into the design of workplaces. Controls Hard hats, safety shoes, and protective eyewear can be introduced to prevent electrical shocks, are examples of personal protective equipment. . —— —

Ch, l—Summary and Options . 9 —

In many cases, especially construction, there are The standards, often not met, are themselves no practical engineering substitutes for such de- limited. Plastic lenses are tested for resistance to vices. Respirators and hearing protectors guard penetration, whereas glass lenses are not; NIOSH against hazardous dusts, fumes, vapors, and loud commented that it would expect most glass lenses noises. to fail the test if it were required. Similarly, hard hats are tested for resistance only to vertical im- Obviously, personal protective devices must be pacts. No tests are required for off-center impacts. worn to be effective, and their successful use requires both that equipment and instruction be The only type of personal protective equipment made available and that use be properly super- that is tested and certified by the Federal Gov- vised. There is evidence that safety equipment, ernment is respirators. NIOSH certifies respirators such as hard hats and safety toe shoes, is worn using laboratory test methods that, in some cases, when required by employers. Because of the clear were developed years ago. Efforts to update the connection between those devices and injury pre- certification requirements have progressed slowly vention, it is reasonably easy to argue that safety and may take years to complete. equipment will provide immediate benefits. On The few tests carried out in the workplace under the other hand, the value of wearing a respirator conditions of normal use show that respirators to protect against a disease that may not mani- often do not provide the level of protection ex- fest itself for several years or a few decades may pected from the laboratory measurements. The not be as immediately clear. In addition, most poorer performance may be due to inappropriate respirators and hearing protectors are uncomfor- use or maintenance or overestimation of perform- table and hamper communication, and respirators ance based on laboratory tests. make breathing more labored. Finally, there is a body of engineering knowledge that can be ap- The Environmental Protection Agency (EPA) plied to reducing or eliminating the need to use formerly required that hearing protectors be rated respirators and hearing protectors. for effectiveness. The effectiveness of probably all hearing protectors is overrated because of sys- Unlike engineering controls that are often tai- tematic errors in tests conducted to comply with lored to a particular workplace, personal protec- the EPA requirements. tive equipment is manufactured and sold for use at many diverse sites. Some Federal regulations Hierarchy of Controls require the use of personal protective equipment. There are, however, no Federal standards for its Using engineering solutions to control hazards performance (with the exception of respirators); at their source or in the pathway of transmission instead, the Government relies on manufacturers is more reliable and less burdensome to the worker to produce and sell equipment that meets stand- than personal protective equipment, Once in- ards adopted by voluntary standards organiza- stalled, these controls work day after day with tions, The American National Standards Institute minimum routine intervention beyond mainte- (ANSI) is the source of most such standards. nance and monitoring. In the mid-1970s, NIOSH purchased samples In keeping with the tenets of professional orga- of personal protective equipment and tested them nizations such as the American Industrial Hygiene against ANSI standards. Many items failed. For Association and the American Conference of instance, the lenses on 11 of 24 models of a type Governmental Industrial Hygienists (ACGIH), of protective eyewear splintered or shattered when OSHA had permitted use of personal protective subjected to the ANSI test for impact resistance; equipment only when engineering controls were only 4 of 19 models of hard hats passed all the not feasible, not capable of reducing exposures ANSI-specified tests. These results are especially to the required levels, or in the process of being discouraging because the employer who purchases designed and installed. This approach has been the equipment and the workers who depend on criticized by some employers who argue that they it must rely on the manufacturer to produce a should be able to substitute personal protective good product. equipment more freely for other types of controls. 10 . preventing Illness and Injury in the Workplace

In 1983, OSHA announced its intention to re- funding to $5.8 million in both fiscal year 1982 consider its policy of relying first on engineering and 1983. Decreases in Federal funding will prob- controls for airborne health hazards. In a more ably result in fewer degree and training programs. specific action, OSHA in 1984 proposed a reduc- Large companies with successful programs em- tion in permissible exposure to asbestos. The phasize that commitment to control of work- agency proposes to allow the use of respirators related injury and illness must begin with top to attain the new standard. If this regulation be- management. Despite that widely held opinion, comes final, it will almost certainly provide an little attention is given to injury and illness pre- argument for primary reliance on respirators in vention in the education of business administra- meeting other standards. Such a change must con- tion students. One attempt at building manager sider the poor results attained with those devices. awareness, the NIOSH and OSHA Project Minerva, OTA’s analysis of the literature indicates that is sponsoring a series of meetings for business edu- respirators provide the protection that is claimed cation teachers to introduce them to the concepts for them only in workplaces that provide scru- of occupational health and safety and to find ways pulous supervision of maintenance and use. Those of bringing those concepts into their courses. conditions are rare. To turn away from the hierarchy of control without careful verification of the The Nation’s engineering schools annually train levels of protection afforded by personal protec- nearly 400,000 students. The accrediting organiza- tive devices is likely to increase exposures to tion for engineering schools requires, in theory, health hazards. that engineering design courses consider health and safety. These courses, in which students learn the fundamentals of designing plants and proc- Training and Education esses, would appear to be especially appropriate for learning about the control of hazards. The OTA finds that programs to educate workers topic apparently receives little attention, however. and health and safety professionals have rarely At one major engineering school, for example, been evaluated, and that evaluation is necessary most faculty interviewed agreed that safety was to know about their effect. Although not sup- important, but few hours were devoted to teach- ported by evaluation, there appears to be general ing it, agreement that they succeed. Evaluation is diffi- The engineering curriculum, which prepares cult because of the difficulty in determining what students for a professional license at the bac- causes changes in illness and injury rates. Never- calaureate level, is acknowledged as one of the theless, such efforts should be encouraged. most course-laden programs at a university. Al- NIOSH funds Educational Resource Centers though adding instruction in health and safety is (ERCs). The centers are to educate occupational attractive, it is difficult to fit this instruction into health and safety professionals, to offer continu- the existing engineering curriculum. ing education programs, to conduct research, and Educating physicians about occupational medi- to provide regional consultation services. They cine falls into two categories: general education are required to provide interdisciplinary educa- about occupational disease and injury, and spe- tion with contributions from occupational medi- cialized training for practitioners of occupational cine and nursing, industrial hygiene, and safety medicine. Improvements can be made in both engineering. These requirements set ERCs apart areas. It is generally accepted that physicians in from other health and safety professional educa- general practice fail to recognize the impact of oc- tion programs. cupational factors on the health of their patients. In 1981, with $12,1 million funding, the ERCs This poor recognition stems from an orientation graduated over 780 professionals from degree pro- toward occupational health that is minimal at best grams and trained over 12,000 professionals in and often nonexistent in U.S. medical schools. To continuing education programs. Since then the accommodate classes on occupational medicine President’s budget has proposed cutting the ERC in the crowded medical curriculum would require funding to zero, and Congress partially restored that other subjects be dropped, a difficult task. Ch. 1—Summary and Options ● 11

Postgraduate, specialty training in occupational Incentives and Imperatives That Influence medicine has traditionally been subsumed under the Decision to Control Hazards preventive medicine, centered in schools of public health, and sometimes criticized for providing Increased knowledge of hazards and improved too little clinical experience. The criticism is be- controls provide the means for protecting health ing muted by the requirement of clinical experi- and safety, but a decision to adopt the controls ence for board certification of physicians in oc- is necessary for them to have any impact at all. cupational medicine. In fact, the first and most important act in workplace health and safety may be the decision to con- Dissemination of Health and trol hazards. At least seven factors may motivate Safety Information the decision to control: ● employers’ enlightened self-interest, Much information about hazards and controls ● information on hazards and controls, is available from NIOSH, OSHA, health and 0 financial and tax incentives, safety professionals’ associations, and the trade ● workers’ compensation and insurance, literature. The volume and unorganized state of ● tort liability, this information impede its use. As a start in mak- ● employees’ rights and collective bargaining, ing information more accessible, NIOSH and the and National Library of Medicine have established ● regulation. computerized data systems that provide useful information for evaluating workplace hazards. The first six factors can be viewed as incentives; the last, regulation, is an imperative. OTA finds OSHA has a consultation program that is de- that while each of these may motivate a decision signed to provide assistance in hazard identifica- to control, the influence of all the incentives and tion and control to employers, especially those the imperative is limited. who run small businesses. It is a potentially valuable tool for the dissemination of information Employers’ Enlightened Self-Interest and may be a way to improve job conditions that is less adversarial than the enforcement of regu- An important motivating factor behind volun- lations through inspections. In fiscal year 1983, tary employer actions concerning health and safe- OSHA funded consultations in more than 30,000 ty is enlightened self-interest and concern for other workplaces. humans, Reinforcing such voluntary efforts are reductions in the costs of absenteeism, workers’ To date, OSHA has not evaluated the effects compensation, or medical care when the decision of the consultation program on injuries and ex- to control hazards results in fewer injuries and ill- posures. Although OSHA urges that employers nesses. share the consultants’ information with employees, this step is not required, and it is probable OSHA has recently instituted several programs that workers are sometimes not informed. Some to encourage voluntary hazard control. In sev- observers have expressed concern that funding for eral States, employers are exempt from programed consultative visits diverts resources from OSHA inspections if they receive an OSHA consultation inspection activities. and thereafter correct all serious hazards. OSHA’s Voluntary Protection Program also encourages Letting workers know about occupational voluntary actions. hazards is now facilitated and required by State and local “right-to-know” laws and the recently Some employers also participate in cooperative issued OSHA rule concerning the labeling of con- efforts to develop voluntary standards that draw tainers of hazardous chemical substances. Such upon the collective information and expertise of information is valuable not only to workers but companies in a particular industry, trade associa- also to owners and managers who purchase chem- tion, or standard-setting organization. Voluntary icals for their businesses and to doctors and other standards are an important source of information health professionals, for employers, workers, and Government agen- 12 . preventing Illness and Injury in the Workplace — cies, and they may move all companies that agree tion of injuries and illnesses is a secondary goal. to them to a common performance level. Although workers’ compensation programs have probably had a positive effect on injury experi- However, voluntary standards are also criti- ence, empirical evidence for this has been diffi- cized for being insufficiently protective. Suggested cult to gather. remedies include having additional input from labor unions and the public when standards are Four factors limit the incentives that workers drafted. Yet unions and public interest organiza- compensation can provide for control of hazards. tions frequently lack the staff and other resources First, all insurance schemes spread losses; there- to participate in voluntary standard-setting. Fur- fore, the insurance function of workers’ compen- thermore, they often do not want to participate sation means that employers who cause injuries because of a history of industry domination and do not bear their full costs, unless they are self- the unenforceable nature of voluntary standards insured or pay premiums that are directly tied to their injury and illness experience. Second, ben- The pressures of the competitive marketplace efit levels represent less than the full social costs substantially limit the ability of individual of injuries and illnesses. Third, some injuries and employers to improve employee health and safety most illnesses are not compensated because a through voluntary actions. If a company devotes claim is never filed, or they are inadequately com- its resources to improving workplace conditions pensated because the claim is delayed or denied. but its competitors do not, the firm can find itself To the extent that these factors reduce the frac- at a disadvantage. tion of the costs of injuries and illnesses that are borne by employers, they reduce incentives for Information on Hazards and Controls prevention. Changes in the system that lead to Timely and accurate data are necessary for a greater proportion of the costs of illnesses and making decisions, and both Government and pri- injuries being paid by employers would enhance vate organizations provide information about the prevention incentives of workers’ compen- hazards and controls. Although necessary, infor- sation, mation alone may have little influence on decisions to control. Tort Liability The last decade has seen spectacular growth in Financial and Tax Incentives the number of cases in which workers sued firms Reducing the costs of purchasing needed equip- that manufactured machinery and other products ment and technology can encourage employers purchased by employers for workers’ use, Such to improve health and safety. Four kinds of tax suits are generally filed against “third parties, ” and assistance programs might be useful for oc- manufacturers and suppliers, because workers’ cupational health and safety: investment tax compensation programs bar suits against employers. credits, accelerated depreciation allowances, di- Tort liability has received special attention be- rect subsidies, and Government loan programs. cause of the number of third-party lawsuits Funds from a Government loan program for small against suppliers of asbestos. If the number of businesses have been used for occupational health third-party suits increases, and if they are success- and safety investments, but that program was ful for hazards other than asbestos, they may be- abolished in 1981. The other three mechanisms come an important incentive for prevention. Even have been used to encourage investments in equip- so, the number of cases may be limited because ment for environmental protection, but not for it is difficult to produce the degree of proof re- health and safety controls. quired by courts in cases of occupational disease.

Workers’ Compensation and Insurance Employees’ Rights and Collective Bargaining The primary goal of workers’ compensation The OSH Act created opportunities for worker programs is to pay injured workers’ medical ex- participation in health and safety activities. The penses and to compensate for lost wages. Preven- act provided that workers can: —.

Ch. I—Summary and Options ● 13

● request OSHA inspections, ployer representatives, especially health and safety ● participate in the conduct of an OSHA in- professionals, accept the need for them. Most of spection, the standards set by OSHA, however, have been ● participate in any of the stages of a pro- criticized by nearly all parties, but for different ceeding before the Occupational Safety and reasons. Labor groups judge the standards as in- Health Review Commission, sufficient to protect health. Business groups see ● contest the “reasonableness” of the abatement them as nit-picking, excessively stringent, unnec- date set by OSHA, essary, inflexible, and too costly. The criticisms ● participate in standards development and the from both sides in part reflect fundamental dif- issuance of variances, and ferences concerning the desirable level and type ● request a Health Hazard Evaluation from of Federal intervention in this area. NIOSH. OSHA’s Standard-setting Criteria.–Since 1981, In addition, the act established a mechanism to OSHA has used four criteria for decisions on protect employees from job discrimination for health standards. First, it determines if the haz- having exercised any of these rights. This provi- ard in question poses a “significant risk” and war- sion prevents discrimination against employees rants regulatory intervention. Second, the agency who refuse work that presents an imminent dan- determines whether regulatory action can reduce ger of injury, although it probably does not ex- the risk. If so, OSHA develops a standard to re- tend to employees who refuse work that the duce the risk “to the extent feasible, ” considering worker thinks presents a health hazard. both technological and economic feasibility. Final- ly, OSHA analyzes the cost effectiveness of vari- Collective bargaining is particularly useful for ous options to determine which will achieve its establishment-specific implementation of controls chosen goal most efficiently. and for monitoring employer actions. It is severely limited because only about 20 percent of the work All OSHA regulatory actions are now reviewed force is unionized and because not all unions have by the Office of Management and Budget (OMB) sufficient staff expertise in industrial hygiene, in- under Executive Order 12291, which, to the ex- jury prevention, or occupational medicine. More- tent permitted by law, requires regulatory agen- over, health and safety provisions must compete cies to demonstrate that their proposed and final with other bargaining issues for attention and re- regulations pass a cost-benefit test. Generally sources. Some people object to collective bargain- speaking, the results of the OMB review and ing for injury and illness prevention because they agency responses have not been made public, believe that health and safety on the job ought making it difficult to determine if OSHA decisions to be an employee right, not subject to nego- have been altered by OMB’s cost-benefit review. tiation. OSHA’s Record of Standard Setting.—There At least 82 percent of union contracts contain is dissatisfaction about the length of time OSHA at least one clause related to health and safety takes to develop, propose, and promulgate new according to data collected by the Bureau of Na- standards or revisions of existing standards. In its tional Affairs, Unions can encourage members’ first 13 years, through December 1984, OSHA participation in health and safety activities, par- issued only 11 new or revised health standards ticipate in worker education in hazard recogni- concerning 24 specific chemical substances and tion, provide or have access to technical exper- one standard covering exposure to noise. Stand- tise, and establish mechanisms for dispute ards for two of the substances and noise were resolution between employer and union, overturned by the courts. Twenty-six new or revised safety standards were completed. In addition, broader regulations concerning employee ac- OSHA Regulation cess to records, a “generic” policy concerning the Mandatory Federal regulations are an impera- regulation of carcinogens (under which no sub- tive for the adoption of controls. Labor represent- stance has been regulated), and the labeling stand- atives insist on mandatory standards and em- ard were issued. 14 . Preventing Illness and Injury in the Workplace

In part because of the slowness of OSHA stand- The current administration has implemented a ard writing, many OSHA standards seriously lag number of changes in inspection and enforcement. behind recommendations and voluntary standards A new type of inspection examines only the em- issued by professional societies and voluntary ployer-maintained injury records if the firm’s standards organizations. injury rate is below the national average for manufacturing. In addition, the number and percent- Most current OSHA health standards are based age of inspections with “serious” and “willful” on the exposure limits published by the Ameri- violations has fallen, and the total dollar amount can Conference of Governmental Industrial Hy- of proposed penalties has been reduced substan- gienists in 1968, and most standards rely safety tially. on American National Standards Institute publications of the 1960s. Those standards were Other new OSHA policies encourage area di- adopted in 1971 under a section of the OSH Act rectors and employers to “settle” citations by re- which gave OSHA authority to adopt established ducing or eliminating penalties in return for an Federal standards and national consensus stand- employer’s promise to abate the hazard and to ards. ACGIH annually updates its limits, in- comply with OSHA regulations. These changes cluding standards for additional chemicals, and may decrease the contentiousness of some OSHA often recommends stricter exposure limits. OSHA proceedings. On the other hand, they may have often does not follow suit. further reduced an already weak regulatory effort. OTA finds that ACGIH exposure limits and NIOSH recommendations, overall, are stricter OSHA’s Effects. —The impact OSHA can have than the OSHA standards. In addition, the 1968 on injury rates is constrained by the small size of ACGIH list covered nearly 400 substances. The the OSHA regulatory effort, which can inspect current ACGIH list covers over 600 substances, less than 4 percent of the Nation’s workplaces but OSHA’S list—with a handful of additions— annually. Most evaluations have searched for remains essentially the same as ACGIH’s 1968 list. OSHA’s effects on total injury rates, which could A mechanism for timely and efficient OSHA con- be masking the success of the agency in prevent- sideration of new ACGIH exposure limits and ing certain specific types of injuries as well as pos- NIOSH recommendations might prevent OSHA sible differences in the effectiveness of each area from lagging behind professional recommen- office of OSHA and of the 25 jurisdictions oper- dations. ating “State programs. ”

OSHA Inspection and Enforcement.—A regu- The research results are mixed. Several re- latory strategy will succeed only if the agency’s searchers have found favorable, but generally enforcement efforts have adequate resources. For small, changes, implying that OSHA activities most establishments the probability of a routine have reduced injury rates. Other researchers have OSHA inspection is very low (there are about not found any significant correlation between 160,000 inspections annually in a total of 4,600,000 OSHA activity and workplace injuries. workplaces). Most inspections take place in manufacturing or construction. But even in those in- Currently, OSHA points to decreasing injury dustries, on average, a plant or site will be in- rates for 1980 through 1983 as evidence that the spected only rarely. For example, the typical agency’s new regulatory approaches are paying manufacturing establishment can expect to be in- off. However, changes at OSHA could not fully spected once every 6 years. In addition, even if account for the declines, for they were not insti- an employer is found not to be in compliance, the tuted until 1981, more than a year after the drop fines issued by OSHA are small, especially when in rates began. Moreover, as indicated earlier, the compared with the costs of many types of con- economic recession, including increased unem- trols. For example, the average proposed penalty ployment and a shift away from “smokestack infor employer violations that threaten “death or dustries, ” is the most important factor behind this serious physical harm” is less than $200, decline. Ch. l—Summary and Options ● 15

There is some evidence that several OSHA reg- tional health and safety. Although quantitative ulations have had a positive effect on exposures estimates are lacking, to some extent the reported to health hazards. The best known case is vinyl declines in injury rates dating from early in this chloride. Exposures declined dramatically after the century may be due to the installation of mod- issuance of a more stringent OSHA standard. ern, safer plants and equipment. A second factor Substantial declines have taken place in asbestos may be general shifts in employment away from exposure levels, perhaps due to OSHA efforts, but industries and operations with greater hazards. more likely due to fears of tort liability suits. Similarly, in some particular cases, exposures to health hazards have declined because of increased A study commissioned for this assessment mechanization, but it is not clear whether ex- found substantial decreases in lead levels in posures to health hazards overall have decreased, workplace air and even more marked reductions remained the same, or increased. in lead levels in employees’ blood in the years since OSHA’s new lead standard was promulgated. Thus, through the process of industrial change Another study commissioned by OTA found sub- health and safety can improve without anyone’s stantial decreases in exposures to cotton dust fol- explicitly “intending” it. In addition, some changes lowing the introduction of a new agency stand- in the workplace have taken place because of ard. The number of workers exposed to levels employers’ desires to minimize the threat of fire above the new, tighter exposure limit for cotton and explosion or to reduce the downtime of plant dust has been halved in the short time since the or equipment. Some changes that lower the threat standard came into effect. Several textile mills ap- of property damage or “down time” also reduce pear to be in complete compliance, while others exposures to toxic agents or the risk of injury. expect to be in the near future. If this country is entering a period of reindus- Measuring OSHA’s impact is difficult. To detect trialization, many opportunities will be available the impact of a small Federal program on some- to improve health and safety. As new plants are thing as large as the Nation’s entire work force built, employers may take advantage of oppor- might be asking too much, Regarding workplace- tunities to install controls as part of initial con- related illnesses, even if the data were reliable, it struction, when they can be put in at lowest cost. is too early to expect that OSHA regulations If the Government provides economic incentives would have much impact on occupational disease. or financial assistance to firms as they modern- On the positive side, however, OSHA standards ize, it can consider methods to encourage the infor vinyl chloride, cotton dust, and lead have stallation of controls. Some of the incentives clearly reduced workplace exposures. Further- already mentioned—including tax breaks and di- more, increased productivity accompanied com- rect financial subsidies, as well as possibly tim- pliance with both the vinyl chloride and cotton ing new OSHA regulations to coincide with in- dust regulations. dustrywide changes —might be useful during a period of reindustrialization. Reindustrialization and Workplace Health and Safety

Over the years, the process of industrial change and renewal has led to improvements in occupa- 16 . Preventing Illness and Injury in the Workplace

OPTIONS FOR CONTROLLING WORKPLACE HAZARDS

Data and Hazard Identification The National Death Index. -Information on death certificates is essential to any epidemiologic Increasing the Usefulness of study investigating causes of death. When sup- Current Data Systems plied with someone’s name and date of birth or Identifying workplace health and safety hazards Social Security number, the National Death In- is the first step in reducing occupational morbid- dex (NDI), a service of the NCHS, can tell epi- ity and mortality. Certain changes in Federal data demiologists whether that person has died and collection efforts can make epidemiologic investi- where the death certificate is located. Until NDI gations. was established, epidemiologists had to contact every department of vital statistics to locate the Mortality Surveys.--One nationwide study of death certificate. Quite simply, the NDI reduces death certificates to examine associations between the number of such inquiries from more than so industry and occupation and mortality was done to 1, although each certificate must still be ob- in the 1950s. Since then, epidemiologists in the tained from the office of vital statistics that holds States of Washington and Rhode Island have con- it. ducted statewide studies. These are valuable not only for identifying high risks associated with The NDI would be more useful if it supplied some types of work but also for indicating occupa- all the information encoded upon death certifi- tions and industries that do not present high risks. cates. Were it to be modified to do that, epide- Yet statewide mortality analyses cannot be rep- miologists could obtain all vital information for resentative of the Nation as a whole and lack the mortality studies from a single inquiry. The ben- statistical power that would be present in an anal- efits of such a change would be to speed up studies ysis of data for the whole country. Nationwide and reduce their costs. Such a change would in- mortality analyses would provide important leads crease the work load at NCHS associated with the for further study to pin down associations be- NDI and require some system whereby State de- tween work and various causes of death, as well partments of vital statistics could still receive rev- as valuable information about hazards in occupa- enue for supplying information. tions that are scattered across the country, e.g., Option 2: The National Death Index could be carpenters or butchers. modified so that all information collected on Currently, NIOSH and the National Center for death certificates can be made available from it. Health Statistics provide instruction and assist- Addresses From Internal Revenue Service ance to a few States that are conducting mortal- Records.—Epidemiologic studies frequently re- ity surveys. A collaborative effort between NCHS quire investigators to interview subjects of the and NIOSH would probably best accomplish the study or their families. One impediment to such nationwide task of carrying out mortality surveys. efforts is the difficulty of locating people. Inter- nal Revenue Service (IRS) records are a reason- Option I: Congress could provide funds and per- ably complete source of recent addresses, but only sonnel for an NCHS/NIOSH collaborative ef- NIOSH and some other Federal agency scientists fort to produce accurate coding of industry and occupation information on death certificates. and persons working on contract to NIOSH can That information could then be used to pro- obtain addresses from IRS. duce mortality analyses for occupations and in- There is some confusion about who can use this dustries either in: “NIOSH window” and clarifications about this s the few States that are establishing mortal- are needed. In addition, it may be desirable to ity surveys or allow a wider spectrum of researchers to obtain ● nationally. addresses from IRS. Any expansion of the win- Ch. l—Summary and Options 17 dow would require safeguards so that addresses has begun detailed investigations of a small num- received this way are used only for epidemiologic ber of fatal injuries. In addition, the BLS has ob- studies. tained information on some types of nonfatal injuries through questionnaires completed by in- Option 3: Congress could direct the Federal agen- jured workers. cies to define clearly who can obtain IRS-held addresses and create procedures to allow a Option 5: Congress might direct OSHA, N1OSH, wider spectrum of researchers to obtain ad- and BLS to devote additional resources to in- dresses from the IRS for use in locating persons vestigating fatal and nonfatal injuries, with the for epidemiologic studies. objective of developing information useful for preventive efforts. Linking Federal Data Systems to Facilitate Epi- demiologic Studies. –The records systems of the BLS Annual Survey Census Bureau, Social Security Administration, Veterans’ Administration, OSHA, and NIOSH The BLS Annual Survey, which collects infor- could be linked together to provide information mation from employer-maintained logs of injuries about medical conditions, work history and ex- and illnesses, is the best source of information posures, and the current address in a single file. about occupational fatalities and nonfatal injuries. Such a link could improve epidemiologic studies; Since 1981, employer-maintained injury records but it increases also the possibility of invasion of and the results of the BLS Annual Survey have a person’s privacy. The option suggested here is been used to grant exemptions from OSHA in- intentionally vague because of the delicate bal- spections. Because of this reliance on the data, ance between improving our capacity to under- assessing the reliability of the responses would be stand disease and protecting citizens’ privacy. prudent. Although epidemiologists are convinced of the In the early 1970s, BLS conducted onsite evalu- value of linking together data systems, few efforts ations of a sample of employer responses to the to do so have been approved. “On Occupational Annual Survey to verify their accuracy. This Cancer Estimation, ” the recent report of the De- “Quality Assurance Program” has not been partment of Health and Human Services’ Com- repeated since 1976. mittee to Coordinate Environmental and Related Option 6: Congress could direct OSHA and BLS Programs, suggests some options for linking data to conduct a new “Quality Assurance Program” systems. to determine the accuracy of employer-main- Option 4: Congress could encourage considera- tained injury records. tion of various proposals to link together Fed- eral data systems for use in epidemiology. Toxicology The Federal Government, especially through Injury Investigation the National Toxicology Program and the Na- OSHA investigates 1,500 to 2,000 accidents in- tional Center for Toxicological Research, is sup- volving fatalities or five or more hospitalizations porting large-scale efforts to improve toxicology each year. Unfortunately, little attention has been so that results will be more predictive of human paid to using the collected information to prevent effects and more readily accepted in the setting future accidents, and for many years it had only of standards. The Toxic Substances Control Act gathered dust in OSHA’s files. The agency has mandates the submission to EPA of information conducted some limited analyses of these inves- about “substantial risks” to human health that are tigations, has initiated a small effort to distrib- identified by companies. This section of the stat- ute summaries of construction accidents to labor ute and the act’s requirement that companies no- unions, trade associations, and other organiza- tify EPA of available toxicologic information tions, and is developing a new data system to pro- before new chemicals are introduced into com- vide information collected during accident inves- merce are important in protecting workers’ health. tigations. Complementing these activities, NIOSH This assessment suggests no particular options re- 18 . Preventing Illness and Injury in the Workplace garding toxicology, but it draws attention to the A third priority area for research in worker importance of those programs. health and safety is new technologies. The hazard potential of new processes, procedures, equip Improved Control Technologies ment, and techniques needs to be evaluated, and attention paid to the development of controls. NIOSH-Supported Research on Controls Early attention to hazards will provide health ben- Provided with sufficient resources, NIOSH, efits to workers; moreover, lower costs are asso- through vigorous grant and contract programs, ciated with building hazard control into the tech- could encourage the application of the techniques nologies at first rather than having to retrofit later. of engineering, epidemiology, ergonomics (hu- Option 7: Congress could expand support of man-factors engineering), industrial hygiene, and NIOSH research and demonstrations in control other disciplines to the development of innovative technologies, using both NIOSH staff and re- hazard control methods. Increasing NOSH’S resources as well as grants, cooperative agree- search in control technologies even five- or ten- ments, and contracts. This expanded research fold need not require a proportional increase in and demonstration effort could be directed at NIOSH staff. Most of the research could be done four different areas: in private sector and university laboratories. ● fundamental engineering research, directed at finding generalizable principles for health Increased research and development of control and safety controls; technologies would enable the Federal Govern- ● applied research and demonstration projects ment to provide new information to improve concerning improved engineering control safety and health. It might also improve cooper- techniques; ation between the Federal Government and occu- research in improved personal protective pational health and safety professionals in the pri- equipment; vate sector. Research in control technologies efforts to track emerging industries and new represented only 12.8 percent or about $7.4 mil- plants, evaluate hazards, and offer advice to lion of the NIOSH budget in fiscal year 1983. firms engaged in new technologies, Three general research areas could benefit from additional funding: engineering controls, personal protective equipment, and new production tech- Private Sector Research niques. Much research, especially that oriented towards Workplaces built some years ago with little at- the development of controls for particular installa- tention to occupational health and safety often tions, is conducted by employers, equipment man- incorporated few injury and illness controls when ufacturers, and the insurance industry. Their ef- they were constructed. Instead, controls—if they forts have produced successful solutions for many are used at all—are added later as retrofits. Addi- occupational health and safety problems, To the tional work is needed to develop general principles extent that they have the appropriate expertise, for designing controls into plant and equipment employers and manufacturers should be eligible in order to increase effectiveness and minimize for NIOSH research grants and contracts. interference with production. Another goal could be improved control at reduced cost. Lower costs Certification and Regulation of might reduce employer and manufacturer resistance Personal Protective Equipment to the installation of controls and the burdens of regulatory standard setting and enforcement. All types of personal protective equipment pose similar questions: What kinds of tests for effec- Research on personal protective equipment tiveness should be required? When should the tests should develop reliable and comfortable devices be done—before or after marketing? Who should and methods to assess efficacy in “real-world” conduct the tests? How should test results be used? conditions. Research on respirators is particularly needed, but investigations of other kinds of per- Option 8: N1OSH could be given resources to sonal protective equipment are also important. establish procedures to test and certify some or —.

Ch. l—Summary and Options ● 19

all types of personal protective equipment; the Workers and Supervisors agency might: Since 1978, the OSHA New Directions Program Option 8A: establish a program of premarket testing that includes, at a minimum, appro- has awarded grants to labor unions, trade asso- priate laboratory evaluation of personal pro- ciations, universities, and nonprofit institutions tective equipment, and, as soon as possible, for developing and conducting training and education programs. The focus has been worker train- testing and certification to reflect real workplace situations; ing, although a number of New Directions grantees have also trained supervisors and produced Option 8B: conduct postmarked surveillance to educational materials useful to supervisors, man- collect reports of equipment failure and defects, and to investigate those reports; or agers, and workers. Option 8C: explore alternative arrangements The New Directions Program, although not so for both premarket testing and postmarked well evaluated as it could be, is seen as a success surveillance of persona! protective equipment. by many health and safety professionals. Cur- rently the grants that were supported by transfer These arrangements could include different of money from the National Cancer Institute to combinations of self-testing and certification by OSHA are being evaluated, and other assessments manufacturers, testing and certification by in- could be encouraged. The characteristics of good dependent parties, “spot-check” testing by and poor projects should be publicized and the NIOSH, and full-scale testing by NIOSH. funding level of the New Directions Program, Although employers and employees rely on ef- which has been decreased, could be reconsidered. fectiveness labeling to select equipment, those Aiding local or industry-centered organizations figures often overstate actual effectiveness. For ex- to find solutions to local problems provides a di- ample, OSHA instructs its compliance officers to rect approach to health and safety problems, assume that hearing protectors provide only so Option 10: Congress might increase Federal sup- percent of the laboratory-measured protection. port for occupational health and safety educa- Option 9: Congress could provide OSHA and tion and training, possibly through the New NIOSH with resources to develop, collect, and Directions Program, by: disseminate information on “real-world” effec- ● involving unions, workers’ organizations, tiveness of currently available personal protec- and trade and educational associations in tive equipment. education and training through the provision of grants to develop informational and educational materials and to hire professional Education, Training, and health and safety staff; Information Dissemination ● supporting education of supervisors and managers in occupational health and safety The Federal Government provides in-house through programs directed at providing training to its own and other employees and grant educational materials to employees. support for various education and training programs. One example of an in-house activity is the The Federal Mine Safety and Health Act of 1977 OSHA Training Institute, which provides con- requires mine operators to provide certain speci- tinuing education to Federal and State OSHA staff fied amounts of safety training to workers. Some (principally inspectors) and, to a limited extent, OSHA standards require employers to provide to individuals from the private sector. Grant-sup- worker training concerning specific hazards, but ported activities are split: OSHA has concentrated there are no requirements for instruction or train- on employee and employer training, whereas ing in most occupations. However, in the absence NIOSH has general responsibility for the educa- of any requirement, some employers provide tion of professionals. health and safety training. Furthermore, some col- 20 ● Preventing Illness and Injury in the Workplace lective bargaining agreements specify that all Some starts have been made (and some aban- workers receive some training and that advanced doned) to extend information about safety and instruction be provided to worker members of health to physicians, engineers, and business health and safety committees. administration educators and students. The De- Option 11: Employers might be required to pro- partment of Health and Human Services sup- vide a certain minimum level of health and ported some efforts to educate physicians in envi- safety training to their entire work force. ronmental and occupational health in several medical schools in the late 1970s, but funds are no longer available. NIOSH has sponsored a series Health and Safety Professionals of workshops on the topic of engineering educa- NIOSH training grants to universities support tion concerning health and safety, two activities: academic programs that train in- Option 13: Congress could provide support for dividuals in a single specialty, and Educational and encourage: Resource Centers, which provide complete pro- ● introducing occupational medicine in medi- grams. Many health and safety experts believe cal school course work; that these funds have been well spent, increasing ● introducing or expanding occupational safety the number of graduated professionals and en- and health into engineering school curricula; hancing the abilities of professionals through con- ● introducing or expanding classes about oc- tinuing education. On the other hand, there has cupational health and safety in business been only limited evaluation of these programs administration courses, or the actual impact that the increased number For example, grants through NIOSH or the Na- of professionals has had on worker health and tional Science Foundation might be used to de- safety. velop training modules for integration into ex- Funding for these programs has been reduced isting courses. in recent years and the current administration has proposed complete elimination of the ERCs. Cut- Expanded Information Services backs in Federal funding in this area are likely to reduce the number of trained professionals. The OSHA consultation program, which was instituted to provide health and safety evaluations Option 12: Congress could continue to fund train- to businesses, especially small firms, is a relatively ing of occupational health and safety specialists, popular program. One possibility is to expand the including the Educational Resource Centers, through the NIOSH training grants program. program to provide consultation to a greater number of employers as well as to employees and unions. This would require funding, as well as the Engineers, Physicians, and Managers creation of procedures for providing these services. The disciplines of engineering and medicine Option 14: Congress could expand the OSHA have a marked impact on occupational health and consultation program by: safety even though most practitioners in these ● providing increased funding for OSHA con- disciplines are not specialists in workplace health sultation; and safety. Neither general-practice physicians ● directing OSHA to explore methods to en- nor engineers receive significant instruction about courage employers to share this information occupational hazards and controls. For physi- with employees and their representatives; cians, the prime need is training to recognize the ● expanding the consultation program to pro- impact of occupational exposures on health. Engi- vide this service to employees and unions. neers need to understand the nature of occupational hazards and to learn the fundamental design Insurance Industry Research techniques useful for prevention of work-related illness and injury. In addition, managers play an Representatives of insurance companies visit important continuing role in decisionmaking more plants than OSHA is able to inspect, and about health and safety. many employers, especially small firms that lack .— .

Ch. l—Summary and Options ● 21 — full-time health and safety personnel, rely on the Workers’ Compensation Programs and advice of their insurers’ loss-control specialists. Tort Liability The establishment of an institute similar to the Workers’ compensation programs, adminis- Insurance Institute for Highway Safety might fa- tered by the States, have been credited with con- cilitate the dissemination of industry-collected in- tributing to the prevention of injuries and illnesses. formation on occupational health and safety. No There is reason to believe that this may be true option is proposed because there would be no Fed- for occupational injuries, although data to sup- eral role in such an institute, port this conclusion are limited. For illnesses, data are even more sparse, and the programs offer Computerized Information Systems fewer incentives for prevention of illness than for There are many useful collections of data. For injuries. instance, NIOSH produces and collects informa- Most potential lawsuits by employees against tion about toxicity, assessment of control tech- their employers for occupational injuries and ill- nologies, and product testing; OSHA collects in- nesses are barred by the statutes that created the formation about hazards and controls during State workers’ compensation systems. It has been inspections, consultations, and courses. Combin- suggested that this prohibition be eliminated in ing information from some or all of these sources some circumstances, but this would involve ma- would produce a data system for use by designers, jor changes in workers’ compensation laws. engineers, workers, employers, and health and safety specialists. Users could be charged for serv- Congress is considering legislation to provide ices to defray expenses and possibly to make the compensation for the victims of asbestos-related service self-supporting. disease. This proposal is a response to perceived problems in both the workers’ compensation and Option 15: The Federal Government could pro- tort liability systems. vide grant or contract money to apply computer technology to the collection and dissem- Prevention should be considered in any changes ination of occupational health and safety in compensation. In general, a compensation sys- information. tem should be designed to encourage prevention. If Federal revenues are used to supplement occupational disease compensation funds, the Fed- Incentives and Imperatives eral contribution might be accompanied by a requirement that companies take concrete steps to Voluntary Implementation of Controls prevent future cases of disease—a suggestion that Voluntary employer efforts to improve health is admittedly hard to implement, Since OSHA and safety are very important. OSHA has initi- would almost certainly already be regulating any ated a program to encourage such efforts, and hazard important enough to require a Federal con- NIOSH has often persuaded employers to contribution to compensation, it is not clear what ad- trol hazards that are not currently subject to ditional requirements might be imposed on com- OSHA regulations. Attempts to encourage “vol- panies that benefit from compensation legislation. untary protection” must be kept in balance, how- But it is also important to consider carefully any ever, with the standard-setting and enforcement changes in either compensation or tort liability required by the OSH Act. to guard against changes that might weaken in- Option 16: Congress could direct OSHA and centives for prevention. NIOSH to increase the attention devoted to encouraging voluntary efforts and to publicize the Labor-Management Committees

firms that have exemplary programs in health Labor-management health and safety commit- and safety, tees exist in many U.S. workplaces, in both union 22 ● Preventing Illness and Injury in the Workplace

and nonunion shops. They offer an avenue for against adopting the recommendations, but it sharing and conveying information about hazards would have to respond or be subject to suit. and controls. OSHA currently supports the for- Without changing the current system of stand- mation of joint committees in companies that ard setting, OSHA inspectors could provide in- participate in the OSHA Voluntary Protection formation to both employers and workers con- Programs. cerning professional recommendations. Although Option 17: Congress could encourage the forma- it would not be legally binding, employers might tion of labor-management committees by: take actions based on this information. ● directing that OSHA expand its Voluntary Option 20: OSHA inspectors could be directed Protection Program; ● to provide information (to employers and em- increasing OSHA funding for training, con- ployees) on current NIOSH recommendations, sultation, and other technical assistance to professional organizations’ recommended ex- workplaces with labor-management com- posure limits (such as ACGIH’s, which are up- mittees. dated annually), and voluntary standards whenever these recommendations and stand- The Role of OSHA ards would affect the hazards found in particular workplaces, Updating OSHA Regulations

It is well known that OSHA lags behind pro- Standard Setting fessional health and safety organizations and consensus standards in responding to new informa- Despite the fact that it did not succeed, a re- tion about health hazards. The agency upgrades cent effort to negotiate a standard for benzene its regulations through the same time-consuming should provide much valuable information about rulemaking procedure it uses to promulgate new the feasibility of using negotiations in standard regulations, and changes are often opposed. setting. OSHA considers NIOSH recommendations Option 21: Congress might encourage OSHA to about exposure limits, but has taken few regula- study possible procedures for negotiation dur- tory actions based on NIOSH criteria documents. ing standards development and implementa- Requiring an OSHA response to NIOSH recom- tion. These procedures will have to assure the mendations would ensure that the regulatory adequate representation of all affected parties. agency considered the research agency’s findings, In the setting of health standards, OSHA has but making it mandatory for OSHA to regulate generally moved substance-by-substance, Each on the basis of NIOSH recommendations might proposed health standard can be, and most have not be useful. The Mine Safety and Health Ad- been, opposed. OSHA has made three attempts ministration (MSHA) is currently required to re- to establish “generic standards. ” The agency pro- spond to certain NIOSH recommendations. How- mulgated a “cancer policy” in 1980 that defined ever, NIOSH has sent no such recommendations what data would be necessary and sufficient to to MSHA, perhaps because of the direct tie be- make a decision about a substance being a car- tween recommendation and regulation. cinogen and the nature of the standard that would Option 18: Congress might direct OSHA to de- then be issued. The “access to records regulation, ” velop methods to respond to changes in na- a generic standard applying to all employer-held tional consensus standards and other profes- health and safety records, guaranteed workers the sional recommendations. right to inspect records and required that employers retain them. The recently promulgated label- Option 19: Congress might require OSHA to con- ing, or “hazard communication” standard also has sider NIOSH recommendations for new or generic aspects. more stringent controls within a fixed period of time—say, 2 or 3 years. At the end of that Generic standards offer greater efficiency in that time, OSHA could adopt, modify, or decide matters of a general nature can be settled once Ch. l—Summary and Options ● 23 rather than being renewed for every specific case. ● increase the number of inspectors, and the There are, however, difficulties in issuing broad level of fines, and change the targeting and regulations that are to apply in many situations. settlement policies to increase incentives for Moreover, there is no guarantee that generic compliance. standards will be used. For example, no carcinogens have been regulated under the agency’s “can- Other Federal Actions Affecting Hazard Control cer policy. ” Various tax and financial assistance programs— Possible areas for generic standards include ex- investment tax credits, accelerated depreciation, posure monitoring and employee training. It may government loan programs, and direct subsidi- also be possible to issue standards that deal with es—might encourage employers to install control groups of, rather than single, substances. technologies. However, all these programs have Option 22: OSHA could be encouraged to issue disadvantages. First, they would reduce Federal generic standards to supplement substance or tax revenues or increase budget outlays. Second, hazard-specific rules. depending on their design, tax-based incentives can be relatively inefficient mechanisms because OSHA Enforcement Activity firms that would have installed controls, even in the absence of the program, would now receive No other OSHA activity stirs up so much emo- a tax subsidy, Third, there will be difficulties in tional fervor as its inspection and enforcement dividing the purchase price of equipment between activities. Many businesses object to inspections features that are health and safety controls and as being nit-picking and unrelated to employee those that are part of the equipment for purely health and safety, Employees and unions, on the productive reasons. other hand, believe that inspections are essential to worker protection and are concerned that Option 24: Congress might enact a tax and finan- OSHA devotes insufficient resources to them and cial assistance program to assist businesses in that inspectors are not vigorous enough in enforc- improving occupational safety and health. ing legal requirements. As the United States considers its economic and Whatever the number of inspections, some vio- industrial policies, it is unclear what balance is lations are found and punished by fines, In most to be struck between updating old-line industries cases, the fines levied by OSHA are less than the and focusing on new industries. In the future, the costs of controlling hazards. One possibility Federal Government may play an active role in would be to increase fines to levels equal to the the “reindustrialization” or “deindustrialization” actual costs of implementing controls. Or fines of America. might be based on a calculation of the amount necessary to have a deterrent effect. If explicit Federal policies are created, they may In some cases, fines equal to the costs of con- include discussions and agreements among inter- trol would exceed the maximum levels established ested businesses, unions, communities, and in the OSH Act. Therefore, the law may have to others, as well as Federal loans and financial be changed to allow higher penalties. Of course, assistance. Information could be developed con- higher penalties will raise the number of contested cerning the health, safety, investment, and pro- OSHA actions and the general level of contro- ductivity needs of various industries. One possi- versy in this field. bility would be to provide financial assistance for health and safety, as well as for productivity in- Option 23: Congress could consider what the vestments. appropriate level of OSHA enforcement activity should be; it could then either: The general disadvantages of these approaches ● continue the current levels of personnel and include the concern that health and safety will funding for inspection activity and the new “take second place” to the push for productivity. policies concerning inspection targeting and In addition, many object to any Federal role in citation settlement; or coordinating or financing industrial investments. 24 ● Preventing Illness and Injury in the Workplace

Option 25: If the United States makes available search efforts and training programs have also funds or tax incentives for the building or re- emerged from collective bargaining. building of industry: A fund could be established with or without ● controls for health and safety hazards could a Government contribution. For example, inter- be eligible for the same funds or tax breaks as other construction costs; ested citizens, employers, workers, foundations, and other groups could make voluntary contri- ● companies receiving reindustrialization as- butions. Or Congress could create a fund. If it sistance might be required to design health and safety into their new plant and equipment, becomes a Federal activity, financing could be either to meet existing standards or to achieve through a payroll tax on employers or, although lower exposure levels or safer processes. this would be more difficult, through a tax or surcharge based on workers’ compensation premiums It has been suggested that regulatory require- (with some adjustments for the presence of health ments have diverted resources from “productive” hazards in various industries). For example, a 0.1 uses and contributed to economic slowdowns. percent employer tax on the total U.S. payroll of However, in at least two cases (standards concern- $1.6 trillion (in 1982) would result in annual ing vinyl chloride and cotton dust), new produc- revenues of about $1.6 billion; a 0.01 percent tax tion processes were developed that both benefited would produce $160 million. A 1.0 percent sur- worker health and improved productivity. Fitting charge on workers’ compensation premiums regulatory activities to productivity concerns can (about $25 billion in 1980) would produce annual be achieved in two ways: either delaying regula- revenues of $250 million. Another possibility tory requirements until they coincide with planned would be to allocate fines collected for violations modernization or using health and safety regula- of OSHA standards to this fund. This would pro- tions to “spur” productivity improvements. duce less money; in 1983, OH-IA’S proposed fines Option 26: Congress could direct OSHA to: totaled $6.4 million. ● delay the required use of engineering con- Several different administrative arrangements trols, so that the installation of these controls for such a fund are available, Congress could fol- coincides with modernization of an industry; low the model of the Work Environment Fund of ● use health and safety regulations to en- Sweden by creating a tripartite board of employ- courage plant and equipment modernization. ers, employees, and Government representatives, or it could delegate administrative responsibilities to NIOSH, since this would be a research and in- Creation of an Occupational Safety and formation dissemination activity. The fund and Health Fund its research and training projects could exist alongside existing projects and arrangements at OTA is aware of concern about recent large OSHA and NIOSH, or Congress could consoli- swings in occupational safety and health policy. date existing research and training activities (in- Two areas—education and training programs, cluding NIOSH extramural research grants and and research on workplace controls—have had training grants, OSHA New Directions grants and funding reduced in the past few years. The crea- OSHA-funded consultations) under one umbrella tion of an Occupational Safety and Health Fund group. might provide more stable and enhanced funding. Although such a fund would enhance the com- Recent U.S. research concerning the use of mitment to research and training, there are dis- “washed cotton” to control the hazards of cotton advantages to consider—primarily that this rep- dust also provides a model for cooperative re- resents a new venture, with all the problems that search. This project was funded by Government such undertakings incur. Moreover, a new tax or and industry, with oversight and direction pro- surcharge, even though of modest size, runs against vided by a group of labor, management, and the desire embodied in recent legislation to reduce Government officials. Jointly administered re- business taxes. Ch. l—Summary and Options ● 25 —

Option 27: Congress could create an Occupational Changed Regulatory Approaches Safety and Health Fund to finance research in control technology, training and education, and Providing protection against occupational in- information dissemination. juries and illnesses in small business establishments presents its own set of problems. It may be cost effective to treat occupational health and safety The Needs of Small Businesses in such firms in a fashion similar to current regu- Loans for Compliance With OSHA Standards lation of consumer products—by regulating machines and products that small businesses Small businesses are often disproportionately purchase. burdened by investments required for health and safety protection. Congress recognized this when Of course, many products purchased by small it passed the OSH Act by also amending the Small businesses are also used in larger businesses, Business Act to allow the Small Business Admin- whose employees would also benefit from such istration (SBA) to make loans for OSHA com- regulation. An important limitation of this appliance. Between 1971 and 1981, when Congress proach is that some occupational hazards are cre- eliminated authorization for this program, SBA ated in the improper installation, use, and main- processed 261 such loans. Now may be a good tenance of machines and products. This time to study this program to learn what effect regulatory approach would have only limited im- these loans had and why so few were processed. pact on those hazards. Following such a study, Congress could consider Option 30: Congress could take actions to im- reauthorizing the loan program. prove the safety of products used by small busi- Option 28: Congress might direct OSHA and/or ness. This might include: SBA to study the results of SBA loans made • directing NIOSH to conduct tests of prod- for compliance with OSHA standards. ucts used by small businesses and to publish the results in a form easily available to such Shared Resources establishments; ● encouraging OSHA, Consumer Product It is inefficient and impractical to require each Safety Commission, and EPA regulatory ac- small business to provide a full range of health tions concerning the products used by small and safety services. Instead, organizations and businesses. programs to serve the needs of a number of small businesses in a given area or industrial specialty Establishment of Occupational Medicine Clinics might be cost effective. Initial funding could come In the United States, most occupational medi- from OSHA or NIOSH, with the hope that these cine is practiced in the workplace by physicians programs would ultimately be self-supporting. employed by industry, especially by large com- The most difficult part of this option is to de- panies. sign a method to sustain the program after the Changes are apparent, however, as small-and startup period. Even though shared programs medium-sized companies are making choices be- should cost less than if a company were to pur- tween contracting with hospital-based clinics for chase the services entirely on its own, some small medical care or maintaining a company medical businesses might find the price beyond their department. The clinics may grow to fill current means. It is unclear how to aid those companies. voids—servicing industries, regions, and employ- Option 29: Congress might direct NIOSH and ers where such services are unavailable or defi- OSHA to encourage the development of shared cient. Clinics might, because of a larger patient programs to provide industrial hygiene, safety load and a staff that consequently sees more pa- engineering, medical surveillance, and worker tients, be able to provide more-knowledgeable health and safety training for small businesses, care and improved physician training. 26 . preventing Illness and Injury in the Workplace

The staff of these clinics emphasize that they it is hard to measure improvements in occupa- will provide advice about prevention as well as tional health. medical care. The combination of staff physicians, Congress in the last few years has already in- industrial hygienists, and engineers could provide dicated a desire for more systematic assessment a critical mass for a great deal of important activ- of Government activities. The Regulatory Flex- ity in hazard identification and control. ibility Act of 1980, for example, requires that reg- Programs concerned with occupational medi- ulatory agencies, including OSHA, review over cine and prevention should consider and study the a 10-year period all regulations that have a sig- choices. They may alter industrial medical care nificant impact on small businesses. and responsibilities of industry and labor, as well This principle of reviewing and analyzing ex- as the relationships between such clinics and the isting programs might be extended to nonregula- private practice of medicine. tory programs. For example, the OSHA New Directions grants program, and the NIOSH train- Assessing Health and Safety Programs ing grants programs could be assessed. In addi- A key final component in improving occupation, periodic assessment could be specified when tional health and safety is evaluating which pro- new programs are established. The principal dis- grams to identify hazards, develop control tech- advantage of such a requirement would be the nologies, disseminate information, and implement diversion of resources from other important areas, controls work and which programs can be im- such as hazard identification and research on con- proved. Assessing or evaluating efforts in occupa- trol techniques. tional safety is difficult because of the many fac- Option 31: Congress could require periodic assess- tors that influence injury rates over time. Some ment of all occupational safety and health proof these may also stymie the evaluation of occupa- grams and provide funds to conduct such tional health activities; more importantly, because assessments. of latent periods and difficulties in recognition, 2.

Data on Occupational Injuries and Illnesses — Contents 2 Data on Occupation; Injuries and Illnesses

In this chapter, OTA presents a summary of ber of occupational injuries in U.S. workplaces. the available statistical information concerning the For occupational illnesses, however, the data are number and distribution of occupational injuries extremely limited. (A fuller discussion of these and illnesses. In general, currently available data topics is found in Working Paper #l of this describe, with reasonable accuracy, the total num- report. )

SOURCES OF INFORMATION

Prior to the passage of the Occupational Safety limitations of the survey methods, appear to be and Health (OSH) Act in 1970, occupational in- reliable. These estimates, however, are subject to jury data collection efforts were limited. One several limitations: source was a series of surveys by the Bureau of ● they are available only since 1972, Labor Statistics (BLS). This information was ● depending on the type of case, they cover limited by its dependence on voluntary reports only two-thirds to three-fourths of the U.S. from employers, by the particular standard most work force, commonly used for recording occupational in- ● they are based on a survey that is adminis- juries, and by incomplete industry coverage by tered only once a year, and the BLS surveys. One study of data from 1967 ● they provide very little detail concerning the and 1968 indicated substantial underreporting of nature and causes of occupational injuries. injuries by employers (186), In addition, some limited data were available from the National A possible fifth limitation is that these estimates Safety Council (NSC), state workers’ compensa- are ultimately based on employer records of in- tion agencies, and employer records. In the OSH juries and illnesses. (The extent of this possible Act, Congress called for the creation of a new, bias is discussed below and in Working Paper #l. ) mandatory system of data collection. Nothing can now be done about the first limita- The system created by the Occupational Safety tion. The second and third could be improved, and Health Administration (OSHA) and BLS re- but these would involve changes in methods and quires employees to keep records using a speci- perhaps require greater resources. To address the fied format. In addition, BLS conducts annual sur- fourth limitation, BLS has initiated two additional veys of a sample of employers. These survey data collection efforts. Since 1976, the Bureau has results are used to compute injury and, to a compiled information provided by 26 to 36 state limited extent, illness rates by industry, as well workers’ compensation agencies, in a data base as estimates of the total numbers of fatalities, lost- known as the Supplementary Data System (SDS) workday cases, and cases without lost worktime (397). And since 1978, BLS has conducted a series but that involved medical treatment. of surveys of injured workers concerning specific types of occupational injuries, published as Work The BLS Annual Surveys (604,606-608) are the Injury Reports (599-601,603,605). best source of statistical information concerning work-related injuries. The published data are Two other Federal systems provide injury data. based on large survey samples, and, within the The National Health Interview Survey (NHIS) of

29 30 ● Preventing Illness and Injury in the Workplace the National Center for Health Statistics (NCHS) cy room admissions to produce information on collects information during household interviews. occupational injuries. Finally, OSHA has pub- A system recently developed by the National lished analyses of several different types of fatal Institute for Occupational Safety and Health injuries using information collected during acci- (NIOSH) uses information from hospital emergen- dent investigations.

OCCUPATIONAL INJURIES

Fatal Injuries an adjustment must be made to generate an estimate for the entire workforce. As discussed in Working Paper #l, for the last OTA used two similar methods, both based on five years, data from the BLS Annual Surveys in- the BLS data, to develop its estimate of the num- dicate that between 3,000 and 5,000 occupational ber of occupational fatalities due to injuries. The fatalities occurred each year in private sector first estimate uses the five-year average of the establishments with 11 or more employees. About number of fatalities in establishments with 11 or 1,000 deaths occur in private sector establishments more employees. For 1979 to 1983, this equals with fewer than 11 employees. The NSC figures about 4,180. Approximately 11 percent of these range between 11,000 and 13,500 for the entire are due to heart attacks. Subtracting these yields workforce. Applying a variety of assumptions to a total of 3,720 deaths due to injury (see table 2- data derived from death certificates yields a range 1). To this should be added the five-year average of estimates from 5,500 to 11,000 for the entire from the BLS Annual Survey of deaths in estab- workforce for 1977. OSHA inspection data sug- lishments with fewer than 11 employees. For 1979 gest that at least 4,500 occupational fatalities to 1983, after adjusting for heart attacks, this was occurred in the private sector workforce in fiscal 930. Thus, the total from the BLS Annual Sur- year 1982. vey is 4,650. The Annual Surveys conducted by BLS are the To this should be added the deaths among the best source of statistical information on occupa- self-employed and public employees. Applying tional injuries. They use a large survey sample that the death rates for private sector workers, ad- is capable of directly measuring the occurrence justed for heart attacks, to these workers yields of injuries. However, because this sample covers a five-year average of 1,640. Adding this to the only private sector employment (the self-em- 4,650 generated directly by the Annual Survey ployed and public sector employees are excluded), yields a total of about 6,300 deaths.

Table 2-1.—Annual Occupational Fatalities From Injuries, Summary of Estimates

Average annual Source of Universe Years total estimate or data All employment ...... 1979-83 12,200 NSC All employment ...... 1979-83 6,000 OTAa Private sector workplaces with 11 + employees ...... 1979-83 3,720 BLS Private sector workplaces with 1-10 employees ...... 1979-83 930C BLS aEgtlmate ig b~~ cm BLS arid OSHA data and Is described in the text bd,l~ minus I I percent of deaths reported es heart attacks (~). c1 ,040 minus 11 percent of deaths reported as heari attacks. (110 NOTE: Because of differing methods and deflnltions, some of these estimates are, strictly speaking, not directly comparable with each other. SOURCE: Office of Technology Assessment. ——— —. . .

Ch. 2—Data on Occupational Injuries and Illnesses ● 31

Alternatively, the BLS figures for establish- Table 2.2.—Occupational Fatality Ratesa ments with 11 or more employees for each indi- By Industry for 1982 vidual year can be adjusted by applying the Industry division Rateb private-sector death rate (for injuries in establish- Mining ...... 44.3 ments with 11 or more employees) to the work- Construction ...... 28.7 ers excluded completely (self-employed and public Agriculture, forestry, and fishing ...... 28.4 Transportation and public utilities ...... 21.9 employees) or for which annual estimates are not Average of private sector ., ...... 7.4 available (small, private sector establishments). Manufacturing ...... 4.5 Using this method, the total ranges from 7,265 Wholesale and retail trade ...... 3.8 Services...... 3.5 in 1979 to 4,600 in 1983, with an average of 6,180 Finance, insurance, and real estate ...... 2.5 deaths per year. aF~r ~~t~bllghrnentg with I 1 or more employees Includes fatal injuries and reported deaths due to illness. bFatalit~eg per 100,000 full-time workers Rounding either of these estimates to the nearest SOURCE: (608) thousand yields OTA’s estimate that about 6,000 deaths due to occupational injuries occur each year. The National Safety Council average of jury. Manufacturing has a death rate below the 12,200 is considerably higher. It is difficult if not average for the private sector, although it has a impossible, to reconcile the estimates from the BLS nonfatal injury rate substantially above the aver- and the NSC. The published data from state vital age. Wholesale and retail trade; services; and fi- statistics are insufficient to do so (see Working nance, insurance, and real estate are safer indus- Paper #l). Moreover, the NSC estimates have tries, with fatality rates between 2.5 and 3.8 been criticized in the past for including duplicate deaths per 100,000 workers. reports and deaths from previous years and for not being based on the results of a probability survey. Instead they are developed using the results Nonfatal Injuries of special studies in combination with a variety of statistics from several sources. These methods Table 2-3 summarizes the estimates of the num- may not generate reliable estimates, either for a bers of occupational injuries. These estimates vary particular year or over time. partly because of differences in the definitions of injuries, the population universes, and methods 6,000 OTA’s estimate of injury deaths per year of estimation. translates into about 25 occupation] fatalities each working day. Rarely, however, does this NSC defines a disabling injury as one that in- daily toll occur at the same time, in the same volves one or more days away from work, some workplace. Usually, occupational deaths occur form of permanent impairment, or death. For BLS one or two at a time in widely scattered work- and OSHA, a lost-workday injury is one that in- places. Because of this, occupational fatalities only volves the employee either not working at all for rarely receive significant publicity. one or more days beyond the day of the injury or reporting to work and being assigned to a Motor vehicles (30 to 40 percent), off-the-road “lighter duty” job (restricted work activity). industrial vehicles (10 percent), and falls (10 percent) are associated with over half of the fatal oc- Combining the BLS estimate of lost-workday cupational injuries. In addition, occupational fa- injuries (2.1 million in both 1982 and 1983) in the talities are unevenly spread among industries. private sector with the estimates for Federal, State, Table 2-2 presents fatality rates for the major in- and local employees (0.1 and 0.3 million) yields dustry divisions in the private sector. Mining, a total of 2.5 million lost-workday injuries. This with a fatality rate of 44.3 per 100,000 full-time compares to the NSC estimate of 1.9 million “dis- workers, is the most hazardous industry. Con- abling” injuries. Using the total of 2.5 million instruction; agriculture, forestry, and fishing; and juries, it appears that each working day results the transportation and public utility industries also in about 10,000 injuries that are serious enough present above-average risks of deaths due to in- to lead to loss of worktime. 32 ● Preventing Illness and Injury in the Workplace

Table 2-3.—Annual Nonfatal Occupational Injuries, Summary of Estimates

Source of Estimate estimate Definition Universe Year (millions) or data Lost-workday cases ...... All private sector workplaces 1983 2.1 BLS Lost-workday cases ...... U.S. Government—civilian and personnel 1982 0.1 OSHA Lost-workday cases ...... State and local government 1982 0.3 SDSa “Disabling” ...... All employment 1983 1.9 NSC Total “recordable” ...... All private sector workplaces 1983 4.7 BLS Total “recordable” ...... U.S. Government—civilian and personnel 1982 0.2 OSHA Total “recordable” ...... State and local government 1982 0.7 SDSa Medically attended or activity restricted . .All employment 1981 11.3 NHIS Treated in emergency rooms ...... All employment 1982 3.2 NIOSH %tlmato Is baeed on data from SDS and Annual Survey and Is described In Working Paper #1. NOTE: Because of differing methods and deflnltlons, some of these estimates are, strictly speaking, not directly comparable with each other SOURCE’ Office of Technology Assessment

The BLS/OSHA definition of “recordable” in- The NHIS includes self-employed persons, but jury includes all lost-workday injuries plus all it is unlikely that approximately 8.7 million self- those that involve medical treatment beyond first employed workers can account for the remain- aid. The National Health Interview Survey in- ing 5.3 million injuries. It could be that the slightly cludes all injuries that are “medically attended” different definitions of the BLS Annual Survey (which may involve only a consultation with a and the NHIS contribute to this discrepancy. It doctor) or cause “restricted activity,” whether or is also possible that employees and their families not that also involves lost worktime. Data devel- are reporting to the NHIS injuries that are not oped by NIOSH through the National Electronic recorded by employers. A number of reasons Injury Surveillance System (NEISS) include all could account for employers not recording an in- cases treated in hospital emergency rooms. jury: an employee may not have reported it to the employer; the employer judged the injury to These various sources again yield differing esti- be a “’first aid only” case that is not required to mates. The sum of BLS data and the Federal, be recorded; or the employer’s records are not ac- State, and local government estimates is 5.7 mil- curate and comprehensive. lion recordable cases in 1982. The number of private sector injuries went down slightly in 1983, The NEISS estimate of emergency room cases Assuming that the number of public sector injuries could be consistent with either the BLS estimate stayed the same (0.9 million), the sum of the BLS or the NHIS figure. The 3.2 million emergency and public sector estimates would be 5.6 million room cases (in 1982) would constitute about 55 in 1983. The NHIS estimates 11.3 million cases. percent of the 5.7 million cases in the private and This translates into a range of between 22,000 and public sectors from (the 1982) BLS and govern- 45,000 injuries each working day. And the NIOSH ment reports or about 30 percent of the 11.3 mil- estimate of the number of emergency room cases, lion cases estimated by the NHIS. The figure of based on the NIESS data, is 3.2 million cases or 30 percent is roughly consistent with a special about 12,000 cases per day. study of data from 1975, which estimated that about 36 percent of all injuries occurring “at job The largest difference in these figures is between or business” were medically attended at emer- the combined BLS and public sector estimate (5.6 gency rooms (388). million cases) and the National Health Interview Survey (11.3 million). Most of this difference re- Table 2-4 summarizes various estimates of the mains even when data from the same year are amount of time lost due to nonfatal occupational used. The 1981 BLS and public sector estimate is injuries, including lost worktime, bed disability, a total of about 6 million cases. and restrictions on daily activity. The BLS esti- Ch. 2—Data on Occupational Injuries and Illnesses ● 33

Table 2-4.—Days Lost Annually From Occupational Injuries, Summary of Estimates

Source of Estimate estimate Definition Universe Year (millions) or data Lost workdays...... All private sector workplaces 1983 36.4 BLS Lost workdays...... U.S. Government—civilian and personnel 1982 1,2 OSHA Lost workdays. ., ...... All employment 1981 60-70’ OTA Bed disability days...... All employment 1981 44.0 NHIS Restricted activity days ...... All employment 1981 214.9 NHIS . - - – - %TA estimate IS based on data-f~om NHIS and Is–described I; Workln9-Paper #l NOTE Because of dlfferlng methods and deflnltlons, some of these estimates are, strictly speaking, not directly comparable w!th each other SOURCE Off Ice of Technology Assessment mate of days away from work and of restricted data collected from other sources. However, ac- work activity (“light duty”) equals 36.4 million counts of working conditions earlier in this cen- days—the equivalent of over 140,000 people tury reveal many instances of job hazards that are working full-time for a year. Time lost by Fed- considered appalling by today’s standards. Many eral civilian employees adds about 1.2 million of these have been improved, and injury rates days to this total. No estimates for State and local have fallen. (Of course, this decline in injury rates employees or for the self-employed are directly may not apply to occupational illness rates. ) available. Trends over the last two decades have not been A higher total of about 60-70 million “’lost constant from year to year and measures of dif- workdays” can be indirectly estimated using the ferent types of injuries sometimes go in different results of a special NCHS analysis of injuries. Part directions. During the 1960s, the BLS surveys of of the reason for the higher figure is that this esti- manufacturing showed rising injury rates. The mate should cover the entire work force. In BLS Annual Surveys from the 1970s show a rela- addition, the varying definitions and survey meth- tively large drop in non-lost-workday injuries ods between the BLS and the NHIS may contrib- from 1972 to 1975, and then a continuing decline ute to the difference. The NHIS also provides data from 1975 to 1983. The BLS lost-workday case on bed disability (44 million days) and of days rate rose during most of the 1970s, falling only of restricted daily activity (close to 215 million between 1974 and 1975 and for the three years days). after 1979 (fig. 2-1). The BLS total case rate, which The BLS figures imply that about 1 in 13 U.S. combines the non-lost-workday and lost-workday workers suffered an occupational injury in 1982. cases, shows a slight rise from 1972 to 1973, and Nearly half of these were serious enough to re- then a decline from 1973 to 1975. This rate then sult in the employee missing one or more days began a slight rising trend until its peak in 1979. of work beyond the day of the injury, On aver- By 1982, the total case rate had declined to 7.7 age each lost-workday injury results in 17 days per 100 workers. This decline continued in 1983, lost from work. The NSC has estimated that the when the total case rate was 7.6 per 100 work- direct and indirect costs of work injuries totaled ers. This is the lowest level reported since 1972, $30.2 billion in 1980 (324), when these data were first collected. Fatality rates calculated from the BLS estimates show a general decline through the 1970s. Injury rates calculated Trends in Injury Rates from the National Health Interview Study show It is probable that occupational injury rates no readily apparent trend from 1962 to 1981, pos- have fallen since the turn of this century. The data sibly because of the relatively larger sampling published by the NSC support this, but these data error for this survey. may not be accurate or reliable. In particular, re- Injury and fatality rates are affected by a num- cent trends in the NSC data are inconsistent with ber of factors. These include the effects of the busi-

34 Preventing Illness and Injury in the Workplace

Figure 2.1.- Occupational Injury Rates, 1972.83 Although it is possible that some of the decline can be explained by these changes, several other features must also be considered. First, as figure 2-1 shows, the decline in injury rates started in 1980—before the changes in OSHA policies that were instituted in 1981. More- over, the installation of controls, changes in employee training, etc., often take place over several years. Thus it is possible that the observed changes from 1979 to 1983 represent the effects of some combination of the policies of the “old” and the “’new” OSHA. Second, the new policy that targets inspections on the basis of injury records may influence employer recordkeeping toward undercounting. Independent verification 21- Lost-workday cases I of employer injury records is necessary to assess the possible impact on changes in employer recordkeeping. Third, as mentioned above, a num- 1972 1974 1976 1978 1980 1983 ber of factors besides the effectiveness of OSHA Year can influence injury-rate trends. Foremost among the factors influencing injury statistics is the business cycle. Since the 1930s researchers have noted that, other things being equal, increased business activity leads to higher ness cycle, various changes in the administration injury rates while decreased activity lowers rates. of workers’ compensation, the practice of occupa- The general explanation for this phenomenon is tional medicine, ‘and other socioeconomic factors. that as business picks up, employers hire more In addition, OSHA and NIOSH and other safety young and inexperienced workers. Both younger and health programs may have contributed to workers and inexperienced workers of all ages changes in rates. But the effects of OSHA activi- tend to have higher injury rates than older, more ties, in particular, may be difficult to discern in experienced workers. Moreover, as production ex- national injury trends because of the low prob- pands, businesses open new plants and bring new abilities of OSHA inspections, the relatively low machinery on-line. For both of these cases there penalties for violations of standards, and the pos- may be a period of adjustment as management sibility of differential effects on various types of and workers learn how to use the machinery safe- hazards. There may also be variations in the ef- ly. In addition, during a business upturn there will fectiveness among the 24 separate State programs be increases in the pace of production, increases and Federal OSHA operations, which cannot be in the amount of overtime worked, less down detected using national data. time, and less time devoted to repair and maintenance, all of which lead to increases in accidents. Injury Rate Trends, the Business Cycle, and During business downturns, all of these elements OSHA Policies are reversed-younger and less experienced workers are laid off while older and more experienced The lack of any dramatic improvements in in- employees are retained, plant operations slow jury rates during the 1970s has been cited to sup- down, and more effort is devoted to repair and port the belief that OSHA has been ineffective. maintenance (254,444). More recently, it has been claimed that the recent declines in the injury rates resulted from the cur- OTA has compared injury rates with several rent administration’s “cooperative, non-adversary measures of the business cycle. Figure 2-2 shows approach to job safety and health” (643,644,650). data for the BLS total recordable injury rate and

Ch. 2—Data on Occupational Injuries and Illnesses ● 35

Figure 2.2.—lnjury Rates and Unemployment— Figure 2-3.– Injury Rates and Unemployment- Private Sector: 1972.83 Private Sector: 1972-83

Unemployment rate (percent) — Injury rate ---- Unemployment rate

Figure 2.4.—Lost-Workday Rate and Employment— the unemployment rate for 1972 to 1983. The total Construction: 1972.83 recordable injury rate declined from 1973 to 1975 and again from 1979 to 1982, simultaneously with rising unemployment rates. In addition, the rising injury rates from 1975 to 1979 coincided with declining unemployment rates. From 1982 to 1983, the total injury rate and unemployment went down slightly, while the lost workday rate stayed the same. In figure 2-3, these two variables have been plotted against each other, with the unemployment rate on the horizontal (or x-axis) and the injury rate on the vertical (or y-axis). Examina- tion of this figure reveals an apparent inverse rela- 0.0 ‘ ‘ 2.0 4.0 6.0 tionship between injury rates and unemployment. Employment (millions) That is, as unemployment rises in a recession, injury rates decline. rates in construction and manufacturing alone. Another possible measure of the business cy- Again there appears to be a close relationship. As cle is to examine the level of employment, as op- employment rises, so do injury rates. posed to the unemployment rate. This must be done carefully because in the last few years, the OSHA has suggested to OTA that the recent changes in the level of employment have not been injury rate declines are not the result of the cur- the same in all industries. In fact, from 1979 to rent recession because “BLS has estimated that 1982, employment in the more hazardous manu- only 16 percent of the decline in injury rates in facturing and construction industries declined, 1982 can be attributed to a disproportionate drop while employment in the other major private sec- in hours worked in high-risk industries” (34). BLS, tor industry groups has stayed the same or in- however, noted that the procedure they used for creased. Figures 2-4 and 2-5 show the relationship this calculation, “does not take into account . . . between employment and lost-workday injury other factors which may also affect the rate, but

36 Preventing Illness and Injury in the Workplace

Figure 2.5. -Lost. Workday Rate and Employment— When the OSH Act was enacted in 1970, Con- gress placed the legal responsibility for preventing occupational injuries and illnesses with employers and created several agencies, including NIOSH and OSHA, to conduct research and administer regulations. Employees, of course, also have a personal stake in preventing disease and injury in the workplace. Some detailed studies on the effectiveness of OSHA in improving the efforts of employers and employees have found a favorable but small impact, while other studies have not found any effect (see ch, 13 for a summary). Even the favorable effects detected in several studies may not be large enough to be dis- 0.0 ‘ ‘ 8.0 12.0 16.0 20.0 24.0 cerned in national injury statistics, while shifts in Employment (millions) the nature of the injury rate-business cycle relationship may be difficult to detect. However, it is clear that if any improvements have been made they have not been large. which have not been measured, such as the dem- . . ographic composition of the workforce, worker Accuracy of Occupation Injury Estimates education, improved safety measures, the role of Questions raised about the accuracy of esti- State and Federal agency compliance programs, mates based on employer-maintained injury technological change, etc” (607). records have intensified with the current admin- But the “demographic composition of the work- istration’s inspection targeting system. OTA has force” and “technological change” are two conducted a limited comparison of data from the variables that are particularly affected by the busi- BLS Annual Survey and from the BLS Supplemen- ness cycle. In addition, the amount of overtime tary Data System (SDS) to see if they are con- worked, the pace of production, and the rate of sistent. Data were available only for years before new hires are affected by the business cycle. The the implementation of the new inspection target- BLS procedure for calculating the effects of the ing system and the conclusions of this analysis, decline in hours worked may not fully capture the therefore, apply just to that period. effects of these other variables on injury rates. Some States participating in the SDS report Additional analysis using variables that directly only lost-workday cases while others report all measure the new hire rate, the number of over- cases involving either lost workdays or medical time hours, the rate of production, and capacity treatment. OTA found that for States reporting utilization in specific industries may clarify this only lost-workday cases to the SDS, the numbers relationship further. (Examination of the influence of cases were not consistently higher or lower of new hires is, however, made more difficult be- from either SDS or the Annual Survey, after ad- cause BLS no longer publishes statistics on labor justing for the minimum waiting periods (as turnover, which included the new hire rate. ) But defined by State workers’ compensation). Al- at present, it appears that the effect of the recent though there were some differences between these recession, especially in construction and manu- two data sources, these differences were not con- facturing, is the most important factor behind the sistent from one State to the next. In States that injury rate declines from 1979 to 1983. In addi- report all cases involving either lost workdays or tion, it appears that national injury rates since medical treatment to the SDS, consistently more 1972 have been largely related to the level of busi- cases were reported to the SDS than would be ex- ness activity. pected from the Annual Survey data. Ch. 2—Data on Occupational Injuries and Illnesses ● 37

In another analysis, OTA compared BLS An- may not be worrisome. Or it could be that em- nual Survey injury rates with rates calculated from ployers do not report certain types of injuries to the estimates of the NHIS. This analysis compared the BLS Annual Survey even though they do sub- information derived from employer records (BLS) mit reports to workers’ compensation. As dis- with that from workers and their families (NHIS). cussed above, the differences between the BLS An- This comparison showed that, in recent years, nual Survey and the NHIS may also stem from overall injury rates based on the NHIS are about employers labeling some injuries as cases involv- one-third higher than those from the BLS Annual ing only first-aid treatment, even though employ- Survey. ees and other family members consider them serious enough to report to the NHIS. Differences among the various sources could arise from different methodologies and, as such,

OCCUPATIONAL ILLNESSES

There is substantially less quantitative informa- from 10,000 deaths to 210,000 deaths. More action on occupational illnesses than on injuries. Al- curate estimates are difficult because of a general though employers are required to include occupa- lack of information on both historical and cur- tional illnesses in their records, it is well accepted rent worker exposures, incomplete knowledge of that employer records and the BLS Annual Sur- the deleterious effects of workplace exposures, and vey estimates, which are based on employer the general problems of assigning single “causes” records, underestimate the magnitude of the oc- to diseases created by multiple factors. cupational disease problem. This is because many Although it is well accepted that employer occupational diseases are indistinguishable from records understate the magnitude of the occupa- non-occupational diseases, because they often be- tional illness problem, it is very difficult to quan- come manifest only after a latent period, and be- tify the extent of this understatement. One pilot cause of a general lack of recognition of the oc- study, conducted by David Discher and colleagues cupational causes of many diseases. For 1983, the (143), explored the usefulness of medical exami- BLS Annual Survey estimate is about 106,000 oc- nations and industrial hygiene surveys for iden- cupational illnesses, but that is almost certainly tifying the extent of occupational illness in sev- an underestimate. eral industries. The researchers administered The most commonly quoted estimates are that medical exams to workers in four industries, con- up to 100,000 deaths due to illness and 390,000 ducted industrial hygiene surveys, and classified illness cases occur each year as a result of work- a total of 451 medical conditions among the sur- place conditions. Although the estimate of 390,000 veyed workers as probably linked to occupational cases was cited during the Congressional debates exposures. Eighty-nine percent of these 451 con- about the Occupational Safety and Health Act, ditions were not noted in either the workers’ com- OTA has not been able to determine the exact pensation claims or the employers’ logs. Although methods for deriving this estimate. this percentage of non-reporting may not be applicable to all workplaces, this study did reveal a The estimate of 100,000 deaths was derived first large number of cases that were not being recog- by a crude technique using the results of three epi- nized by employers and were thus not recorded. demiologic studies and later by an analysis of information in a 1951 British death registry. As a It is also interesting to note that the BLS Sup- result, the 100, 000-deaths figure can only be con- plementary Data System reported only 234 work- sidered an estimate, but it is unclear to what ex- ers’ compensation cases for all cancers in 1980. tent the figure is biased. Peter Barth and Allen This can be compared to the range of estimates Hunt (46) have reported other estimates that range for occupational cancer caused by asbestos alone, 38 ● Preventing Illness and Injury in the Workplace

which is between 4,000 and 12,000 cases annually. The NIOSH National Occupational Hazard Sur- Even if the SDS total is adjusted for all the states vey (564,565,566) and National Occupational Ex- that do not currently report to the SDS, there ap- posure Survey can provide information on the pears to be substantial underreporting of cancer number of employees potentially exposed to haz- cases to workers’ compensation programs. ardous substances, but they do not provide any information on the level of exposure and only The absence of an accurate accounting of the limited information on the duration of exposure. occupational disease toll highlights the need for Although it may be possible to use the data col- accurate and comprehensive information on em- lected by OSHA during inspections to develop ployee exposures, both to provide a basis for more estimates of worker exposures, such analysis de- accurate disease estimates and to measure the ef- pends on further research. fectiveness of current occupational health efforts.

CONCLUSION .

OTA estimates that about 6,000 U.S. workers OTA finds that the effect of the recent reces- die each year from occupational injuries, or about sion, especially in construction and manufactur- 25 each working day. In addition, each working ing, is the most important factor behind the in- day there are at least 10,000 injuries that result jury rate declines from 1979 to 1983, In addition, in lost worktime, and about 45,000 that result in it appears that most of the changes in national restricted activity or that require medical atten- injury rates since 1972 are associated with changes tion. Estimates of the number of nonfatal injuries in business activity. vary partly because of differences in the defini- Compared with occupational injuries, there is tions of injuries, the population universes, and substantially less quantitative information on oc- methods of estimation. Some discrepancies may cupational illnesses, Although employers are re- also be due to differences in how employers and quired to include occupational illnesses in their employees interpret the severity of an injury. records, it is well accepted that employer records Injury and fatality rates are affected by a num- underestimate the magnitude of the occupational ber of factors. These include the effects of the busi- disease problem. However, it is difficult to quan- ness cycle, and various changes in the adminis- tify the extent of this understatement. tration of workers’ compensation, the practice of occupational medicine, and other socioeconomic factors, as well as the possible effectiveness of OSHA in reducing injury frequency and severity. 3.

Health Hazard Identification 42, 42 43 45 45 48 49

52 52 54 ...... 62 ...... 63

TABLES

3-1. The Ten Leading Work-Related Diseases and Injuries: united States, 1982 ...... 43 3-2. some Occupational cancer Hazards ...... 46 3-3. Neurologic Effects of Occupational Toxins ...... 50 3-4. Substances for Which NIOSH Has Recommended Exposure Limits to

Prevent Skin Disorders ● ...... 51 3-5. Workplace Agents That Induce skin Disorders ...... 51 3-6. Industries at Highest Risk for occupational Skin Diseases 52 3-7. General Classification of Tests Available to Determine Properties Related to Carcinogenicity...... 56. . . 3-8. Advantages and Disadvantages of Case-Control and Cohort Studies ...... 59 Health Hazard Identification

Since work occupies a central place in most Traditionally, physicians and groups of work- lives, it is not surprising that it is related to many ers have been the sources of information leading afflictions, nor that in one form or another it con- to the association of particular hazards with dis- tributes to diseases of every system of the body. ease. “Factory fever” (typhus), “mad hatters” (victims of mercury poisoning), and “wrist drop” (lead Some diseases are relatively easily linked to poisoning) were related to workplace exposures workplace conditions, either because the diseases through observation. Recent years have seen in- themselves are distinct or relatively rare, or be- creasing importance being given to epidemiology— cause the particular workplace conditions differ the study of the distribution of diseases—and greatly from ordinary conditions of daily life. toxicology— the study of the dangerous proper- Other diseases are associated with either the work- ties of substances—in identifying workplace place or other activities, or with both; pinning hazards. down causes and preventive strategies of those diseases is more complicated. Case reports from doctors, workers, and em- Occupational diseases have been recognized for ployers can be valuable sources of information centuries, although definitions of disease and ill on hazards and serve to generate hypotheses for health have changed over time. Society—less will- larger studies. But inadequacies in the training of ing to accept adverse effects of any kind and physicians, both those who practice occupational knowing that much disease is preventable—no medicine and those in general practice, limit iden- longer believes poor health to be a necessary con- tification hazards through case reports. comitant of work. Epidemiology relies on observations or sugges- While attention often focuses on new hazards, tions of possible associations between exposures and in identifying and preventing more subtle, or behaviors and disease for hypothesis genera- previously unnoticed effects, professionals in oc- tion. It has limitations in the kinds and magni- cupational safety and health also continue to deal tudes of effects it can detect, but it can provide with many cases of well-known occupational dis- the most convincing evidence of associations be- eases. The still-frequent occurrence of many of tween exposures and behaviors and health. The these older diseases represents a failure to use strengths of epidemiology still remain to be ex- already available knowledge. ploited, and much remains to be learned about diseases and syndromes that are widespread in the Health hazards include those identified as pres- population. ent in the workplace, those present but unidentified, and new hazards, not yet introduced there. Toxicology can garner useful information about The identified hazards include exposures to phys- the possible effects of substances, but large tox- ical agents such as radiation and noise, and ex- icologic studies are expensive, require years to posures to some substances, including chemicals, complete, and produce information that is some- metals, minerals, and vegetable dusts. Present, times difficult to apply to human exposures. Mak- unidentified hazards may be many or few. Con- ing risk assessments from animal data involves tinued observation of workers and testing of sub- both technical problems and assumptions. Al- stances are necessary to determine what exposures though continued attention to toxicology and risk are hazardous. Testing of new substances should assessment may reduce technical controversies, reduce the number of hazards introduced un- the assumptions about the predictive value of knowingly into the workplace. various tests are likely to remain in dispute.

41 42 . Preventing Illness and Injury in the Workplace

Epidemiology and toxicology have not been the Occupational health surveillance is the source of panacea for solving workplace health problems both great promise and great controversy. It can be that some envisioned. The limitations of both used to identify the causes of occupational illness, argue for a continuing role for occupational medi- setting the stage for preventing further illness. But cine in hazard recognition as well as treating there are practical difficulties in implementing sys- workers. That role can be enhanced during the tems that will be statistically useful, concerns about education of physicians. company liabilities after discovering a possible relationship, and concerns that efforts will be made to Computerized information about workplace ex- substitute surveillance activities for preventive ef- posures and workers’ health forms the basis for sur- forts and installation of appropriate controls. veillance systems that aim to identify health hazards.

OCCUPATIONAL DISEASE

Some diseases are always or nearly always to other human activities. In other words, work- caused by conditions at work. These diseases rep- place exposures c ause workers to suffer an in- resent relatively easy cases for health and safety creased incidence of disease, even though these professionals because they can be readily linked diseases also regu arly occur in the general pop- to particular working conditions. In general, iden- ulation. tification of workplace hazards is facilitated by: For example, most lung cancer occurs in smok- ● conditions at work that differ greatly from ers, and it is accepted that there is a causal rela- the normal conditions of daily life, and tionship between cigarettes and lung cancer. Some ● the presence of distinctive or very rare dis- substances encountered in the workplace are also eases in these exposed workers. known lung carcinogens because they increase the occurrence of lung cancer in nonsmokers as well Examples from the early part of this century are as in smokers. In addition, smoking and other car- the occurrence of “phossy jaw” among phosphor- cinogens may also act together to cause cancer. us match workers, the diseases of radium dial However, deciding which exposure(s) caused lung painters, and “wrist drop” caused by lead poison- cancer in a particular smoking worker is a diffi- ing among adult workers. More recently, meso- cult task. thelioma and liver angiosarcoma both occur so rarely in the nonexposed population that when Hazards that increase the incidence of common cases were observed among asbestos and vinyl diseases can be best identified using the techniques chloride workers, respectively, questions of oc- of epidemiology. But even after studies have cupational causation were immediately raised and shown a link between exposures and increased dis- relatively quickly answered. ease incidence for a group of workers, it often remains impossible to determine, for any individ- But relationships between work and diseases are ual worker, that his or her disease was caused by not always so clear-cut. In fact, it is probably occupational exposures. more frequent that working conditions directly cause or contribute to diseases that are also related

MAJOR CLASSES OF OCCUPATIONAL DISEASES

Occupational diseases have been recognized as Ramazzini, the 18th-century physician often called such for centuries. References to almost all classes the father of occupational medicine. Since then, and types of diseases appear in the works of the definition of disease in general has changed, —. —

Ch. 3—Health Hazard Identification ● 43 ——— as has the perception of work-relatedness. Society Leading” work-related diseases, the list includes is less willing to accept adverse effects of any kind. nearly all categories of health effects that have Because so much disease is known to be pre- ever been linked to workplace conditions. ventable, poor health is no longer taken as a con- Six of the categories of diseases listed by NIOSH comitant of certain occupations. Our increasing are discussed in this chapter. Traumatic injuries ability to detect subtle effects allows us to broaden are the subject of the next chapter, and noise- our efforts in prevention. induced hearing loss is discussed in chapter 8. The There is something seductive about new risks, reader is referred to the recent textbooks by Levy and a tendency to focus on new hazards. Al- and Wegman (269) and Rem, et al. (396), for de- though in one sense occupational health is deal- tails of disease and hazard identification. ing with new and ever-subtler effects, the old diseases are still around, in greater numbers than is generally perceived. In 1979, an estimated 84,000 Respiratory Disorders active workers suffered from acute byssinosis and The lungs and other parts of the respiratory at least 35,000 employed or retired workers were tract come in contact with all manner of airborne disabled from cotton dust-related disease. In 1978, materials in the workplace. Gases, vapors, fumes, an estimated 59,000 workers were thought to suf- fibers, and particles all may be inhaled, Of all fer from silicosis. Even as new and perhaps scien- health effects, occupationally related cancers of tifically and medically more intriguing conditions the respiratory tract receive the greatest attention, become issues in occupational health, the old but they are not the only serious respiratory con- problems require continued vigilance. ditions associated with the workplace, and cer- To guide its research priorities, the National In- tainly not the most widespread. Other responses stitute for Occupational Safety and Health (NIOSH) of the respiratory system may be acute irritation, has developed a list of 10 groups of occupational immunologic or allergic reactions, or chronic diseases (table 3-1). Although termed the “Ten changes in the tissues that line the respiratory

Table 3.1 .—The Ten Leading Work-Related Diseases and Injuries: United States, 19828 -. Type of disorder/inju”~ .— Exam Dies 1 Occupational lung diseases asbestos is, byssinosis, silicosis, coal workers’ pneumoconiosis, lung cancer, occupational asthma 2 Musculosketal injuries disorders of the back, trunk, upper extremity, neck, lower extremity; traumatically induced Raynaud’s phenomenon 3, Occupational cancers (other than lung) leukemia, mesothelioma; cancers of the bladder, nose, and liver 4. Amputations, fractures, eye oss, — lacerations, and traumatic deaths 5. Cardiovascular diseases hypertension, coronary artery disease, acute myocardial infarction 6 Disorders of reproduction infertility, spontaneous abortion, teratogenesis 7, Neurotoxic disorders peripheral neuropathy, toxic encephalitis, psychoses, extreme personality changes (exposure-related) 8, Notse-induced loss of hearing — 9. Dermatologic conditions dermatosis, burns (scaldings), chemical burns, contusions (abrasions) 10, Psychologic disorders neuroses, personality disorders, alcoholism,— drug dependency — ‘The condlt!ons I Isted u rider each category are to be v!ewed as sdwfed examples, not cornprehenswe defl n It Ions of the category SOURCE (563) 44 ● Preventing Illness and Injury in the Workplace tract. Some conditions that begin as acute prob- the nose, throat, and larynx-is the most frequent lems progress to chronic states, perhaps the best site of irritation, It is susceptible to highly solu- known being byssinosis–or “brown lung” disease. ble irritants, such as ammonia, hydrogen chloride, and hydrogen fluoride-gases commonly encoun- NIOSH has made formal recommendations for tered in industry. maximum exposure levels to 60 substances, based on their effects on the respiratory system. That Irritants that are less soluble tend to travel far- number is greater than the substances cited for ther down the respiratory tract before they are any other organ system. absorbed entirely, causing irritation in the middle as well as the upper respiratory tract. Chlo- Chronic Conditions rine, fluorine, and sulfur dioxide, all commonly used chemicals, have such properties, The major The most serious conditions are pneumoco- effect on the lungs is bronchoconstriction. nioses, chronic conditions occurring generally after years of exposure to very fine dusts. The Irritants of low volubility may cause only mi- tissue reacts by thickening, producing a condition nor upper respiratory tract problems, but their called “pulmonary fibrosis. ” The best known delayed reaction deep in the lungs, which may oc- pneumoconioses are asbestosis, silicosis, and coal cur as much as a day later, can be very serious. workers’ pneumoconiosis (“black lung”), but simi- Ozone, oxides of nitrogen, and phosgene—again, lar conditions may be produced by a number of commonly encountered in workplaces—are the different materials, such as talc and kaolin. most important hazards in this class. Pneumoconioses are characterized by coughing Asthma and “hypersensitivity pneumonitis” are and shortness of breath, which grow worse over two manifestations of immunologic or allergic time, followed in the later stages by signs of heart type reactions. Bronchial asthma, a condition failure and eventually ending in death. affecting perhaps 4 percent of the U.S. popula- Chronic bronchitis can be caused by a number tion, is also prevalent among certain occupational of occupational hazards but, as the commonest groups. Asthma is a generalized obstruction of the chronic response of the respiratory tract, is also airways in an allergic type of response to some brought on by nonoccupational causes. It may substance. Causes can be of bacterial or animal also be multicausal, as many diseases are, with (e.g., animal dander, small insects, bee toxin) or nonoccupational factors (particularly cigarette plant (e.g., flour, grain dust, fungi, cotton, flax, smoking) interacting with occupational exposures tea fluff, wood dusts) or chemical (e.g., formal- to cause disease. dehyde, certain pesticides, some metals, some acids) origin. Often the condition develops only Emphysema is another chronic condition in re- after a period of sensitization, and for some sponse to many different stimuli. Though there agents, very high percentages of those exposed be- are undoubtedly cases of occupational origin, few come sensitized. It has been reported that nearly convincing, direct correlations between workplace all workers in power plants along the Mississippi exposures and this disease are known. River become sensitized to river flies (396). Beryllium disease (berylliosis) is an example of The causes of hypersensitivity pneumonitis in- granuloma formation in response to foreign clude a variety of organic materials, commonl bodies in the lungs. Granulomas form when body y fungi or bacteria. Beginning with coughing, but cells responding to an “inciting agent” become sur- without the wheezing associated with asthma, rounded by bundles of collagen (a type of con- these disorders can become chronic and disabl- nective tissue). ing. Such conditions as “farmer’s lung, ” “mush- room picker’s lung, “ “cheese washer’s lung, ” and Acute Conditions “paprika splitter’s lung” fall into this category. Inflammations and irritations of the tissues lin- Byssinosis deserves particular recognition. (For ing the respiratory tract occur in response to many further discussion of this disease, see ch. 5.) inhaled substances. The upper respiratory tract— Though it has been known in some sense as a dis- —

Ch. 3—Health Hazard Identification ● 45 ease associated with cotton and other textile fibers strenuous, awkward tasks. The prevalence of for hundreds of years, it was ignored as an occu- repetitive motion disorders is unknown, but more pational disease in this country until fairly and more industries are recognizing that they have recently. The disease begins with tightness in the such problems. chest and a decrease in lung capacity upon exposure. The condition is most severe on Monday mornings. Over a period of years, chronic ob- Cancer structive lung disease may develop, partially or Table 3-2 is a list of recognized occupational totally disabling the worker. The earlier stages of cancer hazards. In most cases, there is convinc- the disease are thought to be reversible, but the ing or very strong evidence that the listed sub- later stages are not. The exact etiologic agent of stances have caused cancer in humans. Inspection byssinosis is not known, but various chemicals of the table shows that many of these substances and organic substances have been suggested. cause common cancers, for instance, of the lung and skin. Except for a few specific and infrequent Musculoskeletal Disorders cancers, there is no way to tell, from examining a cancer patient, what agent(s), exposure(s), or Low back pain is responsible for more lost behavior(s) caused the tumor. work-time than any other medical condition except upper respiratory tract ailments. In terms of The most detailed information about an occu- treatment and workers’ compensation, low back pational cancer hazard involves asbestos. The un- pain is the costliest occupational ailment. More folding of that story illustrates the time necessary than half of all workers will experience low back for association to be accepted and some contro- pain of some kind sometime in their working lives, versies about occupational illness. Individual case but the percentage of those cases associated with studies and reviews of case series relating expo- the workplace is unclear. sures to asbestos with cancer began to appear in the literature in the 1930s (161). According to Low back pain may develop progressively and Selikoff (430), however, the establishment of an insidiously, or it may come on with immediacy. association between occupational exposure to as- Pain may be dull and aching, with fatigue and bestos and lung cancer depended on a classic study stiffness, or sharp and crippling. Surprisingly lit- by Doll in 1955 (147). tle is known about the physiologic and physical causes underlying back pain. Circumstantial evi- Although asbestos was positively identified as dence implicates intervertebral discs in many a cause of lung cancer in the 1950s, and exposure cases. (Discs are cartilaginous structures separat- to it was known to be widespread, no published ing the vertebrae of the spine. ) In extreme cases, estimate of its impact on nationwide mortality a disc may rupture, but physical signs that would was available until 1978, when two estimates were explain the pain are usually absent. Episodes of made. Selikoff (555) estimated that the annual pain, which last usually from a few days to a few number of asbestos-related cancer deaths was weeks, generally resolve with rest. Months or about 50,000. His estimate elicited little public at- years may pass without another attack. tention. Muscles, tendons, ligaments, and bones can The other 1978 estimate (555), entitled “Esti- also be damaged by traumatic events or by re- mates of the Fraction of Cancer in the United peated strains over a long period. Although mus- States Related to Occupational Factors, ” was pre- cle pulls and tears have been recognized for years, pared by the National Cancer Institute, the Na- the “repetitive motion disorders"—those caused tional Institute for Environmental Health Sciences,

b y repeated, often forceful motions, mainly of and NIOSH. Ten employees of those institutions parts of the arm —have come to attention more were listed as contributors to the “estimates pa- recently (see “Carpal Tunnel Syndrome, ” ch. 7), per, ” which was placed in an Occupational Safety Much assembly-line work and food processing, and Health Administration (OSHA) hearing rec- for example, is characterized by repetitive, ord about that agency’s proposed generic cancer 46 . Preventing Illness and Injury in the Workplace

Table 3=2.—Some Occupational Cancer Hazards

Agent Cancer site or type Type of workers exposed Acrylonitrile ...... Lung, colon Manufacturers of apparel, carpeting, blankets, draperies, synthetic furs and wigs 4-aminobiphenyl ...... Bladder Chemical workers Arsenic and certain arsenic compounds ...... Lung, skin, scrotum, Workers in the metallurgical industries, sheep-dip workers, lymphatic system, pesticide production workers, copper smelter workers, hemangiosarcoma of the vineyard workers, insecticide makers and sprayers, liver tanners, miners (gold miners) Asbestos ...... Lung, larynx, GI tract, Asbestos factory workers, textile workers, rubber-tire pleural and peritoneal manufacturing industry workers, miners, insulation mesothelioma workers, shipyard workers Auramine and the manufacture of auramine ...... Bladder Dyestuffs manufacturers, rubber workers, textile dyers, paint manufacturers Benzene...... , . . Leukemia Rubber-tire manufacturing industry workers, painters, shoe manufacturing workers, rubber cement workers, glue and varnish workers, distillers, shoemakers, plastics workers, chemical workers Benzidine...... Bladder, pancreas Dyeworkers, chemical workers Beryllium and certain beryllium compounds . . . . Lung Beryllium workers, electronics workers, missile parts producers Bis(chloromethy l)ether (BCME) ...... Lung Workers in plants producing anion-exchange resins (chemical workers) Cadmium and certain cadmium compounds . . . . Lung, prostate Cadmium production workers, metallurgical workers, electroplating industry workers, chemical workers, jewelry workers, nuclear workers, pigment workers, battery workers Carbon tetrachloride ...... Liver Plastic workers, dry cleaners Chloromethyl methyl ether (CMME) ...... Lung Chemical workers, workers in plants producing ion- exchange resin Chromium and certain chromium compounds. . . . Lung, nasal sinuses Chromate-producing industry workers, acetylene and aniline workers, bleachers, glass, pottery, pigment, and linoleum workers Coal tar pitch volatiles . . . . . Lung, scrotum Steel industry workers, aluminum potroom workers, foundry workers Coke oven emissions ...... Lung, kidney, prostate Steel industry workers, coke plant workers Dimethyl sulphate ...... Lung Chemical workers, drug makers, dyemakers Epichlorohydrin ...... , . Lung, leukemia Chemical workers Ethylene oxide ...... Leukemia, stomach Hospital workers, research lab workers, beekeepers, fumigators Hematite and underground hematite mining ...... Lung Miners Isopropyl oils and the manufacture of isopropyl oils ...... Paranasal sinuses isopropyl oil workers Mustard gas ...... Respiratory tract Production workers 2-naphthyiamine ...... Bladder, pancreas Dyeworkers, rubber-tire manufacturing industry workers, chemical workers, manufacturers of coal gas, nickel refiners, copper smelters, electrolysis workers Nickel (certain compounds) and nickel refining ...... Nasal cavity, lung, larynx Nickel refiners Polychlorinated biphenyls (PCBs) ...... Melanoma PCBS workers

Ch. 3—Health Hazard Identification . 47

Table 3-2.—continued

Agent Cancer site or type Type of workers exposed Radiation, ionizing ...... Skin, pancreas, brain, Uranium miners, radiologists, radiographers, luminous dial stomach, breast, salivary painters glands, thyroid, GI tract, bronchus, Iymphoid tissue, leukemia, multiple myeloma Radiation, ultraviolet ...... Skin Farmers, sailors, arc welders Soots, tars, mineral oils . . . . Skin, lung, bladder, GI Construction workers, roofers, chimney sweeps, machinists tract Thorium dioxide ...... Liver, kidney, larynx, Chemical workers, steelworkers, ceramic makers, leukemia incandescent lamp makers, nuclear reactor workers, gas mantle makers, metal refiners, vacuum tube makers Vinyl chloride ...... Liver, brain, lung, Plastics factory workers, vinyl chloride polymerization plant hematolymphopoietic workers system, breast Agent(s) not identified . . . . . Pancreas Chemists Stomach Coal miners Brain, stomach Petrochemical industry Hematolymphopoietic Rubber industry workers system Bladder Printing pressmen Eye, kidney, lung Chemical workers Leukemia, brain Farmers Colon, brain Pattern and model makers Esophagus, stomach, lung Oil refinerv workers SOURCE (542)

Photo credit OSHA, Office of Inforrnation and Consumer Affalrs Shipbuilding operations present a variety of both safety and health hazards. During World War 11, many workers were exposed to asbestos in naval shipyards 48 ● preventing Illness and Injury in the Workplace policy (the “estimates paper” is available as an ap based on the method used originally in the “esti- pendix in Peto and Schneiderman (371)). mates paper. ” Calculations based on numbers of mesotheliomas and asbestosis produced estimates Had the paper been only deposited in the hear- of between 1 and 2 percent of all cancer deaths ing record it might have passed largely unnoticed. being due to asbestos (148,212,294,370). Its findings, however, were widely publicized when then-Secretary of Health, Education, and The consistency of the projections that asbestos Welfare Joseph Califano cited them in a speech. causes between 1 and 3 percent of current cancer Based on the “estimates paper, ” he stated that deaths (190) has a pronounced effect on estimates workplace exposures caused at least 20 percent of total occupationally related cancers. Most, but of all cancer in this country-with exposure to not all, participants at an international conference asbestos alone responsible for 13 to 18 percent. about occupational cancer agreed that workplace These projections were controversial as soon as exposures cause less than 5 percent (20,000 an- they were publicized, and they attracted many nual deaths) of U.S. cancer mortality (371). critics. They also resulted in a spate of articles pre- Although this number is not as frightening as senting other estimates of the cancer risk associ- saying asbestos causes 13 percent of cancer and ated with occupational exposure to asbestos. that workplace exposures cause at least 20 per- The subsequent papers can be divided into two cent, and perhaps twice that figure, it is still a large general groups. One group used methods similar number of deaths. Furthermore, as representatives to the “estimates paper” to project numbers of can- from all sides—academe, labor, and manage- cer deaths based on estimates of workers exposed, ment—agree, those cancers are preventable. exposure rates, and mortality observed among in- The amount of cancer that is associated with sulation workers highly exposed to asbestos. A workplace exposures is a significant part of the second type of paper measured the number of current debate about the relative importance of deaths from mesotheliomas, which are closely various factors in cancer causation (see (18), the associated with asbestos exposure, and then exchange of facts and opinions in ‘letters” Science multiplied that number by some factor to estimate 224:659 et seq., especially (154) and (19)). all asbestos-caused cancer deaths. Methods similar to those employed in the 1978 paper generated three estimates of total asbestos Reproductive Disorders cancer mortality. Those estimates, lower than the The possibility that people’s occupations are 13 percent figure in the “estimates paper, ” were leading to problems for an unborn generation is 1 percent (162), 2 percent (216) and 3 percent frightening. It is increasingly a concern among (331). The different numbers reflect the authors’ workers, and attention to reproductive disorders different estimates about the numbers of heavily on the part of scientists is intensifying. Few facts exposed workers—estimates that can be criticized are available to either support or quell the fears because they were not made on the basis of ac- that a great many reproductive hazards are pres- tual measurements. As that sort of information ent in the workplace. Relatively few instances of does not exist, however, documented assumptions harm are known when compared with the known are the best that can be provided. effects of workplace hazards on workers them- In the case of asbestos, scientists interested in selves. extrapolating from study-generated data to esti- Initial concerns about reproductive health fo- mates of national cancer mortality are aided by cused almost exclusively on women. Exposure to the fact that asbestos causes asbestosis and meso- the high levels of lead common at the beginning theliomas. Both those diseases are reasonably rare of this century were known to cause menstrual and reasonably diagnostic for asbestos exposure. disorders, sterility, miscarriages, and stillbirths. Although both are subject to undercounting that limits the accuracy of estimates based on them, Much more recently, concern has been extended the estimates from them are congruent with those to males. One episode provided the catalyst. In —

Ch. 3—Health Hazard Identification ● 49 the late 1970s, a number of men working in the Lead is the best-known neurotoxin in the work- manufacture of dibromochloropropane (DBCP), place. More than a million American workers are a pesticide, were unable to father children, In- exposed currently. Mercury, manganese, and vestigation revealed severely depressed sperm pro- other metals, as well as organic solvents and duction. organophosphate insecticides, also pose neurotoxic risks. Table 3-3 lists some known neurotox- Damage can occur in males and females in a ins and their effects. number of ways. In men, successful reproduction depends on proper functioning of the prostate, on Neurotoxins can damage the myelin sheath sur- libido, and on erection and ejaculation. The pro- rounding the nerve fiber or the nerve cell itself. duction and viability of sperm can be affected by Toxins can also interfere with the production and damage to the sperm-producing cells or to the functioning of “neurotransmitters,” chemicals pro- sperm as they develop. duced in the body that are necessary for proper functioning of the nerves. ’some necrologic im- In women, damage can occur in the reproduc- pairment is reversible, but damaged nerve cells tive cells, the oviducts, the endometrium, or to have limited capacity for regeneration and repair. ovarian function. During fetal development in the uterus, humans are most vulnerable to environ- Neurotoxins affect the parts of the nervous sys- mental insults. Death, irreversible structural tem to different degrees. The most commonly changes (teratogenesis), and growth retardation affected are peripheral nerves-those of the ex- are the main classes of effects. More difficult to tremities. Hands and feet are often the first measure or prove are subtle deficits in intellec- symptomatic zones, and numbness and tingling tual capacity and functioning. the first signs. Weakness in the hands and feet follows, and then difficulty walking and an inability The effects of lead have been mentioned. At to grasp heavy objects. Other symptoms include least one form of another heavy metal, mercury, impaired vibratory sense, loss of touch percep- is a known teratogen. Certain pesticides—DBCP tion, and tremors of the hand and other parts of and Kepone for instance-affect sperm produc- the body. tion. Ionizing radiation has a variety of effects, particularly on fetuses—causing growth retarda- A host of behavioral changes can also result tion, for instance, or microcephaly, or having from necrologic insults: Slow response time, im- latent effects, such as leukemias that develop dur- paired hand-eye coordination, irritability, lack of ing childhood. A few organic solvents and phar- concentration, continual emotional instability, maceuticals also are known to affect reproduc- and impairment of recent memory area few such tive health. In all, relatively little is known about signs. (Lewis Carroll’s “Mad Hatter” may have the extent of workplace-induced reproductive been a victim of the necrologic effects of mercury damage, but efforts to find out more are under used in making felt hats. ) way. A current OTA assessment scheduled for Most neurotoxins act through common path- completion in 1985, “Reproductive Health Haz- ways, though some have more specific effects: ards in the Workplace, ” addresses this issue. Carbon disulfide, for instance, acts at all levels on the central nervous system, but also causes Necrologic Disorders conditions as extreme as acute psychosis. A wide variety of metals and organic compounds act on the nervous system to cause phys- Skin Disorders ical and behavioral problems. Since many bodily functions require the participation of nerves, The skin, the largest organ of the body, pro- nerve impairment affects not only sensory abili- vides the first line of defense between workers and ties, but motor (muscular) ability as well as the their environment. Because it is readily observable, functioning of organs. recognition of a problem is relatively easy. For 50 . Preventing Illness and Injury in the Workplace

Table 3-3.—Neurologic Effects of Occupational Toxins

Peripheral Effects Effect Toxin Comments Motor neuropathy ...... Lead Primarily wrist extensors; wrist drop and ankle drop rare Mixed sensorimotor neuropathy . . . . Acrylamide Ataxia common; desquamation of hands and soles; sweating of palms Arsenic Distal parethesias earliest symptom; painful limbs, especially in calves; hyperpathia of feet; weakness prominent in legs Carbon disulfide Peripheral neuropathy rather mild; CNS effects more important Carbon monoxide Seen only after severe intoxication DDT Only seen with ingestions N-hexane and methyl n-butyl ketone (MBK) Distal paresthesia and motor weakness; weight loss, fatigue, and muscle cramps common Mercury Predominantly distal sensory involvement Other Manifestatlons Manifestation Agent Manifestation Agent Ataxic gait , ...... Acrylamide Increased intracranial Chlordane pressure ...... Lead Chlordecone (Kepone) Organotin compounds DDT Myoclonus ...... Benzene hexachloride N-hexane Mercury Manganese Nystagmus ...... Mercury Mercury (especially with Opsoclonus ...... Chlordecone (Kepone) methyl mercury) Paraplegia...... Organotin compounds Methyl n-butyl ketone (MBK) Parkinsonism ...... Carbon disulfide Methyl chloride Carbon monoxide Toluene Manganese Bladder neuropathy...... Dimethylaminopro pionit ri te Seizures ...... Lead (DMAPN) Organic mercurial Constricted visual fields . . Mercury Organochlorine insecticides Cranial neuropathy ...... Carbon disulfide Organotin compounds Trichloroethylene Tremor...... Carbon disulfide Headache ...... Lead Chlordecone (Kepone) Nickel DDT Impaired visual acuity ... , N-hexane Manganese Mercury Mercury Methanol

NOTE: This table Includes most, but not all, of the neurotoxic substances associated with Ilsted conditions. SOURCE: (39a) both these reasons, skin disorders account for both biological and chemical, skin diseases are nearly half of all reported occupationally related manifested in a relatively limited number of clin- illnesses in the United States. NIOSH has recom- ical symptoms: contact dermatitis, infection, pilo- mended maximum exposure levels for about 40 sebaceous follicle abnormalities, pigment dis- agents based on their effects on the skin (see table orders, and cancers. 3-4). Contact dermatitis accounts for 90 percent of all occupational skin disorders. The most com- Chemical, physical, and biological agents, mechanical factors, and plant and wood substances mon manifestations of contact dermatitis are red- ness and swelling, and vesiculation (e.g., a poison are known to cause occupationally related skin ivy rash) in more severe cases. Contact dermatitis disorders (see table 3-5). There is probably no in- may be an allergic reaction or simply due to an dustry without some potential for exposure to one irritant. or more of these agents. The industries with the highest risk for skin disorders are listed in table Bacterial, fungal, and viral infections may be 3-6. Although caused by a large number of agents, contracted from customers or clients by such pro- Ch. 3—Health Hazard Identification ● 51

Table 3-4.—Substances for Which Table 3-5.—Workpiace Agents That NIOSH Has Recommended Exposure induce Skin Disorders Limits to Prevent Skin Disorders Chemical agents Acrylamide Rhus oleoresin (poison ivy and oak) Alkanes: Acids Pentane Alkalis Hexane Solvents Heptane oils Octane Soaps and detergents Arsenic, inorganic compounds Plastics Benzoyl peroxide Resins Benzyl chloride Paraphenylenediamine Carbon black Chromates Chromium (Vi) Acrylates Coal tar products Nickel compounds Cresol Rubber chemicals Epichlorohydrin Petroleum products not used as solvents Ethylene dibromide Glass dust Fibrous glass (dust) Plant and Wood Substances Glycidyl ethers: Pink rot celery Allylglycidyl ether (AGE) Citrus fruit n-Butyl glycidyl ether (BGE) Physical Agents Di-2-,3-epoxypropyl ether (DGE) Ionizing and nonionizing radiation Isopropyl glycidyl ether (lGE) Wind Phenyl glycidyl ether (PGE) Sunlight Hydrazines: Temperature extremes Hydrazine Humidity 1,1-dimethyl hydrazine Phenyl hydrazine Biological Agents Methyl hydrazine Bacteria Hydrogen fluoride Viruses Hydroquinone Fungi Nickel, inorganic and compounds Ectoparasites (mites, ticks, fleas, etc.) Phenol Biting animals Polychlorinated biphenyls: Mechanical factors Chlorodiphenyl (42°/0) Pressure Chlorodiphenyl (54°/0) Friction Refined petroleum solvent Vibration Thiols: SOURCE: (23a). Butyl mercaptan (1-butanethiol) Methyl mercaptan (1-methanethiol) Ethyl mercaptan (1-ethanethiol) tion is common in workers, such as dishwashers, Tin, organic compounds who are frequently exposed to water, because Tungsten: insoluble compounds moist conditions favor the fungus’ growth. Viral soluble compounds infections are acquired by contact with other peo- Vanadium ple and are a particular hazard for workers ex- SOURCE Adapted from (12Sa) posed intimately to other individuals in the course of their work, such as health care workers. fessionals as barbers and hairdressers and by hos- Pilosebaceous follicle abnormalities, generally pital workers. Staphylococcus and streptococcus acne-like lesions, occur after exposures to heavy bacteria may cause a range of skin conditions oils and certain chemicals, particularly chlorinated from superficial to those of deep skin layers. More aromatic hydrocarbons. The example currently serious bacterial infections, such as anthrax in most discussed is chloracne after exposure to sheep handlers and animal hide workers, are chlorinated dioxins, either in the manufacturing rarer. process, or, most dramatically, after industrial ac- Fungal infections often arise in moist, warm cidents involving the generation and release of environments. Ringworm and Candida albicans large amounts of the chemical. Chloracne may infections are common examples. Candida infec- persist for 10 yearn or more after exposure ceases. 52 . Preventing Illness and Injury in the Workplace

Table 3-6.—industries at Highest Risk for Occupational Skin Diseases

Annual reported incidence rate Target population Incidence of lost (per 1,000 (rounded to workdays per Industry workers) x nearest 1 ,000) x Severity a x Durationb = industry per year Poultry dressing plants 16,4 89,800 030 10.0 4,405 Meat packing plants 72 164,300 0.31 4,3 1,561 Fabricated rubber products 55 103,200 0.22 11,5 1,424 Leather tanning and finishing 21,2 22,900 0.34 8.3 1,392 Ophthalmic goods 8.5 38,000 0.52 8.3 1,390 Plating and polishing 8.3 61,400 0.28 9.0 1,270 Frozen fruits and vegetables 7,2 43,200 0.31 12,1 1,153 Internal combustion engines 5.5 75,700 027 8.8 999 Canned and cured seafoods 5.6 19,700 0,36 23,7 934 Carburetors, pistons, rings, valves 70 29,400 024 17.9 895 Chemical preparations 8.3 36,700 0.23 12.3 855 Boat building and repairing 11.1 48,000 0.22 7,4 854 aseverlty ,s defined by number Of lost-workday cases dwlded by fOtd Wnbw Of Cases HI that Industry bDura[lon IS [he number of lost workdays per IOSt workday case SOURCE (23a)

Pigment disorders occur when melanin produc- versible or not, depending on the causative agent tion is either increased or decreased through ex- and on the severity of the insult. Other changes posures to chemicals or from a traumatic event—a in skin color are due to staining of various layers burn, for instance. Loss of pigment may be re- by such substances as heavy metals.

KNOWN AND UNKNOWN HEALTH HAZARDS

Health hazards are agents that can cause dis- cause a variety of diseases, and energy from all ease in people exposed to them. In terms of oc- parts of the electromagnetic spectrum is a hazard cupational health, there are three kinds: under particular circumstances. As the following examples show, associations between agents and ● identified hazards known to be present in the diseases have been made by people from all sec- workplace; tors of society based on laboratory information ● hazards that are present in the workplace but as well as observations of human illness. that have not been identified as causes of disease; and ● new substances or processes not yet introduced Physical Agents into the workplace, that will be hazardous to Sources of ionizing radiation are increasingly human health. common in the workplace. X-ray apparatus and This section reviews the findings that led to radioisotopes are widely used, and nuclear power- some associations being made between particu- plants and scientific research also involve poten- lar diseases and workplace hazards, as well as the tial exposures to ionizing radiation. Very high methods currently employed to identify hazards. doses of radiation can kill workers within a few days, but of greater concern, because the events Identified Hazards are more likely, is low-level exposures, which may last for several years and may cause cancer. The Diseases associated with mining and metal- deleterious effects of radiation were discovered working have been recognized for many years, from observations of disease among early work- to some extent because of the antiquity of those ers in the field and confirmed by analyses of the trades. Some industrial chemicals are known to survivors of Nagasaki and Hiroshima. Ch. 3—Health Hazard Identification 53 —

Nonionizing radiations include ultraviolet, infrared, microwave, and laser. All present hazards Box A.—’‘Phossy Jaw” for workers’ eyes, and there is continued interest Phossy jaw was a disease that resulted from in and study about other effects from microwave inhaling yellow or white phosphorus fumes that radiation. Ultraviolet and infrared radiation as penetrated any defective tooth and killed cells well as intense visible light are generated in weld- in the jaw and surrounding tissues. Invasion of ing, and welders’ goggles and helmets are designed the dead areas by germs from the mouth led to to protect against such hazards. suppurating infection, swelling, and intense pain. Death could result from blood poisoning; surgi- Also in the category of physical agents is noise, cal treatment, which often included removal of which, especially if it is loud and continuous, the jaw, was incapacitating and disfiguring. The causes progressive hearing loss. The impact of oc- disease was first diagnosed in workers in Euro- cupational noise is difficult to separate from the pean match factories in the middle of the 19th effects of aging, but many studies have shown century. workplace noise is a hazard to hearing. (See ch. Up through 1908, there was no recognition of 8 for a discussion of the role of personal protec- phossy jaw as an occupational health problem tive equipment in preventing hearing loss. ) in the United States. A Bureau of Labor study that year of the wages of women and children Vibration, often experienced as a result of the in the match industry revealed 150 cases of use of handtools, causes a number of musculo- phossy jaw. Two years later, the Bureau issued skeletal disorders (see ch. 7). “Phosphorus Poisoning in the Match Industry in Heat, cold, and pressure encountered in under- the United States.” water work are also hazards. These have been One of the surest forms of controlling expo- associated with particular jobs for a very long sures to hazardous substances is to substitute a time, and many of their effects are visible during less hazardous chemical. Phossy jaw was con- or soon after exposure. quered by substituting a different form of phosphorus for the “white phosphorus” commonly Metals used in matches. The Diamond Match Co., which held the American patent for the safe form Hunter (218) divides hazardous metals into (sesquisulphide), waived its patent rights and three groups. Those known since ancient times, made the safe substitute available to the entire such as lead and mercury, were long ago associ- industry (199). In 1912, Congress passed the ated with disease. According to Hamilton (1922, “Esch Act,” which levied a tax on white phos- quoted in 218), the first legislation directed against phorus matches, driving them from the market. an occupational hazard was drafted in 1665 in Idria, now part of Yugoslavia. The workday for cinnebar (mercury ore) miners was restricted to 6 hours as a preventive measure to reduce the of phosphorus in matches. These three metals still occurrence of tremors. Mercury continues to occupy important places in industry and in agri- cause concern today as an environmental con- cultural products. taminant, and it is especially dangerous in the The third group of metals are those recently inorganic (methylmercury) form. troduced into commerce, including some impor- Hunter’s second group, the “other metals, ” are tant in advanced metallurgic technologies and the arsenic, phosphorus, and zinc. He points out that nuclear industry. Toxic effects are definitely asso- the grouping is arbitrary in that arsenic is a ciated with some—beryllium, cadmium, chro- metalloid and phosphorus a nonmetal. These mium, manganese, nickel, osmium, platinum, ra- three elements have been in common industrial dium, ruthenium, selenium, tellurium, thallium, use for a few centuries, and all have caused ill- thorium, uranium, and vanadium. In some meas- ness and death. The recognition of phosphorus ure, because these substances were introduced into as the cause of “phossy jaw” among matchmakers the workplace when industrial hygiene measures (see box A) led to the substitution of a safe form were more common, exposure to many of them 54 ● Preventing Illness and Injury in the Workplace has been well controlled (218). Also important to Environmental Protection Agency’s (EPA) inven- controlling exposures to some of these metals is tory of Chemical Substances, which is a compila- their great expense; uncontrolled losses through tion of chemicals in commerce. About 100 new spills or into the atmosphere as vapors, fumes, chemicals are introduced to commerce each month or dusts entail financial losses as well as health (547a). Many of these substances–pesticides of hazards. various kinds and drugs-are designed to alter normal biological functions, and it is no surprise Many metals are worked in industry with no that some have been found to cause cancer and reported toxic effects. Cesium, cerium, colum- other diseases, and that these substances are of bium, gallium, germanium, hafnium, iridium, special concern (542). lanthanum, molybdenum, rhenium, rhodium, rubidium, strontium, tantalum, titanium, tung- Some of the now-known hazards, such as vinyl sten, and zirconium, for example, have not been chloride monomer, have been discovered as a re- associated with illness in workers (218). Exposures sult of workers who have become sick. (See ch. to many such metals are controlled by standard 5 for a fuller discussion. ) Several years before an industrial hygiene practices, and the fact that some alert physician noted an excess of rare liver tumors of these metals are very expensive also encourages in vinyl chloride workers, the results of an ani- reduced exposures. mal test of the same chemical were announced at a scientific meeting. The animal tests also showed Hunter (218) is a good source of historical in- the chemical to be a liver carcinogen. It can be formation about the uses and effects of the various argued that had the animal results been taken seri- metals and about British approaches to control- ously, exposure to vinyl chloride would have been ling exposures. Rem, et al. (396), discusses clini- reduced sooner. As it happened, the existence of cal symptoms and treatments as well as U.S. ap- the animal studies may have been a factor in the proaches to control, and Levy and Wegman (269) rapid regulatory process that led to significant re- provide a lively introduction to the occupational ductions in vinyl chloride exposures. health and industrial hygiene problems associated with the metals, with less emphasis on clinical Acrylonitrile is a commonly used plastic that, detail than Rem. Tyrer and Lee (483) summarize like vinyl chloride, presents little hazard after it information about acute and chronic health ef- is polymerized. However, animal studies showed fects of the metals and list recommended and reg- that acrylonitrile monomers are carcinogenic, and ulatory limits to exposure. a follow-on epidemiologic study showed an excess of cancer among acrylonitrile production Dusts and Fibers workers. Regulations restricting exposures to the substance were drafted by OSHA; unlike most The hazards of mineral dusts have been known other OSHA health regulations, the final stand- since mining began. Both silica dust and coal dust ard for acrylonitrile was not challenged in court. cause lung diseases. The widespread use of silica There must have been a number of reasons for as an abrasive for “sand blasting” and other that success, and included in them were probably polishing results in many thousands of American the congruence between the results of the animal workers being exposed to mineral dusts that are and human studies and the fact that the methods associated with lung diseases. In addition, cotton developed to control vinyl chloride exposures dust and asbestos are important as causes of bys- were directly applicable to the control of acrylo- sinosis and asbestosis, respectively. nitrile. Chemicals Methods for Detection of Present, Because of the explosion of organic chemistry Unidentified Hazards (chemistry that involves carbon) in the last 100 years, thousands of new chemical substances have Epidemiology, toxicology, and occupational been introduced into the workplace. Currently medicine provide the means for identifying the there are more than 55,000 chemicals listed in the causes of occupational illnesses. In the traditional, Ch. 3—Health Hazard Identification 55 idealized view of the process, physicians gener- stances that have not yet been tested. Toxicology ate hypotheses about possible associations be- can be subdivided in a number of ways. Here, tween workplace exposures and subsequent dis- testing for acute toxicities is discussed first, fol- ease. Hypotheses are tested in epidemiologic lowed by a section on methods for investigating studies so that the associations can be character- chronic toxicities—carcinogenesis, mutagenesis, ized in statements of statistical probability. and teratogenesis. The traditional role of toxicology has been to Acute Toxicity Testing. —Chemical burns and provide information about the mechanisms of dis- immediate difficulty in breathing as a result of in- ease causation, the end results of which are halation of a substance are examples of acute toxic detected by physicians and studied by epidemiol- effects. Animal testing of chemicals for toxicity ogists. Toxicology today is generally thought of has produced a voluminous data set. in different terms. Since the late 1960s and particularly through the 1970s, toxicology has been Increasing concern about animal welfare is seen as a way to identify chemical hazards before causing reconsideration of animal testing meth- their effects appear in humans. The most visible ods. For instance, one of the most venerable acute toxicologic activities are the testing of chemicals toxicity tests is the LD50 test. Designed in the for carcinogenic properties in laboratory animals, 1920s, the test involves the use of 50 to 100 ani- mainly rats and mice (542). The Federal Govern- mals to decide what amount of substance will ment, through the National Toxicology Program, cause the death of 50 percent of the animals. This spent $31.6 million in 1983 on bioassays for that method is coming under increased attack, how- purpose. ever, as being imprecise and causing more animal suffering than is necessary. OTA is studying There is also a certain amount of research now the use of alternatives to animals in research and going on in development of short-term tests (so testing. The report from that project, expected in named because they require significantly less than 1985, will discuss the pros and cons of various the 2 to 5 years for an animal bioassay) as even- animal tests and alternatives to animal tests. tual replacements for and supplements to bioassays, One of the most powerful methods of identify- NIOSH’s 1980 Registry of Toxic Effects of ing associations between workplaces and diseases Chemical Substances lists 45,156 substances. In- is through workers themselves. For instance, the cluded for most of the substances is the LD50 esti- pesticide dibromochloropropane was identified as mate of the amount that will kill half of a popula- a cause of male sterility by workers talking to each tion of test animals. In addition, information other. A possible relationship between office work about the toxic effects of the substance on ani- involving video display terminals and fetal mal- mal skin and eyes is also commonly reported. formations that is now being actively investigated Dosages of ingested or injected substances nec- similarly derives from workers’ observations. In essary to cause effects in animals are expressed many cases, workers’ comments to their physi- as the weight of the substance administered cians lead to epidemiologic and toxicologic in- divided by the animal’s body weight, i.e., milli- vestigations and to medical surveys to decide grams of substance/bod weight in grams or whether a suspected association is real. y kilograms. When the substance is inhaled, the dangerous concentrations are expressed as parts Toxicology per million in air or as the weight of the substance Toxicology is the testing of chemicals in ani- per cubic meter of air. These values provide data mals, plants, or lower forms of life to detect for making estimates of the biological effects of biological effects. In addition to questioning what the substance in humans. Almost always, safety kinds of effects are produced and under what ex- factors of 10 or 100 are used in setting acceptable posure conditions, toxicologists also investigate limits for workers. That is, if 100 parts per mil- the mechanisms by which substances cause dam- lion of a substance causes breathing difficulties age. That information is especially important in in animals, a prudent policy would be to limit efforts to predict the likely toxic effects of sub- worker exposures to 10 or 1 part per million. 56 . Preventing Illness and Injury in the Workplace

Chronic Toxicity Testing.—Structural activity 1) Structural activity relationship analysis, SAR relationship (SAR) analysis, chronic animal bio- uses known information about the properties of assays, and short-term tests are the main tools of a substance to gain insight into the possible and toxicology (see table 3-7) as it relates to car- probable effects of the substance on human be- cinogens, and, in general, chronic health hazard ings. It is a new and still uncertain technique. Sub- identification (542,547a). Finding a toxic effect in stances whose molecular structures resemble those humans is far more convincing evidence about the of known toxic substances come under greater sus- seriousness of a hazard than detecting a toxic ef- picion than those whose structures do not. No fect in animals, which, in turn, is more convinc- firm conclusions can be made based on these anal- ing than results from short-term tests. The weakest yses except in the rare cases where all previously evidence is that derived from projections from known members of an entire class of chemicals structural activity relationships. Although the are known to be hazardous. In general, positive Federal effort devoted to chronic toxicities-muta- results are taken to indicate a need for further genicity and teratogenicity as well as carcino- testing, genicity—is largely directed toward identifying SAR has found most use in making estimates carcinogens, there are some minor stirrings of ef- of the toxicity of “new” chemicals, when no test fort to broaden beyond cancer (595). data are available. However, even there the scien- Carcinogenicity has received the lion’s share of tific underpinnings of SAR are considered by some OSHA’s attention to health hazards. Of the fewer to be very weak, and the conclusions based on than two dozen chemicals regulated through new, it are hotly argued (547a). permanent OSHA standards, all but two—lead 2) Short-term tests. Short-term tests encompass and cotton dust—have been carcinogens. a large collection of methods for measuring tox- Extrapolation problems—that is, how knowl- icity in lower life forms-viruses, bacteria, and edge of effects in animals are projected to make lower plants and animals, such as fruit flies-in predictions for people and how exposure levels cultured cells, or, in a few cases, in specific organ in test animals are related to human exposure systems of laboratory rodents (542). Since their levels–bedevil the use of animal test data. OTA introduction about 15 years ago, they have been (542) has already discussed those problems and characterized as holding great promise for tox- various approaches to reconciling them. icology. A cynic might say that they always will.

Table 3-7.–General Classification of Tests Available to Determine Properties Related to Carcinogenicity

Method System Time required Basis for test Structural activity relationship (SAR) analysis ‘Paper chemistry” Days Chemicals with Iike Structure resembles Chemical may be structures interact (positive) or does not hazardous; that Basic laboratory tests Weeks similarly with DNA resemble (negative) determination requires structure of known further testing carcinogen Short-term tests Bacteria, yeast, cultured Generally few weeks Chemical interaction with Chemical causes Chemical IS a potential cells, intact animals (range 1 day to 8 DNA can be measured (positive) or does not carcinogen months) m biological systems cause (negative) a response known to be caused by carcinogens Bioassay Intact animals (rats, 2 to 5 years Chemicals that cause Chemical causes Chemical IS recognized mice) tumors m animals (positive) or does not as a carcinogen in may cause tumors in cause (negative) that species and as a humans Increased incidence of potential human tumors carcinogen Epidemiology Humans Months to Iifetimes Chemicals that cause Chemical IS associated Chemical IS recognized cancer can be (positive) or is not as a human detected m studies of associated (negative) carcinogen human populations with an increased incidence of cancer SOURCE Adapted from (542) Ch. 3—Health Hazard Identification 57

Running counter to that lack of enthusiasm, re- 3) Bioassays. The bioassay is the mainstay of cent spectacular advances in molecular biology toxicology today. For some questions, answers suggest that short-term tests will grow in impor- involving the biology of whole animals are essen- tance. As more and more insight into the molecu- tial. The technique involves exposing a popula- lar basis of carcinogenesis accumulates, along with tion of laboratory animals, usually rats and mice, rapid advances in methods to manipulate DNA to a suspect toxic agent. After an appropriate and other cellular components (542,548), im- time, about 2 years for carcinogenicity, the dis- proved short-term tests should follow. The limita- ease incidence in the the treated population is tions and uncertainties of testing substances in compared with the disease incidence in a popula- whole-animal bioassays are built into the method tion of untreated controls. The premise underlying itself. No such limits bound potential short-term the mammoth effort in bioassays is that evidence tests for discerning interactions between chemi- of disease in animals is applicable to predictions cal and cellular components. Of course, it will for people; in fact, substances known to be car- always be possible to argue that the short-term cinogens in humans also cause cancer in animals. test system is not sufficiently parallel to human biology to serve as a guide to human risk esti- An entire branch of risk assessment has grown mation. up around the quantitative predictions of effects The critical issue for development of short-term in human beings based on animal evidence, In the tests is defining their current and ultimate value combination of bioassay and risk assessment has in policymaking. The first step is to find out how lain the hope of perfectly protecting workers and well the results of a test represent the “truth,” a the public from chemical carcinogenesis before ef- process referred to as validation. Truth is usually fects appear. On general principles, this appeal- relative, and in the case of the carcinogenic po- ingly simple system may still hold promise for set- tential of chemicals, the convenient measuring ting and defending regulatory goals, but its stick for truth is the bioassay, with its attendant systematic failure to guide regulatory efforts in limitations (542). The acceptability of bioassay specific instances has led to disillusionment. results as a guide to making decisions about health hazards appears, sometimes at least, to be tied to The technical problems encountered in conduct- the financial interest or disinterest of individuals ing bioassays--including questions about high and organizations in the substances identified as doses, and the impossibility of knowing which ex- carcinogens. trapolation model is most appropriate—plague risk assessment. Equally or more important are There is little hope that a single short-term test the assumptions involved. For instance, appar- will ever suffice as a reliable predictor of toxicity ently endless arguments have gone on about in human beings, and hope is pinned on the de- whether liver tumors in mice mean anything in velopment of a battery of tests. Years of discus- terms of human risk; the argument has not been sion and argument will undoubtedly precede the settled by experimentation but is silenced by con- acceptance by scientists and regulators of any set vention (542). of tests. And even then, a “generally accepted” test battery will be challenged in specifics, much Formaldehyde is a case in point. There is gen- as evidence from bioassays currently is. eral agreement that formaldehyde is an animal The development of reliable short-term tests carcinogen. The bioassay was carried out by in- may actually enhance the value of bioassays, dustry’s own toxicology laboratory. But in the which will always find a place in toxicologic final analysis, industry objected to regulating for- testing. Short-term tests can increase the knowl- maldehyde on the basis of the bioassay, and as- edge base for deciding which chemicals should be sessments produced by different organizations tested in animals, and can shed light on the prob- varied in the amounts of human risk they pre- able mechanisms of action of each chemical. dicted. 58 ● Preventing Illness and Injury in the Workplace

Epidemiology There are standard methods for locating people in this mobile society. Asking at places of em- The importance accorded epidemiology reflects ployment and using telephone and city directories a trend toward more systematic, scientific study are common. Mail sent to the last known address of disease. The desire to base conclusions about frequently reaches the person. In difficult cases, causality on something more than individual ob- the epidemiologists can hire private detectives or servation and intuition-the two most valuable credit bureaus to locate persons. The so-called tools of the clinician—calls for describing associations quantitatively, both in terms of strength NIOSH-window facilitates some occupational epi- of association and in terms of the probability that demiology studies. Investigators who are allowed to use it can supply a name and some other iden- the association is not simply one of chance. Care- tifying information (such as the Social Security ful epidemiologic investigations have confirmed number) to the Internal Revenue Service, and the important suspicions about work-related illnesses. The now universally acknowledged case against agency provides the person’s current address. Members of the OTA Advisory Panel for this as- asbestos is built on epidemiologic studies. sessment reported that there is some confusion The strengths of epidemiology still remain to about who can and cannot use the NIOSH win- be exploited. A great deal needs to be learned dow and under what conditions. about diseases and syndromes that are widespread in the population. Certain chronic conditions The Federal Government collects information (cancers in particular) and heart disease are known about places of employment and about what to be associated with various occupations. The hazards or substances are present in them. Such means exists, through the Surveillance, Epidemi- records have obvious usefulness for epidemiology, ology, and End Results Program of the National providing a quick method for identifying persons Cancer Institute (542,683), to enter about one- who may have been exposed to a substance. How- tenth of all U.S. cancer cases on tumor registries ever, all the record systems have flaws that re- as they are diagnosed. This system provides the strict their usefulness (542,557). The recommen- ability to set up large case-control studies with dations made by the Committee to Coordinate relative ease. (See box B.) Environmental and Related Programs (CCERP) of the Department of Health and Human Serv- Cohort studies of large industrial populations— ices provide an excellent grounding for questions which can be assembled by corporations and/or about the current systems and suggestions for unions and facilitated by workplace surveillance changes. systems that have been installed by many companies to track and store various sorts of data— The National Death Index (NDI) can tell epi- also yield valuable information. (These surveil- demiologists that a person is dead and which State lance systems are discussed further in the “Oc- (or other) department of vital statistics holds the cupational Medicine” section. ) death certificate. This speeds up the retrieval of information for studies, but the NDI does not ac- Government Records. —An important and frus- tually provide information on the cause of death trating feature of epidemiology in the United and underlying causes. States is the difficulty of locating and tracking people. In a cohort study, it is critical that the Section 8(e) of the Toxic Substances Control maximum number of cohort members be located. Act (TSCA) requires that manufacturers report If the cohort contains workers employed at a par- to EPA on chemical substances that pose signifi- ticular site S, 10, or 20 years ago, many will have cant risks to human health or the environment. moved. In a case-control study, members of ei- Some companies voluntarily report these results ther population may be identified through hos- to NIOSH and OSHA. In practice, this reporting pital records, and the recorded addresses may no requirement means that an employer that carries longer be current. In either type of study, the out a short-term test, a bioassay, or an epidemi- epidemiologist often needs to locate people for in- ologic study that shows a health risk must report terview and examination. it to the EPA. EPA prepares a report on each 8(e) Ch. 3—Health Hazard Identification ● 59

Case-Control Studies . In a case-control study, persons with the disease under study (cases) are compared with individuals without the disease (controls) with respect to risk factors that are judged relevant. Some authors label this study design “retrospective” because the presence or absence of the predisposing risk factor is determined for a time in the past. However, in some cases the presence of the factor and the disease are ascertained simultaneously. The choice of appropriate controls is rarely without problems. Often, for practical reasons, controls are chosen from hospital records. But they may not be representative of the general population, and they therefore may introduct “selection bias” (2$2).

General Considerations In case-control and cohort studies, the groups selected should be comparable in all characteristics except the factor under investigation. In case-control studies, the groups should resemble each other except for the presence of the disease; in cohort studies, the study and comparison groups should be similar except for exposure to the suspect factor. Since this rarely is possible in practice, comparability between groups can be improved by either matching individual cases and controls (in case-control studies) or by standard statistical adjustment procedures (in either case-control or cohort studies). Demographic variables such as age, sex, race, or socioeconomic status are most commonly used for adjustment or matching. There are advantages and disadvantages in both types of study (see table 3-8). Case-control studies tend to be less expensive to conduct, require relatively fewer individuals, and often have been especially

Table 3-8.-Advantages and Disadvantages of Case-Control and Cohort Studies

Type of study Advantages Disadvantages Case-control ...... Relatively inexpensive Complete information about past exposures often unavailable Smaller number of subjects Biased recall Relatively quick results Problems of selecting control group and matching variables Suitable for rare diseaes Only relative risk is yielded Cohort ...... Lack of bias in ascertainment of risk Possible bias in ascertainment of disease factor status incidence rates as well as relative risk are Large numbers of subjects required yieided Associations with other diseases as by- Long follow-up period product can be discovered Problem of attrition Changes over time in criteria and methods Very costly Difficulties in assigning people to correct cohort SOURCE: (542). 60 ● Preventing Illness and injury in the Workplace

notification and circulates it within the Agency ical schools. Levy (269a) reports that only 50 and to other Federal agencies, including OSHA percent of U.S. medical schools provided some and NIOSH. In addition, periodically the reports class time to occupational health during the 1977- received over a period of time are bound together 78 academic year. This has risen to 66 percent in for distribution to libraries. The 8(e) activities, the 1982-83 academic year. However, the median therefore, provide a way to disseminate health number of required class hours devoted to occupa- hazard information rapidly. tional health remained at 4 hours. Postgraduate, specialty training in this country has traditionally been subsumed under preventive medicine, and Occupational Medicine centered in schools of public health. Recently in- The field of occupational medicine has gone creased emphasis has been placed on clinical ex- through a series of changes during this century. perience in medical schools. The location of the Not long ago, the clinician not only tended the specialty courses is less important than making sick but also filled a number of other roles, insure the programs are well-taught and attractive vestigating possible disease relationships and and that they provide clinical experience. The fostering changes in the workplace. To a certain NIOSH-supported Educational Resource Centers extent, the role of the occupational physician was (discussed in ch. 10) provide postgraduate edu- altered by the rise of epidemiology and toxicology cation for physicians. as separate professions. In the United States, the occupational medical Epidemiology and toxicology have not been the services are usually provided by physicians who panacea for solving workplace health problems are directly employed by or under contract to that some envisioned. Toxicology is limited to employers. Large companies frequently have on- testing under conditions that cannot mimic com- site medical departments, staffed by physicians plex human exposures and behaviors. Epidemiol- and nurses. Medium-sized companies might have ogy cannot begin until it finds subjects for study, the full-time services of an occupational health and it relies on outside input—in particular, clin- nurse, and possibly, the part-time services of a ical observations of possible associations between local physician. Small companies have only rarely exposures or behaviors and disease—for hypoth- provided occupational medical services. esis generation. It has limitations in the kinds and An alternative organizational model is found magnitudes of effects it can detect. The limitations in occupational medicine clinics, which have been of both toxicology and epidemiology argue for growing in the last few years. These clinics are a continuing role for occupational medicine in usually associated with a hospital or university hazard recognition as well as in treating workers. and provide examinations and treatment to work- Better use of physicians’ experience and insights ers. Clinics might, because of a larger patient load will depend on education. There are two catego- and a staff that consequently sees more patients, ries: general education of physicians about oc- be able to to provide more knowledgeable care, cupational disease and injury, and specialized as well as improved physician training. In some education and training for practitioners of occu- cases, the clinics’ staffs include not only doctors pational medicine. An orientation toward occupa- and nurses, but also industrial hygienists and tional health is minimal at best, in most U.S. med- safety engineers. The combination of staff from Ch. 3—Healfh Hazard Identification ● 61 different disciplines can provide a critical mass for In the occupational setting, the necessary com- a great deal of important activity in hazard iden- ponents of surveillance are: tification and control. ● exposure information of some type; These clinics also provide advantages to em- ● records of health outcomes, which may in- ployers, especially small to medium-sized com- clude causes of death; and panies, that previously were not able to provide ● background information about characteris- occupational medical services to their workers. tics of each individual that might influence In the words of the director of an occupational susceptibility to disease. health department at one hospital: Variations in epidemiologic surveillance sys- The larger corporations will undoubtedly con- tems have to do mainly with the quantity and type tinue to have in-plant occupational health serv- of data in each category. “Exposure” can be quite ices. But medium and smaller companies will be basic: for instance, knowing the plant within a forced to make an economic decision on whether company, or the department within a plant, in it is more advantageous to do it themselves or which a worker is employed, and updating it per- farm the occupational health service to others haps yearly. At the more comprehensive end of (Daniel Conrad, quoted in 338a). the spectrum, exposure might contain continuous Some hospitals are apparently establishing these records of personal and area monitors measuring clinics in order to develop new sources of reve- chemicals and other agents in the industrial envi- nue. The staff of these clinics expect to be able ronment. to conduct some research, as well as to provide Health outcomes may be ascertained from in- advice about prevention and medical care to em- dustrial health and accident insurance reports, ployees (338a). which record only the most serious events, These Medical Surveillance Systems. —Computerized can be supplemented by information gathered in information systems have made it possible to store preemployment examinations and nonroutine massive amounts of data. Information about ex- visits to physicians, as an intermediate approach. posures in the workplace and the health records At the extreme, to the above information could of workers can form the basis for surveillance sys- be added the results of periodic medical screen- tems that aim to identify health hazards. Surveil- ing for many diseases or other abnormalities. Ba- lance is defined as the “collection, collation, and sically, the simpler systems are considered passive, analysis of data and its dissemination to those using data collected for other purposes (person- who need to know” (474). Public health surveil- nel records, insurance data); systems can be pro- lance techniques were developed in the last cen- gressively more active in seeking data expressly tury to identify foci of pestilential diseases such for health surveillance (312). as cholera, smallpox, plague, and yellow fever, Routine analyses of data collected in surveil- so that appropriate control measures could be in- lance systems are seldom sufficiently rigorous to stituted. In the workplace, the value of surveil- evaluate possible instances of occupational dis- lance is to alert workers and employers to unusual ease. Their broad, sweeping monitoring of health patterns of morbidity or mortality. events is more of a hypothesis-generating device. It provides the means to make epidemiologic Concerns today center on chronic rather than studies as targeted and as timely as possible. acute diseases; the technical problems of linking cause and effect are heightened by the remoteness A sign of growing interest and activity in oc- of disease from exposure. Computerized informa- cupational health surveillance, and medical infor- tion systems in industry, including their use for mation systems in general, was a meeting of the medical and exposure records, have enabled mas- American Occupational Medical Association’s sive amounts of information to be stored and cor- Medical Information Systems Committee in 1981. relations to be produced. Papers presented at that meeting, which described 62 . Preventing Illness and Injury in the Workplace

19 such systems, were published as a supplement requires reporting of such findings, making them in the October 1982 issue of the Journal of Oc- public and available to potential litigants. On the cupational Medicine (238). other hand, some companies expect that acting responsibly will provide some defense against tort In the same issue of that journal, computer soft- action. The problem of deciding how to use suspi- ware companies advertised their ready-made pro- cions that may be generated by routine match- grams for instituting surveillance systems. The ing of health and surveillance information is a literature packets behind those systems, which ap- very real one. peared to be directed at smaller companies, describe convenient ways to classify and store large A second policy issue in this field concerns the amounts of information about workplace expo- proliferation of data collection systems for health sures and employee health, What is missing, at and exposure information that are accompanying least in the prospecti, are discussions about the the microcomputer age. There appears to be little ultimate value and potential contribution of such thought given to the ultimate value of these sys- information to detecting problems in the work- tems in improving workplace safety and health. place. Although the systems may facilitate rec- Certainly for small companies, the targets of much ord keeping that already goes on, they may fail advertising, the number of workers will be too to have a serious impact on safety and health, as small ever to detect all but the most obvious ex- they are promoted to do. cesses of disease. There may be scope for using computer networks to pool data, but these activ- Occupational health surveillance remains a ities bring their own problems. (See ch. 10. ) source of both great promise and great controversy. If it could be used just to identify the Another pertinent issue is the substitution of causes of occupational illness, setting the stage for surveillance for prevention, particularly preven- preventing further illness, there would be little to tion in the form of controls on workplace ex- say against the idea. As a purely scientific con- posures. Union officials and many health profes- cept, it is unassailable. In practice, from the point sionals fear that the creation of surveillance of view of companies, the collection and particu- systems will lead to the impression that “some- larly the analysis of data about exposures and thing is being done” to improve health, resulting health outcomes raises legal issues of responsibility in less emphasis on controls and paralyzing ac- and liability. From the employees’ point of view, tion against hazards until large numbers of peo- there is a fear that surveillance will be adopted ple become sick or die. as an alternative to installation of controls. There is anecdotal evidence that some com- New Hazards panies that had maintained surveillance systems have now dismantled them. Although the same In some measure, “familiarity breeds contempt”- data may still be collected for administrative rea- even when the subject is hazards-and there may sons, they are not being assembled in a form for be a human tendency to fear new hazards more analysis of possible relationships between expo- than old ones. The emphasis placed on identify- sures and disease. This step may at least in part ing and understanding “new” hazards grows stem from the unknown consequences of finding partly from that psychology and partly from the the suggestion of a health problem—for instance, realization that it is easier to control hazards a slight excess of some particular cancer. Further before they become established in commerce and study would certainly be necessary to confirm the economically important. association, yet the liability associated with even Epidemiologic studies and occupational medi- suspecting that a problem exists cannot be known cine are of no value in learning whether a new at this time. agent is hazardous before people are exposed to Some employers are concerned that discover- it. The introduction of a new substance or proc- ing a possible association may make them liable ess into the workplace that is subsequently shown in tort actions. In addition, section 8(e) of TSCA to be a hazard must be regarded as a failure of . .

Ch. 3—Health Hazard Identification ● 63 — —- preventive health measures. Analysis of the chem- tional exposures to chemicals described on PMNs ical structure of a new substance can be used to (547a), and it has established informal communi- estimate what toxic properties are associated with cations with OSHA and NIOSH staff about con- it, but many people consider that technique to be trols. For instance, EPA has required the use of unreliable. Toxicologic techniques can be used to respirators in the manufacture of some new chem- learn about the hazards of new substances, but icals described on PMNs. According to EPA offi- the associated costs place some restrictions on cials, the Agency consulted with NIOSH about their use. appropriate respirators. Toxicology costs money, and manufacturers The PMN program provides an important op- will not spend great sums on testing a newly de- portunity to identify hazards before they become veloped chemical before they know there is a mar- established in the workplace. Although EPA reg- ket for it. Some manufacturers argued during the ulated pesticides under a licensing law before debate when the Toxic Substances Control Act TSCA, its regulatory concern about other chem- was passed that they did enough toxicologic icals was restricted to those that became pollut- testing to be assured that new chemicals would ants. Under the PMN program, it has authority not pose unreasonable risks. TSCA set up two to regulate chemical substances before they get programs to gather information about new into the workplace. chemicals. Significant New Uses The Premanufacture Notification Program TSCA anticipated that the uses of a chemical The Toxic Substances Control Act requires that described on the PMN might not be associated manufacturers prepare a Premanufacture Notice with an unreasonable risk, but that a different use, (PMN) and submit it to EPA at least 90 days called a “significant new use, ” might. TSCA before starting manufacture of a chemical sub- directs EPA to write a significant new use order stance for use in commerce. The PMN is to con- about new chemicals that fall into this category. tain any information available to the manufac- In practice, EPA has restricted some chemicals to turer about the toxicity of the chemical, Some particular uses and required submission of more PMNs contain many items of information bear- data about the chemical before it could be more ing on the properties of the new chemical, while widely used. One example of this process concerns others contain none or only a few, and there are a surfactant for cleaning. Concerned about pos- disputes about how useful the reporting has been sible dermatologic effects, EPA did not object to to date (547a). its use by professional cleaners, because those workers could be instructed in the proper use. It is clear from EPA’s experience with the PMN However, if the surfactant is considered for use program that a common plain for potentially haz- in consumer products-a significant new use— ardous exposures to newly introduced substances more information must be provided to EPA. is in their manufacture. EPA has used formal and informal regulatory procedures to reduce occupa-

SUMMARY

Preventing workplace-related disease requires vapors from manufactured and naturally occur- that associations between activities and exposures ring substances—illustrate the diversity of ex- and diseases be identified. The known health posures. In addition to identified hazards, present hazards-extremes of heat and cold, radiation of but so-far-unidentified hazards are also a concern. various kinds, noise, and some dusts, fumes, and Finally, increasing attention is being focused on 64 ● Preventing Illness and Injury in the Workplace —. assessing the possible hazards of new substances and locations, are especially useful in epidemi- and processes before they are introduced into the ologic studies. workplace. Toxicology provides information about the po- Some health hazards that have been known for tential hazards of substances by testing them in centuries were obvious because of the particular animals or other systems. With the passage of the nature of the diseases; for instance, lead poison- Toxic Substances Control Act, which requires that ing symptoms were sufficiently distinctive to make companies notify the Environmental Protection the association between exposure to lead and dis- Agency of their intention to manufacture new ease apparent. Three disciplines—occupational chemicals, the government is in a position to ob- medicine, epidemiology, and toxicology—have tain information about chemicals before they enter been important in describing associations. All commerce. Although there are conflicts about three are currently used in investigations of cur- how much information EPA needs to protect rent exposures that may be hazardous. Toxicology human health, it is clear that workplace exposures is especially important to learning about “new, ” are being identified as concerns in the case of some possibly hazardous substances before they are new chemicals. Toxicology plays the central role introduced into the workplace. in identifying hazards from new chemicals. Some of the most successful efforts at preven- TSCA also requires that companies notify EPA tion, such as the marked reductions in exposure about substances present in commerce that are to vinyl chloride, began with a physician noting substantial risks, and the Agency then dissemi- an unusual cluster of diseases. The importance of nates that information. All three disciplines— this source of information draws attention to med- occupational medicine, epidemiology, and toxi- ical school teaching about the role of work in cology—have contributed to the identification of health and disease. Unless medical students learn substantial risks. The NIOSH Health Hazard Eval- the value of taking an occupational history as part uation program investigates possible associations of the medical examination, associations may be between exposures and illness at the request of missed. Occupational physicians, familiar with employers or employees or on its own initiative working conditions and exposures and often inter- (see ch. 10). It, too, relies on all three disciplines. acting with industrial hygienists and safety engi- Hazard identification is not a smooth path; neers, can be especially important in hazard iden- arguments and conflicts abound. Evidence that tification. Workers’ own observations and convinces some people leaves others unmoved. complaints, brought to the physician, are often The methods that were used in the past, improved the first indication of a hazard. by better training and techniques, continue to be Epidemiology is important less in initial iden- of value today. More attention during the edu- tification of hazards than in providing evidence cation of physicians and other medical personnel for or against an association. In making decisions to the influence of work on health, better use of about which hazards are “real, ” positive Federal records, where appropriate, to facilitate epidemiologic studies are the most convincing evi- epidemiology, and continual research to make dence, but there are often protracted arguments toxicology more predictive all offer opportunities about the appropriateness of study methods and to improve hazard identification. However, as is the conclusions drawn. Companies, trade asso- made clear in other parts of this assessment, haz- ciations, unions, and government agencies all ard identification alone is not sufficient. Making commission epidemiologic studies and comment a decision to control a hazard requires that the on studies done by others. Government records, hazard be identified, but identification, by itself, which contain information about vital statistics is not sufficient for control. 4.

Safety Hazard Identification LIST OF TABLES Table No.

. .

,. 4 Safety Hazard Identification;

Many safety hazards are obvious: a punch press An epidemiological approach analyzes sets of sim- ram that descends every five seconds, a wet floor, ilar injury-related incidents to find common cir- an unstable ladder. This is true at least for the cumstances contributing to their cause. At the causes of acute injuries, though it is not neces- most specific level, individual injury-related in- sarily true for cumulative injuries. In either case, cidents are examined to determine cause; several what is often unclear is the complex of events methods have been developed at this level. through which the potential of the hazard is realized and an injury occurs. During this century the The nearly 4,100 work-related fatalities (includ- theory of the cause of injuries has evolved, and ing heart attacks) that occurred in private sector is still evolving, from attributing most injuries to workplaces with 11 or more employees during “unsafe acts” of workers, to identifying conditions 1982 were not evenly distributed over all indus- that increase the probability of an injury occur- tries (see table 4-1). Mining, accounting for 2 per- ring. Under the first approach, the preventive cent of employment, had 11 percent of the fatali- remedy is to install perfect workers in jobs. Under ties, Construction, 5 percent of employment, the second—by identifying all possible contrib- accounted for 18 percent of reported on-the-job uting factors—preventive strategies can be ap- deaths. The wholesale and retail trades represent plied, or at least considered, in different ways. 25 percent of employment, but recorded only 12 percent of the fatalities. Finance, insurance, and The causes of work-related injuries can be ex- real estate, along with the service industries, ac- amined at two levels. The “macro” approach uses counted for 31 percent of employment, but only aggregate statistics, such as those produced in the 12 percent of fatalities. Bureau of Labor Statistics’ Annual Survey, to examine the distribution of various types of injuries A further breakdown reveals the injuries that according to several variables: industry, occupa- resulted in death by industry categories, both by tion, size of establishment, sex of worker, and the distribution of causes within an industry cat- others (see ch. 2, Working Paper #l). These dis- egory (see table 4-2), and by the distribution of tributions provide general clues to injury causes. each cause across all categories (see table 4-3). The “’micro” level identifies specific injury causes. Overall, motor vehicle accidents account for 27

Table 4.1.–On-the-Job Fatalities by Industry Division, in Private-Sector Units with 11 Employees or More, 1982

Annual average employment Fatalities Industry division Number Percent Number Percent Private sector (total) ...... 62,629,000 100 4,090 100 Agriculture, forestry, and fishing ...... 729,000 1 180 4 Mining ...... 1,070,000 2 440 11 Construction ...... 2,898,000 5 720 18 Manufacturing...... 18,267,000 29 770 19 Transportation and public utilities...... 4,629,000 7 970 24 Wholesale and retail trade ...... 15,603,000 25 490 12 Finance, insurance, and real estate ...... 4,252,000 7 100 2 Services ...... 15,181,000 24 420 10 NOTE Because of rounding, components may not add to totals SOURCE (608)

68 . Preventing Illness and Injury in the Workplace

Table 4-2.—Causes of On-the-Job Fatalities in Private-Sector Units with 11 Employees or More, by Industry Division, with Distribution by Industry, 1981 and 1982a

I I I

Causeb 1 Total—all causes ...... ,-.., ...... - 100 100 100 100 100 100 Over-the-road motor vehicles ...... , . . . 27 18 26 15 20 52 20 35 29 Falls ...... 12 12 9 31 10 6 5 9 10 Heart attacks ...... 10 6 8 8 10 6 12 23 16 Industrial vehicles or equipment ...... 10 27 21 17 9 3 4 0 9 Nonaccidental injuries ...... 7 3 <1 <1 2 2 30 8 15 Struck by objects other than vehicles or equipment ...... , . . . 6 1 9 5 8 0 2 Electrocutions ...... , . . 6 16 4 11 5 0 5 Caught in, under, or between objects other than vehicles or equipment...... 6 1 3 4 17 2 Aircraft crashes...... 4 2 5 1 77 Fires ., ...... 3 8 7 1 1 <1 Plant machinery operation ...... 3 1 1 2 0 <1 Explosions ...... 2 0 2 2 0 1 Gas inhalations ...... 2 1 3 1 0 1 Another ...... 3 4 3 3 <1 3 the industry division level sampling errors are large Therefore, the results are for both Years rather than a comparison between them. as the object or event associated with the fatality. and nonmetal mining, and railroads for data are available ‘Excludes NOTE: Because of rounding, percentages may not 100 SOURCE (608)

Table 4-3.—Causes of On-the-Job Fatalities in Private-Sector Units with 11 Employees or More, by industry Division, with Distribution by Cause, 1981 and 1982” -

Over-the-road motor vehicles ...... 100 3 6 10 18 37 11 4 11 Falls ...... ,, . .,,..,.,, 100 4 4 47 20 8 62 8 Heart attacks. ..., . . ..., ..., ...... , 100 3 516 24 11 18 7 16 Industrial vehicles or equipment ...... 100 12 14 32 23 5 6 0 9 Nonaccidental injuries ...... , ..., ..., 100 2 <1 <1 6 5 63 3 20 Struck by objects other than vehicles or equipment .,,,...... 100 <1 10 16 33 8 31 0 3 Electrocutions ...... , ...... 100 12 4 34 21 18 3 0 8 Caught in, under, or between objects other than vehicles or equipment...... 100 1 4 12 21 30 20 9 3 Aircraft crashes...... ,. 100 2 9 6 22 32 6 5 19 Fires .. ...,,...... 100 12 15 8 46 14 1 2 2 Plant machinery operations ..., . . . . ,... 100 1 2 11 78 2 7 0 <1 Explosions . . ,...... , ...... 100 0 5 14 47 20 8 0 6 Gas inhalations ...... , ...... 100 2 9 10 48 14 12 0 5 All other ., ...... 100 6 5 18 33 13 16 1 8 is changes precisely the division level sampling errors are large Therefore, the are for both Years rather than a comparison between them Is defined as the object or event associated the and nonmetal mining, and railroads for which data are available ‘Excludes railroads NOTE rounding, percentages may not 100. SOURCE (608) Ch. 4—Safety Hazard /identification 69 — ..—— ————- — percent of fatalities, but in transportation and deaths in the work force (46 percent) and deaths public utilities the figure is 52 percent. In construc- from plant machinery (78 percent) occur in man- tion, falls were responsible for a greater propor- ufacturing. Additional sources of aggregate injury tion of deaths (31 percent) than were over-the- statistics are described in Working Paper #l. road motor vehicles (15 percent), although indus- Aggregate statistics can guide injury prevention trial vehicles and equipment were associated with by highlighting general hazard categories in spe- 17 percent of deaths. cific industries. Immediate or underlying condi- Nearly half of all fatalities (47 percent) resulting tions related to an individual injury can only be from falls occurred in the construction industry determined through case study. (see table 4-3). Most fire- and explosion-related

BASIC THEORIES OF INJURY CAUSATION

Traditional Approach—Unsafe only “’unsafe acts. ” Thus he produced a well- Conditions or Unsafe Acts known and often cited figure that 88 percent of injuries are due to “unsafe acts” by workers (380). In the 1920s, Heinrich proposed a theory of in- Heinrich attributed only 10 percent of injuries to jury causation that many safety professionals have unsafe conditions and considered the remaining followed ever since. Simplified, the theory states 2 percent of injuries to be unpreventable (207). a domino sequence: • Injuries are caused by accidents. • Accidents are caused by unsafe acts of per- Critique of Traditional Approach sons or by exposure to unsafe mechanical conditions. Arndt (24) has noted that, although there has • Unsafe acts and conditions are caused by been little research published to support Heinrich’s faults of persons. theory of injury causation, Heinrich’s ratio of 88 ● Faults of persons are created by the environ- percent unsafe acts to 10 percent unsafe condi- ment or acquired through inheritance. tions is commonly cited. In fact, the published research on this topic uniformly refutes Heinrich’s Using this approach, Heinrich analyzed 12,000 theory. cases of injury from insurance claim records plus 63,000 cases from the records of plant owners, Heinrich himself pointed out two other studies. for a total of 75,000 cases. Seventy-three percent The first, by the National Safety Council (NSC), of the injuries were classified as due to “unsafe concluded that unsafe acts contributed to 87 per- acts” by workers. Heinrich noted that 25 percent cent of the cases examined, while mechanical of the cases examined would, according to the causes contributed to 78 percent. The total of 165 usual methods employed at the time, have been percent is due to NSC’s considering multiple charged to defective or dangerous physical or me- causes of accidents. An analysis in 1940 by the chanical conditions. However, he concluded that State of Pennsylvania showed that an “equal num- many cases in this group of 25 percent were caused ber” of injuries resulted from unsafe acts and me- either wholly or chiefly by worker failure, and chanical causes. Heinrich recognizes the discrep- only partly by physical or mechanical conditions. ancy, which he attributed largely to the fact that He decided to classify as “unsafe conditions” only the NSC and Pennsylvania studies allowed both those cases that were wholly caused by physical an unsafe act and an unsafe condition to contrib- or mechanical failure. The injuries not wholly due ute to a single injury. Heinrich’s methodology did to physical or mechanical failure (15 percent) were not permit such multiple assignment of cause grouped with the 73 percent of cases that involved (207), 70 ● Preventing Illness and Injury in the Workplace

Table 4-4 presents a summary of other research presses allow operators to insert their hands in- aimed at apportioning injury causes between un- side the “point of operation” of the press. It is not safe acts and unsafe conditions. Heinrich’s study surprising, then, that someday someone places a is the only one to attribute more than 35 percent hand or arm inside such a press to adjust the piece of injuries primarily to unsafe acts by workers. being worked on or to clear a jam, and then is The other research has generally categorized most unable to remove it quickly enough. For exam- injuries as resulting from a combination of un- ple, if a press operator produces 5,000 pieces a safe acts and unsafe conditions. day, then the operator’s hands are placed in front of the press ram every 5 seconds, which means Arndt (24) examined nearly 1,000 injuries asso- about 25,000 times per week or 1.2 million times ciated with mechanical punch presses. He devel- a year (24). It may be only a matter of time before oped eight mutually exclusive categories to dean operator commits an “error” and loses a fin- scribe the circumstances of the injury. These ger in the press. included operator timing errors, inadvertent tripping of the press, other operator errors, tripping There are, however, machine designs that can of the press by a second person, and machine reduce and nearly eliminate this particular haz- malfunctions. He found that 53 percent of the in- ard. Machines can be designed to operate with juries resulted from something other than machine two-handed controls, so that the operator must malfunctions. All of that 53 percent would be at- have both hands on the controls. In Arndt’s anal- tributed to “unsafe acts” in a dichotomous sys- ysis, 60 to 70 percent of the injuries from presses tem, like Heinrich’s, for recording causes of ac- activated by foot pedals, by one-handed controls, cidents. The machine malfunctions, including or automatically were related to “unsafe acts, ” and broken parts and accidental recycling of the press, only 10 to 20 percent related to “unsafe condi- which would generally be labeled “unsafe conditions. ” For presses with two-handed controls, the tions, ” amounted to 18 percent of cases. Arndt fraction due to “unsafe acts” was only 35 percent. was unable to classify 29 percent of the cases be- “Unsafe conditions” were cited in about 54 per- cause of a lack of information. cent of these cases. Arndt’s paper does not present any information on the injury rates associ- Thus, under the traditional breakdown between ated with the various kinds of presses because data unsafe acts and unsafe conditions, about three on the total numbers of each control type are not times as many injuries in Arndt’s study would be available. But it is clear that the design of the press classified as due to unsafe acts rather than unsafe has a dramatic affect on the number and percent- conditions. But Arndt observes that a very large age of cases attributed to “unsafe acts. ” number of those classified as “unsafe acts” occurred on presses activated by a foot pedal, by The traditional partition between unsafe acts a one-hand control, or automatically. These and unsafe conditions unfortunately often draws

Table 4-4.—Estimated Percentages of Accidents Due to Unsafe Acts versus Unsafe Conditions . .—— —. Percent due to Percent due to Percent due to ‘Percent - Study unsafe acts unsafe conditions combination unknown Furniss . . .:...... ~...... — – 16 84- — Pennsylvania Department of Labor...... 2 3 95 — National Safety Council...... 19 18 63 — Mintz and Blum ...... 21 — 79 — Hagglund a ...... 26 58 2 13 Hagglund b ...... 35 54 4 7 c 10 — — Henrich ...... 88 —— — — alnvestigations of reported fatalities. bRandom sample of accident reports cHenrlc h classified 2 percent as ‘unpreventable ‘‘ SOURCES (30,37,207,316) Ch. 4—Safety Hazard Identification 71 attention away from the job or equipment rede- point of impracticability. The procedure, there- signs that can remove or minimize hazards. In the fore, provides for recording of one pertinent fact 1959 edition of his textbook, Heinrich himself cau- about each accident in each of the specific cate- tioned safety professionals not to neglect work- gories or classifications. To insure uniformity in place conditions. He expressed confidence that the selection of items to be recorded in each category, the items are specifically defined in terms safety professionals would not “ignore the very which eliminate any necessity for decision as to first common-sense step. . . of safeguarding [the] the relative importance of multiple items falling mechanical environment” (quoted in 462). in the same category (emphasis added). The catchall category of “unsafe act” or “human Instead of collecting information on all the cir- error” has greatly restricted advances in injury re- cumstances leading to the accident, the ANSI search and the application of control techniques Standard Z16 allows “only one pertinent fact” to in workplaces (380). The label “unsafe act” has, be recorded concerning the nature of the injury, unfortunately, often led to a failure to recognize the source of the injury, the type of accident, the how the design of workplace equipment can min- hazardous condition present, and any unsafe act. imize the occurence of “unsafe acts” or reduce the This standard facilitates the administration of in- probability and severity of human injury. jury data collection because of its simplicity, but The seriousness of this limitation is clear from it is inadequate for research on causation. Unfor- one commonly used system for recording infor- tunately, the most common entry under this sys- mation about injuries. The American National tem is simply to attribute the injury to “worker Standards Institute (ANSI) Standard Z16.2, error. ” Method of Recording Basic Facts Relating to the According to Purswell and Stephens (380), at- Nature and Occurrence of Work Injuries, used tributing responsibility for accidents to human er- widely for employer injury investigation and rec- ror, with no significant information as to why the ordkeeping. ANSI itself sees inadequacies in its error was committed, is not limited to the ANSI method, as the text of Z16.2 indicates: system. It is found in other workplace-injury data It is recognized that the occurrence of an injury collection systems as well as those for collecting frequently is the culmination of a sequence of data on non-workplace injuries. For example, re- related events, and that a variety of conditions searchers in the field of highway safety have noted or circumstances may contribute to the occurrence that there is no place on standard police forms of a single accident. A record of all these items to record many items that relate to features of the unquestionably would be useful to the accident vehicle or the road that contributed to the injury. preventionist. For the most part, these forms are oriented around Any attempt to include all subsidiary or related recording information on the driver (41). facts about each accident in the statistical record, however, would complicate the procedure to the

OTHER MODELS OF INJURY CAUSATION

Purswell and Stephens (380) describe a num- Some safety specialists believe that a dispropor- ber of other models of injury causation and in- tionate number of injuries are incurred by a hand- vestigation. These include behavioral models, ful of individuals who are especially prone to ac- management models, epidemiological models, and cidents. Accident proneness has been, at one time ergonomic or human factors models. or another, ascribed to recent immigrants to the United States, to certain ethnic/racial groups, or Behavioral Models to certain personality traits (380). Thus efforts were made to identify these workers and either The underlying concept of behavioral models fire them or not hire them in the first place. Later is that of the “accident proneness” of individuals. researchers have been unable to find similar traits 72 . Preventing Illness and Injury in the Workplace that will reliably predict which workers will be energy should loss of control or improper injured. Unfortunately, the belief that there are flow occur. “injury-prone” workers is still commonly held. The next major outgrowth of the domino the- According to a Bureau of National Affairs (78) ory was based on the idea that multiple factors report, 65 percent of the businesses surveyed can combine in a random manner to produce ac- stated that their safety programs attempted to cidents and injuries. Such causation models focus identify “accident-prone” individuals. not only on unsafe acts of the injured person, but Other behavioral models have considered mo- also on unsafe acts of coworkers and unsafe con- tivational factors, the rewards of working safely, ditions that existed at the time. Attention is and the level of satisfaction received from work- ultimately drawn to failures in management sys- ing safely (368). It has been observed that many tems that permit the multiple factors to converge workers perceive little positive reward for and produce an injury. working safely. One of the best known management-oriented approaches to accident causation is called the Management Models Management Oversight and Risk Tree (MORT), Bird (58) revised Heinrich’s domino theory to developed by Johnson (235) for use in the analy- emphasize management’s responsibility for injury sis of complex systems related to atomic energy. causation. His revised domino theory is: It could also be called a systems model (discussed later in this section). MORT employs a large sche- ● Injuries are caused by accidents. matic to inductively trace events of a work-related ● For each accident there are immediate causes injury back in time, to identify the sequence of that are symptomatic of problems in the unwanted energy flow, and to evaluate the ade- overall system. quacy of barriers to unwanted energy transfer to ● There are basic causes in the overall manage- persons or equipment. Along the route, hazards ment of the system that produce the imme- arising from specific accident circumstances, from diate causes of the accident. risks acknowledged or assumed by management, ● The lack of management control permits the and from general management systems and pol- basic causes of accidents to exist in the icy weaknesses are identified. system. Still, several difficulties exist in adopting MORT Bird’s approach therefore shifted the emphasis to general injury investigation or applying it to from the worker as the cause of injuries to the most industrial workplaces. Its use as an indus- management system in which the worker exists. trial injury investigation procedure is limited by Zabetakis (684) of the Mine Safety and Health its complexity. The method is more suitable for Administration Academy added the idea that in- investigating large-scale incidents, especially sit- juries are due to an unplanned release or flow of uations holding the potential for public disaster, energy, again following the approach of the domi- such as nuclear powerplants. While it is an excel- no theory. Energy release is considered in the lent approach for these situations, in its present general sense—mechanical, electrical, chemical, form it is much less useful for explaining most thermal, or ionizing radiation. Since unwanted work-related injuries. But it may be useful as a energy flow is a fundamental source of injuries, blueprint for the optimal allocation of resources Zabetakis claimed, a system maybe evaluated and for building a safety program (380). improved by studying: Epidemiologic Models the sources of energy existing in a system, the means available to reduce the energy Epidemiology has been described as the search levels, for causal association between diseases or other the means of controlling the flow of the biologic processes and specific environmental ex- energy, and periences or exposures. The epidemiologic model the methods available for absorbing the applied to injury research seeks to explain the Ch. 4—Safety Hazard Identification ● 73 occurrence of injuries within the system of host assessing the reliability of hardware and equip- (injured victim), agent (means of injury), and the ment. Another systems model, “fault tree analy- environment (physical, psychological, and social sis, “ involves building a logical “tree” of events factors related to the event). that can lead to undesirable outcomes. The ana- ● lyst examines component failures, which can in- Using such a model, it should be possible to clude both hardware and human errors, and at- identify features common to a set of injuries or tempts to learn what might cause these failures accidents, and either identify causes directly or and what effects on system safety they might find clues to causation. This approach has advan- have. tages over investigating each incident separately, Looking at a group of off-the-road industrial vehi- The systems-safety models have been applied cle accidents, for instance, it might become appar- most extensively in military and aerospace en- ent that one company’s products are involved in deavors with the focus on potential failure points a disproportionate number of incidents (40). in system hardware. Few quantitative data exist about human error rates, so including the human (187) (295) Gordon and McFarland were two component of this system is frequently precluded. early proponents of epidemiologic models of injury causation. Haddon (197) was successful in implementing an epidemiologic approach to trans- Ergonomic/Human Factors Models portation accidents while directing the National Highway Traffic Safety Administration. Baker The injury causation models developed by and her coworkers (40) have successfully used epi- human-factors engineers or ergonomists attempt demiologic techniques to describe work-related to provide insights into the problems of “unsafe deaths in Maryland. acts” or human error that are lacking in other in- The epidemiologic method requires the collec- jury causation models. tion and study of far more information about the Ergonomists generally analyze the interactions host, the environment, and their interactions than between workers and their machines for the the behavioral and management models do. This sources of injury causation. The limits of human approach, which recognizes the interactive nature beings to perform consistently and without errors of the injury process, is a significant advance over are important issues to the ergonomists. Rather earlier models. In fact, it has provided a frame- than viewing operators’ errors as merely “unsafe work for the application of many systems ap- acts” that can only be addressed through train- proaches that incorporate human operator vari- ing and motivation, the ergonomic approach ables, environmental factors, and task demands. looks to see if various features of the machine or The National Institute for Occupational Safety the design of the work might themselves be in- and Health is currently conducting two epidemio- ducing worker errors. These features can include logic studies to evaluate the role of personal, the presentation of information to workers managerial, and work environment factors in the through displays, the design of machine controls, etiologies of fall-from-ladder accidents and ac- and the relationships between displays and con- cidents that result in fatal injuries. Both are case- trols. In addition, ergonomists analyze the phys- comparison studies that should produce a scien- ical capacities of workers, such as lifting or tific assessment of these causal factors. reaching ability, to determine whether the task places undue stress on specified parts of the body Systems Models or leads to excessive fatigue. The emergence of systems engineering as a dis- Because of the importance of this discipline for cipline in the 1960s gave rise to many new applica- the prevention of both acute trauma and cumu- tions of systems theory, including systems safety. lative trauma, as well as its potential usefulness The various models that have been used include in the field of general workplace and equipment “failure mode and effects analysis” and “criticality design, ergonomics is discussed in greater detail analysis.” Both these are largely oriented towards in chapter 7. 74 ● Preventing Illness and Injury in the Workplace

CRITIQUE OF INJURY CAUSATION THEORIES AND MODELS

The inherent simplicity of Heinrich’s domino The human factors/ergonomics models focus theory and its historical availability no doubt ac- on the human/machine interface and thus pro- count for its widespread acceptance. A minimum vide a much-needed emphasis on understanding amount of training is required to understand its the interaction of worker and machine in order application and it does provide an answer to the to achieve a safe working environment. The thrust question of “cause,” even if it is a superficial one. of the practice of ergonomics is designing the work Since most injuries are classified as resulting from tusk, rather than merely installing machinery and “unsafe acts, “ it unfortunately allows more fun- letting the worker find a way to adapt. Thus in- damental features of workplace design that lead jury prevention can bean integral part of job de- to injuries to be ignored. sign. The principal shortcoming of this model is the absence of any analysis of hardware failures Behavioral models initially contributed to our beyond the human/machine interface. However, understanding of the human component of injury compared with the injuries that occur at that in- causation, although approaches based on study- terface, hardware failures are relatively rare. ing “accident proneness” have contributed little or no useful information for prevention. It is un- Purswell and Stephens (380) conclude that no fortunate that many firms still expend resources single model provides a wholly satisfactory ap- trying to identify “accident-prone” individuals, proach to explaining the various facets of injury rather than pinpointing features of workplace de- causation. They suggest that, for the present, the sign that lead to injuries. epidemiologic model is useful for identifying major categories of causal factors in the workplace, The adaptations of Heinrich’s theory that place and that these major categories should be stud- the responsibility for unsafe acts and unsafe con- ied in-depth using the human factors/ergonomics ditions on the management system of the enter- model. prise represent a major step forward in preventing occupational injuries, The causal explanations The quest for causal models should not be the are still too simplistic, although these approaches sole object of research on injury prevention. What do provide a limited ability to predict the occur- is even more important is the design of interven- rence of hazards in the workplace. tions to eliminate or reduce the injury hazards faced by workers. In fact, one distinguished re- The chain-of-events or multiple events models searcher has concluded that the search for causal (MORT is an example) recognize that many fac- models for injuries may ultimately be fruitless. tors influence injury causation and thus represent Singleton has recently stated that “there can never progress over single-event models. However, the be a theory which will predict an accident and current models do not have a sufficiently simple even accident rates are subject to too many organizing structure to make them useful across variables for prediction to be meaningful. ” But a wide range of industries. he adds:

The epidemiologic model has value as an orga- It does not follow that we must abandon hope nizing framework for the systematic study of the of controlling accidents. The same problem oc- factors related to various types of injuries. It is curs in other complex practical situations. The physician, for example, is often faced with a pa- limited in that, in general, it cannot adequately tient with a disease which he cannot readily iden- explain why injuries happen or how corrective tify. . . However, this does not mean that nothing measures can be identified and applied. can be done, The physician has certain general The systems models have been developed pri- principles; the temperature must not be allowed to get too high, the body must not get dehydrated marily to evaluate system, subsystem, and com- and so on. He can take action on the basis of these ponent failures. The primary focus has been on principles without waiting to identify the cause nonhuman or hardware failures. Although the po- of the symptoms. Similarly in accident preven- tential exists for incorporating human error rates tion we can take action to increase safety with- into these analyses, the data to do so are currently out waiting for a theory of accident causation very limited. (442). Technologies for Controlling Work-Related Illness

. Contents Page Control Systems ...... 78 Control at the Source ...... 80 Controlling Dispersion...... 82 Control at the Worker...... 85 Integration of Health Hazard Controls into Workplace Management ...... 85 Case Study: Controlling Worker Exposure to Cotton Dust ...... 85 Byssinosis ...... 85 The OSHA Cotton Dust Standard ...... 86 Changes in Cotton Dust Levels ...... 86 Costs of Compliance with the Cotton Dust Standard ...... 88 Case Study: Controlling Worker Exposure to Silica Dust...... 89 R,gulatory Activities for Silicosis Control ...... 90 Control Technologies: Engineering Methods ...... 90 Control Technologies: Personal Protection and Administrative Controls . . . . 91 Strategies for Silica Dust Control ...... 91 Case Study: Controlling Worker Exposure to Lead ...... 92 Some Features of the OSHA Lead Standard ...... ,...... 92 Control Methods: Engineering and Respirators...... 93 Medical Removal Protection ...... 93 Changes in Air Lead Levels ...... 94 Changes in Blood Lead Levels ...... 95 Costs ...... 96 Summary of Improvements ...... 96 Extent of Control Technology Usage in the United States ...... 97 Information About Controls and Areas for Research ...... 97 Summary ...... 99 LIST OF TABLES Table No. Page 5-1, Principles of Controlling the Occupational Environment ...... 80 5-2, Benefits of New Technologies for Controlling Worker Exposure to Vinyl Chloride Monomer ...... 81 5-3. Suggested and Recommended Levels for Cotton Dust Exposure ...... 88 5-4. Cotton Dust Measurements Before Promulgation of the OSHA Cotton Dust Standard and Percentage of Companies Claiming Compliance with the Standard in North Carolina ...... 88 5-5. Estimated and Realized Costs of Compliance with the OSHA Cotton Dust Standard ...... 89 5-6. Measures To Reduce Air Lead and Blood Lead Levels ...... 93 5-7. Blood Lead Levels That Trigger Medical Removal From and Return to Lead-Contaminated Atmospheres...... 93 5-8. Medical Removal Protection Transfers in a Sample of Lead Industry Plants ...... 94 5-9. Reductions in Average Air Lead Levels, 1977-78 and 1981-82...... 95 5-10. Average Blood Levels Before and After Promulgation of the OSHA Lead Standard ...... 95 5-11. Projected Industry-Wide Annual Costs of Compliance With Air Lead Levels of 50 to 150 micrograms/m3 ...... :...... 96 , LIST OF FIGURES Figure No. Page 5-1. Generalized Occupational Exposure ...... 77 5-2. Generalized Model for Control of Workplace Hazards ...... 78 S-3. Vinyl Chloride Reactor System ...... 81

5. Technologies for Controlling Work-Related Illness

This chapter describes the principles and tech- Figure 5-1 .—Generalized Occupational Exposure nologies for controlling workplace health hazards—toxic substances and harmful physical agents found in the workplace. For clarity and since the control principles are similar for both toxic substances and harmful physical agents, discussion focuses on control of the former. Empha- Source Workplace Worker sis is given to technologies proven to be the most effective for protecting workers’ health—those that prevent hazard generation or that prevent ~ worker contact with the hazard. Three case stud- Generation Transfer Exposure ies commissioned by OTA illustrate these prin- —Gas — Respiratory ciples and technologies as applied in controlling — Liquid — Dermal work-related exposure to cotton dust, silica, and —Solid — Ingestion — Energy lead. In addition, the extent of the use of control technologies in United States workplaces is discussed. Health hazards, as defined by public health science, cause disease by an agent (hazard source) transmitted through the environment by a vector (transmission of hazard) to a host or a receptor (worker) who is affected. This model includes er, such as protective clothing or a respirator, are workplace hazards to which workers are exposed generally called “personal protective equipment.“ (see fig. 5-1). For workplace hazards, the source– A hierarchy of control methods is commonly the point at which the hazard is generated—may used. The first choice is control at the source, be a gas, a liquid, or a solid if it is a substance, which can be done by design or modification of or a form of energy if it is a physical agent. Trans- a process or equipment or by substitution of less mission or dispersion of the toxic substance or hazardous materials. If the source is unalterable harmful physical agent is generally through work- through design or substitution, the next choice is place air or by direct contact. The worker at risk to control or contain the dispersion of the con- may receive (absorb) the hazard through inges- taminant by isolation of the source, preventing tion, the skin, or by inhalation (see fig. 5-2). the toxic substance from becoming airborne, or A control technology system can include hazard by removing the contaminant through local ex- control at any or all of these three points—source, haust or general dilution ventilation. Finally, con- transmission, or worker. Hazard controls applied trol at the worker may include administrative con- at the source, such as isolation of a process, or trols, personal protective equipment, and work in the transmission or dispersion path, such as practices. (Personal protective equipment is dis- local exhaust ventilation, are generally called cussed in ch. 8, and the hierarchy of controls is “engineering controls. ” Those worn by the work- discussed in ch. 9.)

78 ● Preventing Illness and Injury in the Workplace

Figure 5-2.—Generalized Model for Control of Workplace Hazards

Zone I source

Zone Ill receptor

Production process

CONTROL SYSTEMS There have been many attempts to define control technology. Brandt (71) described it as a system designed to control contaminant emission and dispersion along the pathway to the worker. Bloomfield (61) cited ventilation to reduce levels of airborne contaminants as the primary means of engineering control. The International Labour Office (229) includes several techniques in control technology: ventilation; process changes; substitution of process, equipment, or material; isolation of stored material, equipment, process, and workers; and education of management, engineers, supervisors, and workers. Caplan (96) defined engineering controls for industrial hygiene purposes as “installation of equipment, or other Photo credit: NIOSH physical facilities, including if necessary selection This electrostatic precipitator is used to remove oil mists from the atmosphere of a machine shop and arrangement of process equipment, that significantly reduces personal exposure to occupational hazards.” Smith (450) defined control technology as substituting less dangerous substances, limiting contacts between worker and toxic mate- equipment, or processes; limiting releases or pre- rials by personal protective equipment; and in- venting buildup of environmental contamination; troducing administrative changes.

Ch. 5—Technologies for Controlling Work-Related Illness ● 79

For this assessment, a hazard control system includes: 1. control at the emission source by substitution of materials, change of process or equipment, or other engineering means, 2. control of the transmission or dispersion of the contaminant by isolation, enclosure, ventilation, or other engineering means, and 3. control at the worker by personal protective equipment, work practices, administrative control, training, or other means. The controls in No, 1 and No. 2 are commonly called “engineering controls. ” Training workers, supervisors, managers, engineers, and other concerned persons about a hazard and its control underlies the effectiveness of control solutions. Hazard-free operation requires rigorous maintenance of controls, and good housekeeping is essential to control secondary sources of contamination. Work practices (e.g., instructions that liquids should be poured away from the worker) and administrative procedures (e.g., that workers spend limited time in the presence of hazards) are also important parts of a control system. Table 5-1 is a compilation of hazard control principles and includes examples of control measures.

One tenet of effective hazard control is that a Photo credif: OSHA, Office Of Informatlon and Consumer Affairs system should be designed in a way that the con- Engineering controls include the enclosure of trols are automated or inherent in the operation operations and using remote controls. This photo of the system. Thus, hazard controls should func- illustrates equipment designed to handle very toxic tion even in the absence of continuous worker and radioactive materials manager attention. For instance, enclosing a process to prevent emission of toxic substances to Because of the continuing need for human in- workplace air is a more reliable, and likely less tervention and attention in the use of personal expensive, control than respirators, where effec- protective equipment, practicing industrial hy- tiveness is difficult to measure, protective fit is gienists employed by business, government, and difficult to achieve. Although systematic design unions have long recognized that such equipment will consider a variety of control methods and should be turned to only after other means of pro- combinations, engineering solutions are preferred tection have been exhausted (see ch. 9). Occupa- because they depend less on routine human in- tional Safety and Health Administration (OSHA) volvement for effectiveness. For example, ground- standards require the use of engineering and work ing home electrical appliances provides greater practice controls except for the time period nec- protection against electrical shock than instruc- essary to install such controls, when engineering tions to remember not to simultaneously touch and work practice controls are infeasible (in- an ungrounded appliance and a metal surface. cluding many repair and maintenance activities), 80 . Preventing Illness and Injury in the Workplace

Table 5=1 .–Principles of Controlling the The industrial hygiene literature repeatedly points Occupational Environment to source control as the most effective means of Point of application of preventing work-related illness. the control measure Control measure At or near the hazard Designing Controls zone...... Substitution of nonhazardous or less Designing equipment to eliminate contact be- hazardous material tween hazard and worker is the most effective way Process modification Equipment modification to control exposure (71). The control of vinyl Isolation of the source chloride monomer (VCM) provides an example Local exhaust ventilation of successful design eliminating a health hazard Work practices (housekeeping) (see also box N in ch. 12). In the 1960s, before To the general workplace VCM was recognized as a carcinogen, it was iden- environment ...... General dilution ventilation tified as a cause of acro-osteolysis (bone deteriora- Local room air cleaning device tion, especially in the finger tips). This finding led Work practices the American Conference of Governmental Indus- (housekeeping) trial Hygienists (ACGIH) to revise the Threshold At or near the worker . . . . . Work practices Limit Value (TLV) exposure limit from 500 parts (housekeeping) Isolation of workers per million (ppm) to 200 ppm in 1970 (5). Personal protective equipment Revision of the exposure limit meant that the Adjuncts to the above firms that followed ACGIH recommendations had controls ...... Process monitoring to find ways to reduce worker exposure. Analy- systems Workplace monitoring sis by design engineers identified two methods by systems which the high exposures associated with clean- Education of workers and ing the VC reactor vessel could be reduced: elim- management Surveillance and ination of reactor fouling or mechanical or chem- maintenance of controls ical removal of the polymer buildup. Hydraulic Effective process-people reactor cleaning technology was adopted that re- interaction and feedback duced the frequency of worker cleaning from once SOURCE (576) per several reactor charges (loading the reactor) to once per 25 to 30 charges and thereby reduced when they are insufficient, and in emergencies (see worker exposure (256). ch. 9). For instance, engineering solutions to reduce airborne lead concentrations to the OSHA When VCM exposure was recognized in 1974 standard are difficult to apply in lead smelters, as strongly related to angiosarcoma of the liver and OSHA allows respirator programs while the (a rare and deadly cancer) by health professionals, solutions are engineered. OSHA mandated a permissible exposure limit of 1 ppm. Feasible engineering and work practice Of course, the nature of some jobs requires reli- controls were required to reduce exposure below ance on personal protective equipment. For in- this level (617). stance, firefighters depend on self-contained breathing apparatus when fighting fires. Again, industrial hygiene analysis determined that exposure to gases during reactor cleaning was Control at the Source a major problem. Re-investigation led the design engineers back to earlier considerations, of either Control at the source can be achieved by de- eliminating the fouling or finding an automated sign of new or modification of existing processes cleaning method. But this time the design criterion or equipment, or by the substitution of less haz- was to reduce drastically exposure from over 200 ardous materials-all done, preferably, before the ppm down to 1 ppm, and mechanical cleaning process or equipment is installed and operated. alone was found to be inadequate. However,

Ch. 5—Technologies for Controlling Work-Related Illness ● 81

spraying a simple coating solution on interior re- Figure 5-3.-Vinyl Chloride Reactor System actor walls before mixing each batch prevented polymer buildup, Automating and enclosing the reactor cleaning process by installing a permanently mounted nozzle inside the reactor (see fig. S- Coating Water 3) very effectively contained the VCM gases and solution greatly reduced worker exposure (256). Commercial use of this design demonstrated that the new reaction vessels needed cleaning only Internal flush once every 500+ polymerization batches, greatly nozzle improving the productivity of the process. The Coating developer, B.F. Goodrich, now uses the innova- solution Reactor tive process in its vinyl chloride monomer plants tank both here and abroad and also licenses it worldwide to other chemical manufacturers. Table 5-2 shows the benefits of this control technology I (256). I This example illustrates the advantages of applying engineering controls to the prevention of work-related illness. Engineers sought solutions to a recognized health problem by first considering methods that would eliminate exposure such as by automating cleaning or by preventing buildup of materials that require removal. This example also shows that production costs can be reduced and productivity increased, as Brandt Table 5-2.—Benefits of New Technologies for Controlling Worker Exposure to postulated some 35 years ago in his book on oc- Vinyl Chloride Monomer (VCM) cupational health engineering (71). Reduction in worker exposure to VCM. Health hazards can also be eliminated or con- Reduction in VCM emissions to the atmosphere. trolled by changing an industrial process. For ex- Closed reactor operation—entry only for normal ample, the National Institute for Occupational maintenance. Savings in labor. Safety and Health (NIOSH) recently conducted Reduction in reactor downtime due to cleaning and, as a a study of dry cleaning machine operators exposed result, increase in productivity. to perchloroethylene, a widely used solvent, Polymer buildup lost as scrap is eliminated. Reduction of rupture disc changes due to polymer known to cause contact dermatitis, central ner- buildup. vous system depression, liver damage, and anes- Constant and maximum process side heat transfer thetic death. NIOSH investigators found higher coefficient in the reactor. exposure levels of perchloroethylene vapors in SOURCE: (25S). processes involving separate washing and drying itself harbor toxic properties. For example, asbes- machines than in processes that combined these tos, an excellent insulator, is found widely in two steps in one machine. The two-step process buildings, ships, and other places requiring ther- requires manual transfer of clothes, resulting in mal insulation. However, as its toxic properties, unnecessary worker exposure, which is avoided especially its carcinogenicity, were recognized, in the combined process. other materials were considered as a replacement. Several materials are suitable, depending on the Substitution application and the temperature range to be in- Substitution of a less toxic agent for a more sulated. These include insulating concrete, ver- toxic one is an important means of control, but miculite, fiberglass, and rockwool. While none care must be taken that the substitute does not of these is yet known to cause cancer, precautions 82 ● Preventing Illness and Injury in the Workplace should be taken to control exposure to these tion (to prevent dust, fumes, or vapors from materials during installation (80). escaping into occupied areas); automation, through the use of unattended machines; and dis- Silica dust, which can cause lung disease, is one tance, to place operations away from workers. of the oldest known occupational health hazards, and its control well illustrates the principle of sub- Isolation by enclosure has been used effectively stitution (see case study, later in this chapter). to reduce silica exposure in foundries (359,569, Silica dust is a problem in “sand blasting, ” in 577). Abrasive blasting operations maybe located cleaning and polishing moldings and metals, and in enclosed, ventilated booths. Enclosure is also in mining and quarrying, where it is generated by used to reduce worker exposure in the asbestos explosives and mining machinery. textile industry. Card machines, among the dusti- est parts of the asbestos textile manufacture proc- In foundries, silica dust is generated during ess, can be completely enclosed and asbestos dust cleaning, during chipping and grinding of castings filtered from the air exhausted (80). Enclosure has because some sand from the cores and molds re- been applied successfully in containing contamina- mains on the castings, and during abrasive clean- tion from radioisotopes since the beginning of the ing, which generates airborne silica dust. If nuclear industry. A variation is to protect work- abrasive cleaning is performed by sand blasting, ers from physical and chemical hazards by locat- silica dust may be generated from both the blast ing their work stations in ventilated control sand and the mold and core sand. booths. The most direct method of eliminating silica Many jobs with risk of exposure to toxic sub- dust is to make substitutions for silica-containing stances can be automated. For instance shakeout material. A number of silica-sand substitutes are (a method for removing foundry sand from molds available for abrasive blasting, including metallic or parts) in a foundry can be done by ventilated shot and grit, garnet, nut shells, cereal husks, and machines rather than by hand. Automobiles may sawdust, and have been widely used in abrasive be spray painted or welded by automated ma- blasting operations and to some extent in found- chines to remove workers from exposure to spray ries (560). paint and solvent and welding fumes, respectively. In some cases, silica dust can be eliminated by Finally, explosive or extremely toxic materials substitution of a nonabrasive process—by clean- can be stored in remote and inaccessible areas and ing castings by the salt bath process, acid pickl- hazard-generating operations may be removed ing, or ultrasonic cleaning. Water jetting and la- from areas where workers are concentrated. ser cutting to remove excess metal from castings Open-air sand blasting can be done at a distance have been considered as alternatives to chipping from other work sites to reduce the number of and grinding (435). workers at potential risk. Persistently leaky pumps and piping for the transport of toxic sub- Controlling Dispersion stances can be isolated by placing them in areas If a source cannot be altered through design or remote from workers. substitution, the next choice is to control or contain the dispersion of the contaminant. This may Wetting be done by isolating the source, preventing the Wetting dust to prevent it from becoming air- toxic material from becoming airborne, or by ven- borne is used to reduce worker exposure. Spray- tilation. ing is a primary means of dust control in mining, but it is considered to be inadequate alone and Isolation is usually used in conjunction with ventilation Isolation of a process involves the placement (230,394). Substitution of wet processing and of a barrier between the process and the worker. spraying for dry operations has been widely used In dusty operations for example, there are three to control silica dust. In foundries, adding mois- basic means of isolation: enclosure of an opera- ture to sand has been found to reduce dust con-

Ch. 5—Technologies for Controlling Work-Related illness ● 8.3 centrations substantially (435,569). By contrast, ducing control effectiveness. Furthermore, energy wet processing in the manufacture of portland ce- costs are increased to heat the air that replaces ment appeared to have no effect on respirable dust the exhausted air. levels (419). Using a jet of air “upwind” from the exhaust pushs the emissions toward the exhaust. This is Local Exhaust Ventilation commonly referred to as push-pull ventilation. Local exhaust ventilation is one of the most NIOSH showed that push-pull ventilation con- commonly used engineering controls. It aims to trolled emissions from chrome plating tanks with protect the worker by capturing generated gases, just 25 percent of the exhaust needed if only pull vapors, fumes, or particles in an exhaust air was used. The system thus controlled emissions stream and discharging them away from work- and reduced energy costs (582). ers. Examples are laboratory fume and kitchen- A successful local exhaust ventilation sys- range hoods, both of which use fans to exhaust tem.—As already indicated, controlling exposures contaminated air, Industrial operations are often is best done by considering design of the health placed in hoods to obtain maximum contaminant hazard control at the time a process is established control with minimal exhaust air volume. and carefully monitoring performance of the sys- For example, local exhaust can be applied in tem. Anderson (20) describes the effective design aluminum reduction operations to reduce worker of a control system in a large electronics plant. exposure to carcinogenic particulate, in spray paint booths to control paint mist and solvent The process begins when a manufacturing engi- vapors, in garages to control carbon monoxide neer asks to add or change a chemical process. from auto exhaust, and in foundries to control The request is submitted to the facilities engineer- silica exposure from abrasive blasting and grinding. ing department and an engineer is assigned responsibility for installing the equipment to satisfy NIOSH is currently investigating “push-pull” process, safety, health, and other requirements. ventilation. Generally, local exhaust ventilation Part of the facilities engineer’s responsibility is to depends on “pulling” air away from the opera- review the need for local exhaust ventilation with tion and exhausting it at some distance from the the industrial hygienist, who is responsible for pro- worker, If the emission source is over two feet viding health protection information including de- from the exhaust, a great quantity of room air tails about hood design and air volume require- must be pulled into the exhaust, significantly re- ments. The preliminary design is then reviewed by the environmental engineering department to determine the need for air cleaning devices and emission permits. After the process design is completed, it is given a final review by the industrial hygiene, environmental engineering, safety, maintenance, and manufacturing engineering departments.

Installation is supervised by a coordinator who ensures that contract specifications are followed, Changes must be approved by the facilities engineer. The contract coordinator informs the facilities engineer when the job is done and puts a warning tag on each completed hood. Before the hood can be used it must be adjusted to meet design specifications by the facilities engi- This hood in a secondary lead smelter illustrates the neer and the maintenance ventilation technician, use of local exhaust ventilation who enters information about the system in a data 84 . Preventing Illness and Injury in the Workplace

base for scheduling preventive maintenance and the gradual introduction and mixing of fresh air who also tags the hood to indicate that this has with, and exhausting of, workplace air. Con- been done. After this the hood is inspected by the tinuous air exchange in buildings reduces non- industrial hygienist, who reviews its use with the taminants that resist other control means while workers and ensures that the proper chemical contributing to maintenance of a comfortable en- identification labels are placed at each station. vironment. General dilution ventilation is defined as “the process of supplying or removing air by Hood effectiveness is measured periodically and natural or mechanical means, to or from any data entered into a computer. Each week the com- space” (71). The air circulation systems found in puter system generates a card for each hood per- most buildings are examples of general dilution forming below specified levels for review by the ventilation. industrial hygienist. If the hood is in need of attention, the card is forwarded to building main- This technique requires careful planning, and tenance. If that department is unable to fix the it can fail if inadequate consideration is paid to hood, the facilities engineering department treats contaminant generation rates. Furthermore, pro- the failure as a unique project, and then follows vision must by made for adequate fresh “makeup” the same procedure that is used in designing a new or “replacement” air, for heating or cooling the hood, makeup air, and for avoiding contamination of makeup air. If a hood is found to be dangerously deficient by the ventilation technician, it is tagged “Do Not Recent interest in energy conservation has Operate” and immediately reported to the depart- added new considerations. Increased building in- ment manager, facilities engineering department, sulation has greatly reduced the flow of air from and industrial hygiene department. “leaks,” which requires more makeup air. Chap- ter 16 describes particular problems among office The main features of this well-thought-out sys- workers in new “tighter” buildings, Office work- tem for designing and managing controls are: ers report health effects from microorganisms, ● coordination among all concerned parties, organic chemicals, asbestos, tobacco smoke, and . integration of occupational health concerns other sources in buildings with inadequate ven- at the beginning and throughout the design tilation (25). process, . integration of occupational health concerns Control by general ventilation is aided by removing sources, such as smoking, and by clean- following installation, and ing air. Since most building ventilation systems ● execution of a well-planned preventive main- now recirculate air, cleaning the air becomes espe- tenance program. cially important. This is a relatively new prob- The company has found that this approach lem; before energy conservation was given em- greatly lowers costs by reducing the need to phasis, accepted engineering practice was to retrofit processes. Before this method was completely exchange building air to avoid con- adopted, newly installed exhaust systems fre- tamination buildup. Now, building air is often quently failed because of improper design or in- cleaned and then recirculated to reduce energy stallation. Post-installation approval guarantees cost. Systems are available for cleaning both gas all concerned parties that the system works from and particulate, but care must be taken to ensure the start as it was designed. A well-planned, that the system is reliable and the cleaning com- computer-based, preventive maintenance pro- plete (563). gram assures continued effectiveness. Neither local nor general ventilation acts to prevent generation of hazards; it can only capture or dilute contaminated air and take it to another General Dilution Ventilation location. The air may still have to be cleaned While local exhaust systems are applied at a before discharge to the ambient environment, to particular point to remove contaminants at rela- meet Environmental Protection Agency or other tively high rates, general dilution ventilation is ambient-air standards (6,562,563). Ch 5- Technologles for Controlling Work-Related Illness ● 85

CASE STUDY: CONTROLLING WORKER EXPOSURE TO COTTON DUST 86 . Preventing Illness and Injury in the Workplace —

The quote was made in reference to a respira- adopted as a startup standard by OSHA in 1971 tory disease suffered by workers in English- tex- (see Ch. 12). tile mills 150 years ago. That disease—byssinosis, In 1974, ACGIH revised its TLV downward to or “brown lung’’—was recognized in this country 3 200 micrograms/m (the method of measurement much later than in Europe. The reasons for the changed also, and the “new” 200 micrograms/m3 late recognition are complex. Many occupational is not directly comparable to the ‘ old” 1,000 health authorities suggest ignorance or refusal to micrograms/m 3). That same year, the Director of recognize particular respiratory diseases that were NIOSH recommended that exposure to cotton common to mill workers to spare employers the dust should be reduced to the lowest feasible level, costs of installing controls. In addition, social con- and that it should in no case exceed 200 micro- ditions inhibited workers from making their com- grams/m3. plaints known and prevented actions on those complaints. Also, local and State Governments In 1976 OSHA proposed a 200 micrograms/m3 were reluctant to act because they feared the loss standard. The final standard, issued in 1978, set of textile industry jobs as a result of requiring pre- three different exposure limits—200 micrograms/m3 vention of work-related injury and illness. Finally, for cotton yarn manufacturing, 750 micrograms/ a lack of scientific studies showing an association m3 for “slashing and weaving” operations, and 500 between cotton dust and illness in the United micrograms/m 3 for exposures in other operations. States contributed to the tardy recognition of the This standard was contested by the textile indus- disease and inhibited action to prevent it until try through legal suits. While the Supreme Court OSHA came into being (124). upheld the standard for the textile industry in 1981, in the same year the current administration moved to reconsider it. This action is pending. The OSHA Cotton Dust Standard Table 5-3 shows how the suggested and recom- Although the exact disease-causing agent with- mended levels for cotton dust came downward in cotton dust has eluded identification, it is after the substance was regulated as a health haz- known that the dusts from the early stages of ard in the United States. processing are more hazardous than those from later stages. Opening cotton bales and sorting, Changes in Cotton Dust Levels picking, and blending raw cotton present greater Table 5-4 presents North Carolina Department risks than do weaving and finishing. of Labor measurements of the percentage of tex- In 1964, the American Conference of Govern- tile plant departments that were in compliance mental Industrial Hygienists considered the evi- with the OSHA cotton dust standard i n 1981. As dence for establishing a recommended limit for can be seen, just two years after promulgation of cotton dust exposures. Two years later, the Con- the new standard and during the period the stand- ference agreed on a Threshold Limit Value of ard was being challenged in the courts, over half 1,000 micrograms/m3 as the maximum exposure the departments complied with the standard. that was consistent with maintaining workers’ Some problems remain, as higher frequencies of health. In 1969, the Secretary of Labor incor- noncompliance were found in the early stages of porated ACGIH’s recommended TLV into Fed- the process-opening, picking, carding, drawing, eral standards for employers with Government and combing. In these stages, workers are exposed contracts (see ch. 11 for a discussion of the Walsh- to the more hazardous dusts associated with un- Healey Act). processed cotton. overall, however, the cotton industry is coming into compliance with the new The Occupational Safety and Health Act of standard. 1970 required that the newly established OSHA adopt the Walsh-Healey Act standards and apply The industry trade association, the American them to all the Nation’s workplaces. Thus, the cot- Textile Manufacturers Institute, estimates that ton dust standard of 1,000 micrograms/m3 was about 75 percent of the industry was in compli-

ance within two years of the standard being introduced. Some plants that have been completely modernized are in full compliance (413). In 1981, the U.S. textile industry purchased $1.6 billion worth of new machinery. About 70 percent of those purchases were for the purposes of modernization to increase productivity (413) in Photo credit O. SF/A, Office of Information and Consumer Affairs the face of increased foreign competition and, to The spinning of cotton fibers into yarn and weaving some extent, to comply with the OSHA stand- yarn into fabric are two of the operations regulated by ard for reduced cotton dust levels. the OSHA cotton dust standard. In recent years, the textile industry has invested heavily in modernized Ruttenberg (413) concludes that it is impossi- equipment in order to comply with the standard and ble to decide the relative importance of increas- to improve productivity ing productivity and compliance with OSHA regulations in the modernization of the American and purchasing. Machine suppliers modified textile industry’, but that both have made a con- equipment to comply with OSHA regulations and tribution. this equipment has been accepted on a worldwide basis as well as in the USA. The dust controls The U.S. Occupational Safety and Health have also contributed to much better operating Administration dust regulations have had a dra- results. . . (U.S. Department of Commerce (551). matic effect on . . . processing equipment design Quoted in 413). 88 . Preventing Illness and Injury in the Workplace

Table 5-3.-Suggested and Recommended Levels for Cotton Dust Exposure

Level a Organization Year (micrograms/m3) American Conference of Governmental industrial Hygienists (ACGIH) ...... 1964 1,000 tentative recommendation ACGIH recommendation ...... 1968 1,000 formal Secretary of Labor ...... 1968 1,000 Walsh-Healey Act standard Occupational Safety and Health Admin- istration (OSHA)...... 1971 1,000 OSHA standard British Occupational Hygiene Society ...... 1972 500 recommended standard for Britain ACGIH recommendation ...... 1974 formal National Institute for Occupational Safety and Health (NIOSH) ...... 1974 200 recommendation OSHA ...... 1976 200b proposed standard OSHA ...... 1978 200 final standard ~he levels from 1964 through 1972 were based on techniques that measured the concentration of total dust in the workplace atmosphere. From 1974 on, the levels are based on the use of the vertical elutriator —a device that measures the quantity of small, resplrable dust particles. Levels based on the these two methods are not dirwctly comparable %he 200 limit is for yard manufacturing, 750 for slashing and weaving, and 500 for all other processes The Iim!t goes up as the cotton dust becomes cleaner SOURCE Adapted from (413).

Table 5-4.—Cotton Dust Measurements Before Promulgation of the OSHA Cotton Dust Standard and Percentage of Companies Claiming Compliance with the Standard in North Carolina

Companies claiming Range of measurements Limit under compliance in before OSHA standard OSHA standard North Carolina Area of plant (micrograms/m3) (micrograms/m 3) (percent) Opening ...... 300-3,000 53 Picking...... 700-1,700 200 61 Carding ...... 300-1,800 200 52 Drawing ...... 400-800 200 63 Combing ...... NA 200 61 Roving ...... NA 200 81 Spinning ., , ...... 200-300 200 83 Winding ...... 1,200 200 76 Twisting...... 1,200 200 80 Slashing...... NA 750 100 Weaving ...... 400-1,000 750 96 Knitting , ...... NA 100 Waste Processing . . . . . NA 85 Other ...... NA 500 97 SOURCE: (413).

Tougher government regulations on workers’ Costs of Compliance with the health have, unexpectedly, given the [U. S.] indus- Cotton Dust Standard try a leg up. Tighter dust-control rules for cotton plants caused firms to throw out tonnes of old in- OSHA contracted for an economic analysis of efficient machinery and to replace it with the latest the expected costs of compliance with the cotton available from the world’s leading textile machin- dust standard, and the contractor assumed that ery firms. (The Economist (160). Quoted in 413). compliance would be accomplished by “add-on” Ch. 5—Technologies for Controlling Work-Related Illness ● 89 ventilation equipment. However, the availability costs for compliance were nearly $2 billion (in of newer production equipment, which increased 1982 dollars). At the time of promulgation in productivity and reduced cotton dust exposures, 1978, OSHA estimated costs of just under $1 bil- resulted in much lower costs than those estimated lion (in 1982 dollars). Thus, while cost estimates at the time the standard was considered. As table plummeted more than 50 percent by the time the s-5 indicates, the initial 1974 estimates of capital standard was issued, the reduced estimate was still almost four times higher than the actual costs re- Table 5-5.—Estimated and Realized Costs of ported in 1982 in a poststandard contract report. Compliance with the OSHA Cotton Dust Standard Although most of the more productive, less Millions of dusty machinery now in use in U.S. textile mills 1982 dollars was available in the mid-1970s, its potential use Preregulatory estimates OSHA contractor, 1974 ...... 1,941 was ignored in the early estimates of compliance Revised OSHA contractor, 1974 ...... 1,388 costs. Even if purchase of new technology had a ATMl contractor, 1977 ...... 875 been anticipated, it would have been difficult to OSHA, 1978 ...... 970 assign the proper fraction of its costs to dust con- Postregulatory estimate OSHA contractor, 1982 ...... 245 trol. In the event, new technologies greatly re- aAmer~can Textile Manufacturers !rTStitUte duced the costs. SOURCE (413)

CASE STUDY: CONTROLLING WORKER EXPOSURE TO SILICA DUST

Silica is a major component of the earth’s crust; may result in respiratory failure and secondary it is the sand covering the beaches, the sand heart disease. Tuberculosis and other pulmonary sprinkled on icy winter streets, the grit in the dust infections may complicate acute or chronic sili- on windy days—it is everywhere. It is also widely cosis and significantly shorten life expectancy. used in industry. Over 402 million tons of silica- Hickey, et al. (210) discuss these silica-related containing sand were produced in the United health problems and reported associations be- States in 1980. Of this total, nearly 300 million tween worker exposure to silica dust and an in- tons were used for glassmaking, as molding sand creased risk of lung cancer. in foundries, and as industrial abrasives. Since it is ubiquitous, silica is frequently found as an un- Since diagnostic procedures do not detect wanted constituent of ores mined for other minerals. silicosis at a reversible stage, and since medical In those cases, it must be removed and discarded. treatment will not alter the course of the disease after it is found, emphasis on exposure control Silicosis is a disabling lung disease resulting is imperative. Yet, even though the cause of the from the inhalation, deposition, and retention in disease has been well understood and technologies the lungs of respirable crystalline silica dust. Acute for controlling exposure have been available for silicosis can occur within six months following ex- decades, silicosis continues to occur in the United posure to extremely high silica dust concentra- States at an alarming rate. A minimum of 59,000 tions. Silicosis victims appear to suffer more cases of silicosis may be expected based on knowl- episodes of chest illness than workers without the edge about current exposure levels and numbers disease. The mortality for nonmalignant respira- of exposed workers at risk in 1980 in U.S. indus- tory disease is significantly higher among work- try (210). ers receiving compensation benefits for silicosis than in the general population. A complication Hickey, et al. (210) estimate that there are 1.3 of silicosis, progressive massive fibrosis, results million production workers with potential expo- in significant impairment in lung function and sure to silica dust—40 percent of whom are in 90 ● Preventing Illness and Injury In the Workplace workplaces lacking exposure control. Historically However, due to the serious silicosis problem, the most severe exposures to silica have occurred OSHA has made a special effort to enforce the in granite and stone working, foundries, mining, existing silica standard. In 1972, silica was one and abrasive blasting. Workers producing and of five major health hazards selected for special using silica flour (silica ground so fine that it ap- enforcement efforts in the ‘Target Health Hazard pears to be refined grain flour) have recently been Program” (414). Silica was again given priority recognized to be at high risk for silicosis, because in the 1975 National Emphasis Program, as one of the extremely fine size of the particles produced. of the major worker health hazards in foundries (339). In both cases OSHA industrial hygienists Regulatory Activities for focused health inspections on plants where silica Silicosis Control was likely to be found. The current OSHA standard for silica is based Control Technologies: on an equation that limits the total amount of free Engineering Methods silica to 100 micrograms per cubic meter. This standard was adopted as a start-up standard in Silicosis is an entirely preventable disease. Ex- 1971 (see ch. 11). Evaluation of the silica stand- posure occurs whenever materials containing crys- ard shows that it may be inadequate at its pres- talline-free silica are processed and dust is gener- ent level. In 1974, NIOSH recommended limiting ated. Processes include abrasion (sand blasting, silica exposure to 50 micrograms per cubic me- grinding, milling, etc. ) that creates dusts of par- ter—half the current level. The studies on which ticularly small particle size (less than 5 microme- NIOSH based its recommendation used pulmo- ters in diameter). These dusts are too small to be nary function performance as the measure of easily seen as a “cloud.” Too small to settle, they health effect—a more sensitive indicator of remain airborne and “respirable’’—-that is, they silicosis than X-ray methods. may readily pass through the upper respiratory passages and be deposited in the alveolar spaces In certain circumstances, such as in abrasive of the lung (the small air sacs deep in the lung blasting where alternatives to silica are available, where gas is exchanged with the blood). substitution may be the most appropriate method of control. The United Kingdom banned the use The most direct method of eliminating silica of silica sand for abrasive blasting in 1948, and dust is to substitute less hazardous materials for NIOSH has recommended a similar prohibition the silica-containing material. This control has in this country (560). Sweden banned silica as an been widely used in abrasive blasting operations abrasive in manual abrasive blasting in 1981 (210). and to some extent in foundries. Silica-sand sub- A California standard requires that prior to use, stitutes include metallic shot and grit, garnet, nut not more than 1 percent, by weight, of abrasive shells, cereal husks, and sawdust. Olivine (mag- sand must pass a No. 70 U.S. standard sieve (0.3 nesium iron silicate) has been used for mold mak- mm). After use, the sand must have no more than ing in foundries to reduce silica dust exposure, but 1.8 percent of its weight as particles 5 micrometers it is not clear how effective this method will be or less in diameter (211). These restrictions on size (210). reduce the number of respirable particles. Process change may also be used to control In 1978, OSHA conducted a technological fea- silica dust exposure. For instance, water may be sibility assessment and economic impact analy- added to foundry molding sand or sprayed on at sis for a specific standard addressing use of silica the point of dust generation in granite sawing and sand in abrasive blasting (211). The study con- processing of portland cement. In some situations, sidered three alternatives: banning use of silica dust-producing abrasive processes may be re- sand in abrasive blasting, setting minimum cri- placed by other types of cleaning such as salt teria on size and hardness of blasting sand, and baths, acid pickling, or ultrasonic cleaning. Water controlling exposure through work practices. To jetting and laser cutting for removal of excess date no revised standard has been issued. metal from castings have been considered as alter- ——

92 ● Preventing Illness and Injury in the Workplace substitutes that are suitable for replacing silica. should be refined to provide a better way of Also, for those situations where engineering con- worker protection. Information about the toxicity trol maybe infeasible, further improvement in res- of silica and technologies for controlling exposure pirator effectiveness is necessary. Medical proce- could be provided to workers and employers using dures for detection of the early stages of silicosis it.

CASE STUDY: CONTROLLING WORKER EXPOSURE TO LEAD

Early efforts against industrial lead intoxication improved as measured by reduced lead levels in in this country were championed by Alice Hamil- their blood. ton. Her autobiography, Exploring the Dangerous Trades (199), presents many examples of terrible Some Features of the OSHA exposures that were corrected when managers and Lead Standard owners were convinced that lead was causing the “colic, “ “lead fits, ” and blindness that occurred The lead standard sets limits on ambient con- in lead workers. Until they were convinced, own- centrations of the metal in workplace air, requires ers and managers preferred to believe that the ill- engineering controls and work practices to reach nesses resulted from bad personal habits—drink- those limits, and requires that workers be in- ing, smoking, or the consumption of coffee. formed about lead, its effects, and the methods used to protect against them. TWO features— Some firms refused to act voluntarily, and Medical Removal Protection (MRP) and the ex- states began passing “lead laws” in the 1910s that tended time periods granted to selected industries set limits on occupational exposures. These early before engineering controls are required—dis- efforts were the forerunners of the revised OSHA tinguish the lead standard from other OSHA lead standard, which was issued in 1978. health standards. The current standard regulates exposure to lead MRP requires employers to measure workers’ in over 40 different industries. With only few ex- blood lead levels regularly. If the measured ceptions, most industries comply with the sO concentration of lead in the blood exceeds certain micrograms/m3 permissible exposure limit for limits, the worker must be removed from lead ex- workplace air concentration. The exceptions in- posure until the level drops to an acceptable value. clude primary and secondary lead smelting and For up to 18 months, the employer must main- lead-acid battery manufacture, where controls are tain the worker’s wages and seniority status even most difficult and economic conditions have been if the person cannot perform his or her regular unfavorable. (Primary smelters purify lead from job. lead concentrate, which is lead ore enriched by milling. Secondary smelters recover lead from OSHA requires that air lead levels be reduced discarded lead-containing products—in particu- to an effective concentration of 50 micrograms/m 3. lar, worn-out batteries. Battery plants make lead- Since reported exposures have ranged above 2,000 acid batteries. ) Although the standard was con- micrograms/m 3, reaching the regulatory limit tested by both union and management and it is poses many problems for employers. The regu- impossible to be certain of the future of these in- lation gives companies 3 to 10 years to attain the dustries or of the burdens placed on them by the so micrograms/m3 limit through engineering con- standard, it is clear that workers’ health has been trols; in the meantime, employers can require the

Ch. 5—Technologies for Controlling Work-Related Illness ● 91 natives to chipping and grinding in foundries. and other controls are available, remains a prob- Vacuum cleaning may be substituted for dusty lem. Hickey, et al. (210) note some possible compressed air cleaning and screw conveyors used reasons: instead of dust-producing pneumatic conveyors. ● the current OSHA standard is inadequate However, care must be taken to assure that such and based on outdated information, treatment, while suppressing visible dust, also ● compliance with the inadequate standard is controls the smaller, more hazardous, respirable insufficiently monitored, silica dust particles. ● accurately measuring silica concentrations in Where silica remains in use and worker expo- respirable dust samples is difficult and costly, sure is possible, local exhaust ventilation may be and used to capture and carry dust away. Environ- ● there is too much reliance on after-the-fact mental Protection Agency or other ambient-air control methods that control the dust after standard regulations may require that ventilated it is generated rather than on methods that air be cleaned before discharge to the outside. eliminate silica dust. An underlying reason for failure of worker pro- Control Technologies: Personal tection against silicosis is the cost of controlling Protection and Administrative Controls exposures. Respiratory protection and face, eye, and body To attack this problem, Hickey, et al. (210) sug- protection against physical injury are also re- gest promulgating a protective standard based on quired by OSHA in specific regulations for abra- the latest medical knowledge and streamlining en- sive blasting. NIOSH has specified the respirator forcement by developing an accurate, inexpensive, types required for protection from various air con- and rapid measurement method. These initial centrations of silica, but these often prove to be steps will provide the basis for developing more inadequate in practice (210). Employer-provided effective technology to prevent generation of silica and -maintained protective clothing and facilities dust. Greater emphasis should be placed on pre- for changing at work plus training about personal venting generation than on refinement of meas- hygiene prevent exposed workers from exposing ures for control after the dust is generated. Re- family members to silica dust when taking work search should be conducted to find nontoxic clothing home. NIOSH (and others) recommend: administrative measures that help reduce risk of silicosis; training managers and workers about the hazards of silica dust; the effective use of personal protection equipment; and work practices that prevent the generation of silica dust, Dust-reducing practices include vacuum cleaning, regular maintenance of dust-producing and dust-controlling systems, and good housekeeping. Dusty work may be scheduled or located to reduce the number of workers at risk. However, Hickey, et al. (210) report that company dust-control policies are often unenforced.

Strategies for Silica Dust Control

One might ask why a well-recognized, entirely Photo credit OSHA, Office of Information and Consumer Affairs preventable, work-related illness, for which the Abrasive blasting workers are frequently exposed to etiology is understood and for which engineering high levels of silica dust ——

Ch. 5—Technologies for Controlling Work-Related Illness . 93 use of respirators to reduce workers’ exposures battery manufacture that reduce worker exposure to airborne lead. to lead. A major source of lead exposure here has been the breaking open of old lead storage bat- Control Methods: Engineering teries. Goble, et al. (184) mention two new proc- and Respirators esses that significantly reduce the liberation of lead in that process. In addition, technological changes Table 5-6 lists categories of control measures recently introduced in the manufacture of new that can be employed to reduce lead exposures. lead storage batteries reduce worker exposure In general, major changes in processes will be in- while increasing productivity. troduced only when a plant is rebuilt for other Table 5-6 includes personal protective equip- reasons. (An example of the costs involved in sub- ment as well as business cycle factors that influ- stituting a new process in primary smelters com- ence the number of workers exposed. The role of pared with adding on controls is presented inch. respirators in providing protection until engineer- 16.) Add-on controls, in particular better ventila- ing controls are installed is clearly recognized in tion, are probably the most common form of engi- the OSHA standard. The standard does require neering controls, although far simpler controls— that ultimately compliance shall be achieved such as covering stockpiles and putting tops on through the use of feasible engineering controls. reaction vessels—are an important part of engineering controls. Medical Removal Protection A number of process innovations are being made in the secondary smelting industry and in The OSHA lead standard provides that when the amount of lead in a worker’s blood exceeds a trigger level, he or she is to be removed from Table 5-6.—Measures To Reduce Air Lead and exposure or placed in an area of lower exposure Blood Lead Levels until the blood lead level drops (see table 5-7). A. Measures that affect air lead levels in the plant When the amount falls to a specified reinstatement 1. Changes in production processes (direct smelting level, the worker can return to his or her regular processes, more automated battery production lines) job. 2. Add-on controls (ventilation systems) 3. Changes in operating practice (keeping floors When the OSHA standard was being consid- cleaner) ered, employers pointed at MRP as a source of 4. Greater or lesser use of lead-emitting equipment high costs. They argued that older, more experi- B. Measures that do not affect air lead levels but limit times workers spend in lead-contaminated at- enced workers who were paid a premium for their mospheres knowledge would be removed to less skilled jobs, 1. Isolation booths with filtered air supply causing losses in productivity. In addition, since 2. Changes in work practices to limit time in high lead areas MRP requires that the worker’s wages be main- c. Measures that do not affect air lead levels but limit workers’ lead absorption 1. Respirators Table 5-7.—Blood Lead Levels That Trigger 2. Showers, changing clothes before and after enter- Medical Removal From and Return to ing work areas Lead-Contaminated Atmospheres 3. Business cycle factors: layoffs, overtime D. Measures that do not necessarily affect exposure of Blood lead Ievelsa the work force as a whole but affect the distribution (micrograms/100g blood) for of exposures among the work force Date Removal Return 1. Monitoring of workers and removing those with biological indicators of exposure to areas with March 1979 ...... 80 60 lower lead contamination March 1980 ...... 70 50 2. Rotation of workers September 1981b . . . ~ . . 60 40 3. Firing of highly exposed workers March 1983b ...... 50 40 E. External measures that impact on lead exposure %Vorkers’ blood levels are to be monitored quarterly except workers with 10WJIS 1. Changes of lead level in out-of-plant environment greater than 40 micrograms/100g are to be monitored monthly. bMany firmS have been given extensions of the time fOr the 60/40 and 50/40 2. Changes of lead content in food and water trlgffers SOURCE (164) SOURCE (164) 94 Preventing Illness and Injury In the Workplace tained, experienced workers doing less skilled jobs well with achievement of 100 micro~grams/m3 air would still receive the pay associated with their lead levels, supporting the conclusion that effec- previous positions. tive air lead levels are between 50 and 100 micrograms/m 3. Given that blood lead levels are related Table 5-8 summarizes three years’ data about to worker health, these changes are evidence that medical removal from companies seeking relief lead-related diseases and disorders should be from the lead standard. These data represent a declining. worst-case group and may not be representative of the industry. In both the primary smelter and The number of terminations of workers because the battery industries reported, the percentage of blood lead levels remained above the reinstate- workers on MRP transfer and the share of work- ment values even after removal to lower exposure time spent on transfer peaked in the second year. situations is apparently small. An examination of The data for primary smelters is reasonably com- the new-hire and termination rates before and plete, based on 5 of 7 smelters and about 2,120 after imposition of the OSHA lead standard did workers each year, compared with a total of about not show an increase. That observation is incon- 2,500 workers; it is less complete for the battery sistent with the idea that employers would ter- industry, based on only 8 plants and about 1,300 minate “leaded-up” workers and replace them workers in an industry that employs about 30,000 with new hires. people. In the secondary smelting industry, the percentage of workers on MRP and the proportion of worktime on MRP transfer increased each Changes in Air Lead Levels year. The data in this case are certainly incom- Although some data about air arid blood lead plete, and the facilities reported may not be rep- levels are available, they are often unsuitable for resentative of the entire industry; the data in table making precise estimates of levels, of high ex- 5-8 are based on about 640 workers out of a total posure. For instance, although 67 percent of sec- of some 3,000 workers in the industry. If the data ondary smelter workers in 1977 were exposed to are representative, the secondary smelters are en- 3 greater than 200 micrograms/m airborne lead, countering greater problems complying with the neither the maximum exposure level nor the aver- OSHA standard. age exposures of the highly exposed workers in Goble, et al. (184) compared the percentage of this group were reported. Goble, et al. (184) made total worktime on MRP transfer to projections of a number of assumptions and then calculated ap- transfers that had been made based on assump- proximate average air lead exposure levels in the tions of so or 100 micrograms/m3 air lead levels three industries in 1977-78 and in 1981-82 (see in the industries. They found that the reported table s-9). Air lead levels dropped by about one- percentages of transfer worktime agree reasonably quarter in primary and secondary smelting and

Table 5-8.—Medical Removal Protection Transfers in a Sample of Lead Industry Plants

Average number per plant Percent Plants Lead exposed Workers on worktime Industry Year In survey In industry workers MRPa transfer on MRP 1979 5 7 465 21 1.0 Primary lead smelting ...... 1980 5 7 419 31 2.1 1981 5 7 492 18 1.3 1979 6 36 120 4 1.0 Secondary lead smelting. . . . . 1980 6 36 104 9 4.6 1981 6 36 96 11 6.9 1979 8 136 176 2 0.4 Battery manufacture...... 1980 8 136 140 8 1.9 1981 8 136 162 6 1.5 aMflP, M@c~ Removal Protactlon. SOURCE: (1S4 from data available in 103). —

Ch. 5—Technologies for Controlling Work-Related Illness ● 95

Table 5-9.— Reductions in Average Air Lead Levels, River Associates (103) report prepared for OSHA, 1977-78 and 1981-82 and those are probably more reliable. Average air A satisfying drop in blood lead levels was seen lead levels in less than 5 years between 1977 and 1982. Not (micrograms/m’) Percent Industry 1977-78 1981-82 reduction shown on the table is the finding that the num- Primary lead smelting. . . . . 740 565 24 ber of workers with blood lead levels greater than Secondary lead smelting . . 285 205 28 80 micrograms/100g blood dropped from l,553 Battery manufacture . . . 160 80 50 (2 percent of 2,200 primary smelter workers plus Seven battery plants ...... 160 90 50 16 percent of 3,170 secondary smelter workers SOURCE (184) plus 6 percent of 16,700 battery workers) to about by half in battery plants. Confidence about the 20 (0.1 percent of 2,470 primary smelter workers validity of these estimates, especially for battery plus 0.6 percent of 3,000 secondary smelter work- plants, is increased by the access Goble, et al. had ers and no battery workers). to detailed, company-collected exposure data from seven battery plants. The percentage reduc- Furthermore, the number of workers with tion observed in those plants is the same as the blood lead levels above 40 micrograms/100g calculated reduction for the industry overall. dropped from 17,217 to 6,738. This significant decrease is especially important because that is the The data in table 5-9 show what are probably lowest action level required at any stage of MRP. minimal estimates of reductions in air lead levels In other words, the almost 9,000 workers who because of systematic errors in the calculations. have moved from the over-40 to under-40 micro- Clearly, however, levels are coming down. Equal- grams/100g category are now at a level that ly clearly, there is some distance to go before the means they would not have to be removed from 3 eventual goal of 50 micrograms/m is reached. their current jobs even as the threshold level for OSHA recognized that engineering control of air medical removal drops. lead levels would take time, up to 10 years in some industries. The decreases shown in table s-9 were In 1978, OSHA had estimates prepared of the achieved in less than 5 years and during the period blood lead levels to be expected if the statutory limits for lead were set and realized at 50, 100, when the standard was still being challenged in 3 the courts. or 200 micrograms/m . The levels were expected to fall as exposures decreased and workers elimi- Changes in Blood Lead Levels nated some of the lead accumulated during their previous high exposures. Data on blood lead levels for the period before Measured blood lead levels two-and-a-half promulgation of the lead standard are not so plen- years after the introduction of the standard were tiful as air lead data. The estimates shown in table consistent with projections made on the basis of 5-10 for 1977-78 are from information presented achieving a level near so micrograms/m3 in the in OSHA hearings. The data shown for 1981-82 battery industry and 100 micrograms/m3 in the are from measurements reported in a Charles other two industries (184). These measurements are somewhat surprising because the air lead levels Table 5-10.—Average Blood Levels Before and After Promulgation of the OSHA Lead Standard in the industries are above so or 100 micrograms/m3. Effective respirator programs and at- Approximate average tention to personal hygiene have probably con- blood lead levels (micrograms/100g blood) tributed to the lowering of blood lead levels. Industry 1977-78 1981-82 Difference Although no blood lead level has been estab- Primary lead smelting. . . . . 49.4 41.6 7.8 lished below which symptoms are never found, Secondary lead smelting . . 56.5 44.2 12.3 Battery manufacture ...... 53.2 42.4 10.8 and there is no level at which symptoms will nec- Seven battery plants ...... 53.0 38.3 14.7 essarily occur, there is agreement that lower blood SOURCE (184) lead levels are associated with lower risks (174). 96 ● Preventing Illness and Injury in the Workplace

OSHA has established 40 micrograms/100g as an micrograms/m3. The costs are quite close. One action level; when the lead standard is fully im- reason is that (according to engineers employed plemented, workers with blood levels above 50 by Charles River Associates (184)) the best con- micrograms/100g must be removed from lead ex- ventional engineering controls will not reduce exposure until their blood lead levels drop below posure to 150 micrograms/m3. Another reason is 40. The Centers for Disease Control (558) have that isolation booths, if installed, could reduce ex- concentrated on 30 micrograms/100g as a level posures to less than 50 micrograms/3 for about at which concern should be raised. the same as it would cost to reach 150 micrograms/m 3. costs Major process changes, although costing more Capital expenditures for current controls run in capital expenditures, are expected to result in at about $1,000 to $1,500 per worker each year. operating savings. In general, the capital costs of To that must be added the expense of respirators, process change may be appropriate if a new plant clothing, and facilities for personal hygiene is to be built, but they outweigh the costs of add- (showers, changing rooms, etc.)–between $1,000 ons in an existing plant unless significant tax sav- and $1,700 per worker per year. Monitoring and ings or credits accompany installation of the new medical surveillance cost about $500 per worker process. annually, and the tranfer costs under MRP are expected to run between $300 and $600 per work- Summary of Improvements er yearly. Taken altogether, complying with the The data about workplace air lead and blood lead standard is estimated by Goble, et al. to cost lead levels show that both have decreased since between $2,800 and $4,300 per worker yearly. the issuance of the OSHA lead standard. While In addition to the current costs, Goble, et al. the air lead levels have dropped about 25 percent (184) project that future conventional industrial in primary and secondary smelters and about so hygiene controls will cost between $8,000 and percent in battery plants, they still remain much $9,000 per worker per year in secondary smelters higher than 50 micrograms/m3 that is the goal of and battery plants. Future costs in primary smelt- the standard. At the same time, however, blood ers are expected to be lower, about $5,200. lead levels have dropped appreciably, and in general are close to the levels predicted for reaching Table 5-11 presents estimates of the engineer- air lead levels between 50 and 100 micrograms/m3. ing cost of reducing air lead levels to 50 or 150 A number of factors—including decreases in lead uptake from the environment in general, changes Table 5-1 I.—Projected lndustry-Wide Annual Costs of in the methods for measuring lead, errors in the Compliance With Air Lead Levels of 50 and 150 micrograms/m3 model that is used to project blood leads based on air leads, and greater-than-expected impacts Millions of 1962 dollars of respirator programs and hygiene practices— Industry 50 150 have contributed to the apparent better realiza- micrograms/m 3 micrograms/m 3 tion of reductions in blood levels than was pre- Primary lead smelting . . 15.5 16.0 dicted. Whatever combination of factors is respon- Secondary lead smelting ...... 24.5 26.4 sible, the falls in blood lead levels are gratifying Battery manufacture . . . 97.4 not done and bode well for better health among lead SOURCE: (1S4). workers. Ch. 5—Technologies for Controlling Work-Related Illness ● 97

EXTENT OF CONTROL TECHNOLOGY USAGE IN THE UNITED STATES

The National Occupational Exposure Survey With the control questions asked in NOES it (NOES) (see chs. 2 and 12) includes data that de- is possible to analyze the extent of engineering scribe the extent of the usage of control technol- control usage in the manufacturing sector of the ogies for the prevention of work-related illness. country. Areas include practices of material sub- NOES, conducted from 1980-82, estimates the ex- stitution, process change, and the management tent of worker exposure to potentially hazardous of personal protective equipment programs. These workplace agents. This survey was conducted as data are unique in that there are no other com- a followup to a similar survey, the National Oc- prehensive assessments of work-related exposure cupational Hazard Survey, conducted in 1972-74. control practice. Control technology usage may be classified by plant size and by industry, allow- The sample of businesses in the NOES survey ing distributions to be done for comparison. consists of approximately 4,000 establishments in 67 metropolitan areas throughout the United These data may be used to pinpoint patterns States. The sample represents all nonagricultural of control technology use within and among in- businesses covered under the Occupational Safety dustry groupings, giving insight about areas where and Health Act. Data were collected onsite by improvement is needed. This analysis may also teams of engineers and industrial hygienists spe- be used to assist in setting priorities for control cially trained for the survey. technology research. NOES was conceived for the purpose of recording specific worker exposures to potential work- Information About Controls and place health hazards. Among the questions that Areas for Research the survey attempted to answer were: The vinyl chloride, industrial solvent, lead, cot- ● What occupational groups are exposed to ton dust, and silica examples show that control what types of potential health hazards in the technologies for workplace exposures can be engi- United States? neered once commitment to control is made. ● In what types of industries are these hazards Commitment, however, is often difficult to found? achieve. For example, in the regulatory proceed- ● What control technologies are present to pre- ings concerning new health standards, arguments vent work-related disease in terms of plant are often raised about the harmful health effects operation and occupational safety and health of existing exposure levels, and the costs and fea- practice? sibility of controls (see ch. 14 and box 12-1 in ch. ● What are the exposures by intensity, dura- 12). In addition, opposition to some governmental tion, type of control? regulation may result simply from employers’ con- ● What trade name products were present? cern that an outside authority is telling them what they must do to protect workers. Both surveys included questions about demog- raphy and occupational safety and health prac- However, as shown by the vinyl chloride and tice, followed by a walk-through survey of the cotton dust examples, the installation of technol- plant work area to inventory potential exposures. ogies to control workplace hazards can be accom- A series of questions specifically aimed at the prac- panied by greater productivity. As seen in the case tice of using controls was asked in NOES. of the ventilation control system in the electronics 98 . Preventing Illness and Injury in the Workplace industry, there are advantages to planning, install- There is also a dearth of new approaches in this ing, and maintaining control technologies in a sys- area. For instance, First (173) pointed out that little tematic way. Anticipation of work-related health has been added to the theory of ventilation since problems very often reduces the cost of their two Ph.D. theses done at Harvard in the 1930s. control. The tendency has been to retrofit control solutions after problems appear rather than to an- Access to information about control technol- ticipate them. Yet there is promise of new meth- ogies for workplace health and safety could be ods on the horizon. improved. Perhaps the greatest current need is for published information about controls in the occu- Brief and colleagues (74), recognizing the limita- pational safety and health literature. While there tions of retrofit solutions in preventing work- are journals dedicated to toxicology and epidemi- related injury and illness, have explored tech- ology, there are none specific to industrial hygiene niques for designing new plants with new control engineering. Industrial hygiene journals infre- systems built in. They have found that in the past quently and engineering technical journals only rarely include articles about technologies for con- retrofit control procedures were recommended trolling worker exposure to hazardous materials. without being able to judge the effectiveness of Yet it has been suggested that such information controls, until after installation and operation. should be part of every engineer’s training and This retrofit approach is probably not as cost ef- be readily available as reference material to the fective as designed-in controls, although cost effectiveness was rarely tested. In many cases ad- practicing engineer (587). ditional administrative and personal protective Published information about specifics of work- programs were used to achieve desired worker place control is sparse for several reasons. First, protection. and probably most significantly, companies that We have embarked on a new era involving develop controls simply do not take the time to some major companies and government agencies investigating the impact of engineering design on publish details since it is not their business. On the workplace environment. The objective is sim- the other hand, it is likely that some consider the ple. It states that we will attempt to design into information proprietary and keep it unpublished our plants and operating facilities the necessary for competitive reasons. engineered controls to meet occupational health In some cases, such as for the control of expo- standards. Intuitively, we believe that it is more cost-effective to install engineering controls in new sures to vinyl chloride, a few companies market plant designs than to retrofit later. Equally as im- new technology for preventing work-related in- portant is the practicality of having an environ- jury and illness. This, however, appears to be in- mentally sound plant at the start, rather than one frequent and be limited to very large companies which requires modifications later. Retrofitting . such as B.F. Goodrich and Dupont. Probably controls may be difficult to implement due to most companies that have found and use inno- physical factors and the time to implement the vative control technologies in their plants simply changes after the plant is running. have yet to explore workplace control technologies as a market. In this innovative approach, design is based on selection of process equipment controls appropri- University and government researchers have ate to the process. The key is to determine emis- published some practical information that can be sion rates of contaminants from each type of proc- used by design engineers but the volume of this ess equipment used. These data may then be used material is limited. One widely used handbook to build a near-field dispersion model (a mathe- specific to ventilation is the ACGIH Ventilation matical expression of the release and buildup of Manual that is published annually (6). Programs contaminant in workers’ breathing zones) to cal- such as the NIOSH Control Technology Assess- culate collective concentrations to which work- ments have produced useful information for haz- ers could be exposed. By trying various combina- ard control in some specific and some generic tions of equipment and controls in the model and manufacturing processes (see ch. 12). testing them against recommended health stand- Ch. 5—Technologies for Controlling Work-Related Illness 99 ards, engineers can predict potential worker ex- plied, particularly in plant design stages. Technol- posure and thus design processes with optimum ogies are available but information about specific worker protection and production. These in- solutions is difficult to find because it is seldom vestigators stress the need for interaction between published. Retrofit is the dominant mode even engineers and occupational safety and health pro- though there is recognition that solutions should fessionals at the design stage for this to succeed. be designed into new processes. Thus, control technology for work-related illness prevention is possible but insufficiently ap-

SUMMARY

Workplace exposures to toxic substances can Data about workplace air lead and blood lead be controlled at their source, during transmission, levels show that both have decreased since the is- and at the worker. Control at the source includes suance of the revised OSHA lead standard in changes in the design of a process and substitu- 1978. The possible factors to explain the improve- tion of nontoxic or less toxic materials. Control- ments in blood lead levels include changes in ex- ling the transmission of a toxic substance can be posures to lead in the workplace air, the use of done by isolating or enclosing hazard sources, medical removal protection, decreases in the wetting toxic dusts to prevent dispersion, install- amount of lead absorbed from the environment, ing local exhaust ventilation to capture and carry changes in lead measuring methods, and improve- toxic substances away, or reducing toxic concen- ments in respirator programs and hygiene practices. tration through the use of general dilution ventilation. Control at the worker includes the use of Silicosis—a disabling lung disease-is caused personal protective equipment (see ch. 8), work by silica dust. Control measures include substi- practices, and administrative procedures. Engi- tution with safe abrasives, ventilation, wetting, neering controls that can be designed into a work as well as the use of respirators, work practices, process to control hazard sources and dispersion maintenance of ventilation systems, and good of contaminants are preferred to other measures housekeeping practices. that may provide less reliable protection. Train- A considerable amount of information about ing (see ch. 10) of supervisors and workers is re- how to design and implement control technology quired to make sure control programs are ef- for worker protection is available but is not widely fective. disseminated. Research on improved control tech- Three case studies prepared for this assessment nology design and implementation is also needed. provide information on controls for health haz- For example, little has been added to the basic ards. Exposures to cotton dust cause a debilitating theory about ventilation since the 1930s. The Na- respiratory disease known as byssinosis. In the tional Occupational Exposure Survey conducted years following the issuance of a revised OSHA by NIOSH collected information which will give health standard concerning cotton dust, the U.S. estimates of the extent of worker exposure to po- textile industry has invested heavily in moderniz- tential hazards and the current practice of con- zing its operations. The new equipment has led to trol technology use. These data can potentially improved productivity in this industry, as well assist in setting priorities for research on improved as reduced worker exposures. controls. 6.

Technologies for Controlling Work-Related Injuries Contents

Page Preventing Motor Vehicle-Related Injuries and Fatalities ...... 106 . Preventing Construction-Related Injuries ...... 106 . Preventing Manufacturing-Related Injuries ...... 108 . . Woodworking Processes ...... 108 . Metalworking--Cutting, Welding, and Cold Forming...... 109 Metalworking-Hot Working...... 111 Preventive Maintenance...... 9... 112 Fire and Explosion Prevention ...... 113 . Injury Control Programs Industry ...... 116 The DuPont Injury Control Program ...... 117 summary .. ● ● ...... ●. ●. ●. . ●. ●. ●. ●...... ●.. ●. ●. ●. . . . ●. . . ● ● . . ●. ● ● 118 LIST OF TABLES Table No. Page 6-1. Principles of Preventing Work-Related Injury ...... 104 6-2. Prevention Checklist To Be Used Before Starting a Production Plant . . . . 105 6-3. Examples of Injury Control in Plant Layout .. .. a s ...... 105 6-4. Technologies for Preventing Work-Related Injury From Woodworking Machinery ...... 109 6-5. Technologies for Preventing Work-Related Injury From Metalworking Machinery ...... 111 6-6. Technologies fo r Preventing Work-Related Injury During Work with

Hot Metal in Foundries . . . . . ● ...... 112 6-7. Technologies for Preventing Work-Related Injury Through Maintenance of Structures ...... 113

6-8. common Ignition sources of Industrial Fires ● ● ● . ● . .... ● .. ● ● ● ● ● . 114 6-9. National Fire Protection Association Fire Classification ...... 116 6-10. Du Pont Calculations of Cost Savings From InjuryControl ...... 118

Technologies for Controlling Work-Related Injury

Injuries are caused by “abnormal energy trans- from management. The stronger the commitment fers or interferences with energy transfer” (198). at the top, the greater the likelihood of success. One analytical method breaks the injury-causing The success of one such program is illustrated by event into three parts: 1) the source of hazard, a discussion of the injury prevention program of 2) its transmission, 3) and the worker; this method one large company—Du Pont. is patterned after the traditional public health Workplace injuries can be prevented by proper model of disease transmission (“agent,” “vector,” design of structures, machines, and operations (see and “host”). table 6-l). Proper design considers the stresses to Control technologies suggested by this ap- be placed on the building structure, the arrange- proach consider: control at the source of energy, ment of spaces for the work to be accomplished, control of transmission of energy, and control at and specific safety requirements. For example, the worker. Although there are many similarities falls on working surfaces, the single most com- between safety engineers’ approach to injury con- mon cause of workplace injuries (13.5 percent of trol and the public health approach, their ter- all nonfatal injuries), can be largely prevented by minology and methods have usually been quite designing proper walking and working surfaces. different. Safety engineers have tended to use Appropriate design, including safe construction codes, standards, and models of “good practice” plans, can prevent work-related injuries as build- that are oriented around particular topics: fire pre- ings go up and equipment is installed. Structure vention, electrical safety, design of machinery, collapse during concrete pouring may kill or in- plant layout, etc. jure a number of workers at once. Such catastro- This chapter describes how designers and engi- phes get media attention whereas the bulk of neers can plan sites, plant layout, and equipment work-related injuries and fatalities that occur design in order to prevent work-related injury. singly do not. One disaster at Willow Island, WV, “Safe” design presents particular difficulties in the construction industry where constantly changing conditions create constantly changing workplace hazards against which workers must be protected. In manufacturing, the worksite is relatively more stable but there is still a great deal of worker exposure to hazardous releases of energy. Fire and explosion prevention is an area that not only can prevent human deaths and injuries, but also can prevent very large economic losses. Probably for that reason, fire and explosion prevention has received a great deal of attention from the safety profession. Finally, this chapter discusses injury prevention

programs. Because management has the primary Photo credit: Office of Technology Assessment responsibility for prevention of work-related injury and illness, a successful injury prevention The rear of this construction site shows the use of program must start with a strong commitment shoring to prevent the collapse of surrounding soil

103 104 . Preventing Illness and Injury in the Workplace

Table 6-1.-Princlples of Preventing Work-Related Injury

Injury-prevention Relevant objective Examples control principle 1. To prevent the creation of the hazard. One-story buildings reducing need for ladders Elimination, substitution 2. To reduce the amount of hazard. Reducing speeds of vehicles. Process design 3. To prevent release of the hazard. Bolting or timbering mine roofs. Enclosure 4. To modify the rate or spatial Brakes, shutoff valves, reactor control rods. Ventilation distribution of release of the hazard. 5. To separate, in time or space, the Walkways over or around hazards, evacuation. Isolation Administrative hazard and that which is to be controls protected. 6. To separate the hazard from workers by Operator control booths. Isolation, Personal interposition of a material barrier. protective equipment 7. To modify relevant basic qualities of the Using breakaway roadside poles, making crib Process design hazard. slat spacing too narrow to strangle a child. 8, To make what is to be protected more Making structures more fire- and earthquake- Process design resistant to damage from the hazard. resistant. 9, To counter the damage already done. Rescuing the shipwrecked, reattaching severed NA limbs, extricating trapped miners. 10. To stabilize, repair, and rehabilitate the Posttraumatic cosmetic surgery, physical NA object of the damage. rehabilitation, rebuilding after fires and earthquakes. SOURCE: (71,197). in 1979 was found by the Occupational Safety and Information about these and other failures is Health Administration (OSHA) and National Bu- now available in a data base maintained by the reau of Standards investigators to be related to University of Maryland’s Architecture and Engi- concrete failure from improper pouring and from neering Performance Information Center. This insufficient allowance of curing time. The Na- computer compilation of analyses of design er- tional Institute for Occupational Safety and rors enables designers and engineers to search for Health (NIOSH) (584) is refining equations to pre- and compare information on failures of similar dict more accurately concrete’s curing time as an designs. aid to preventing similar disasters. Checklists have been developed for engineers, In 1981, 11 workers were killed and 23 injured architects, and designers to guide their attention in Cocoa Beach, FL, when a five-story building to methods of reducing injury risks. An abbrevi- collapsed during placement of a concrete roof ated example of such a list is given in table 6-2. slab. Analysis of this catastrophe showed two fac- Codes and standards for building structure, for tors caused the failure: steam, heating, and electrical systems, and for fire ● A design error: a check for punching shear and injury prevention are also used as design cri- (the stress or force around holes cut in beams) teria. The sources for these codes include the was omitted. American National Standards Institute, the Na- ● A construction error: supports for reinforc- tional Electrical Codes, the National Fire Protec- ing steel other than those-specified by the de- tion Association (NFPA) Codes, OSHA regula- sign were used and proved to be inadequate. tions, and recommendations made by NIOSH. A failure at one column precipitated a progres- Site planning and plant layout for location of sive failure of the slab, which, when it fell, caused buildings, facilities, and processes and other de- successive collapse of all lower floors (271). Trag- sign practices can be done in ways to prevent ically, this disaster closely resembled a collapse work-related injury. Table 6-3 provides examples of similar construction in Jackson, MI, in 1956. of injury control practices that are possible

Ch. 6–Technologies for Controlling Work-Related Injury 105

Table 6-2.—Prevention Checklist To Be Used Before Starting a Production Plant

Section l—Boiler and machinery review A. Boilers B. Pressure vessels C. Piping and valves D. Machinery Section n-Electrical safety review Section Ill—Fire protection review Section lV—Personnel safety review A. Project site location B. Building and structures C. Operating areas D. Yard Section V—Process safety review A. Materials B. Reactions C. Equipment Section Vi—Environment control audit A. Atmospheric discharges Photo credit: Department of Labor, Historical Office B. Liquid discharges Electrocutions account for about 6 percent of reported C. Solid discharges work-related fatalities Section VIl—Periodic plant loss prevention review (manufacturing) to: A. Keep operating personnel alerted to the hazards. through plant layout. The isolation of hazardous B. Determine whether operating procedures require materials and machinery, for instance, can be revision. achieved through building special rooms or build- C. Carefully screen the operation for changes that may have introduced new hazards, or changes that ings. Falls can be prevented by providing adequate should be made to reduce existing hazards. walkways and lighting. Again, specific codes are D. Reevaluate property and business interruption loss available to guide designers and planners in these exposures. E. Uncover potential hazards not previously recog- areas. nized, especially in the light of experience or new information. Many vehicular-related injuries and fatalities SOURCE (172) in and around factories maybe prevented through

Table 6-3.—Examples of Injury Control In Plant Layout

Source of injury Design solutions Contact with moving machinery Adequate space between and around machines Potentially explosive or inflammable processes Remote siting; separate buildings to contain explosion or fire; and substances isolated storage areas Crane loads striking worker Site cranes away from work areas Building fire Adequate space between buildings to prevent spread of fire Chemical burns Use corrosive-resistant containers, remote siting, and special handling procedures Falls from trestles Provide adequately wide walkways along trestles, and adequate lighting Falls in pits or bins Provide grating screens and covers, and adequate Iighting Falls on stairs and from ladders Design stairs with non-slip tread; provide adequate lighting Electrocution and electrical burns Adequate grounding; wiring inaccessible to inadvertent insulation wear Being caught in machinery during maintenance Interlocks and tag-out procedures to prevent inadvertent start-up Being caught in machinery during operation Adequate illumination Explosion related to broken light bulbs Special lamps and enclosures where physical conditions may shatter ordinary bulbs SOURCE (322) I&i ● Preventing Illness and Injury in the Workplace

careful planning of transportation facilities, in- reduce overexertion and back injury by provid- cluding shipping and receiving departments, railing working surfaces at correct height and allow- road sidings, parking lots, and roadways. Work- ing room for mechanical lifting devices. Shipping site roadways and walkways can be laid out and and receiving docks can be isolated to prevent designed according to safe engineering practices harm from mishap when loading or unloading in- to prevent traffic-related injuries. Adequate an- flammable, explosive, or extremely toxic sub- ticipation of traffic to, from, and during work and stances and from falling objects. landscape design to eliminate blind spots can also Sensible plant layout, including consideration reduce risk. Railings on stairs and walking inclines of headroom, aisle width, and access for mainte- prevent falls. nance can both lower injury risk and increase pro- Shipping and receiving, whether by truck, ductivity (322). Clearly, the risk of injury is re- train, ship, or airplane, pose special potential for duced by designing work stations that are located work-related injury. Facilities can be designed to away from hazardous areas.

PREVENTING MOTOR VEHICLE= RELATED INJURIES AND FATALITIES

Analysis of injury statistics shows that motor precautions. For example, braking systems on vehicle-related fatalities account for 30 to 40 per- earth-moving equipment, large trucks, and long cent of work-related fatalities and are among the distance tractor-trailers should be maintained with 5 leading causes of work-related injuries (see emphasis on safety rather than mere schedules. Working Paper #l). While a good deal of public Drivers should be properly trained to operate attention is given to “defensive driving, ” speed equipment safely under different road and weath- limits, collision protection through passive and er conditions. Vehicles used for employee trans- active restraints, and reducing the number of port should meet appropriate requirements for drunk drivers, little of the industrial injury-pre- both driver and passenger safety; for example, vention literature relates to the occupational use proper emergency exits should be available to of motor vehicles (40). allow escape. An insurance company has prepared a handbook that describes both routine and particular

PREVENTING CONSTRUCTION= RELATED INJURIES

The lost-workday rate from job-related injuries Seat belts or other restraining devices protect vehi- is much higher in the construction sector than the cle operators from harm during collisions. all-industry average. The constantly changing A frequently reported cause of injury is falls conditions of construction sites create a variety from heights—31 percent of construction fatali- of workplace hazards against which workers must ties. These may occur from inappropriate ladders, be protected. improperly erected scaffolds, poorly designed Specialized, often large, machines that come temporary stairs, or inadequately protected open- and go to construction sites bring their own me- ings in floors, elevator shafts, and roofs under chanical energy-related hazards. The “beep-beep- construction. Yet available equipment and pro- beep” of backing earth-moving equipment alerts cedures can protect against each of these hazards. workers on the ground; the cages around drivers’ Manufactured ladders that meet codes and lad- seats protect against falling and swinging objects ders constructed on the job that meet minimum and from being crushed if the machine tips over. requirements will reduce the number of falls. Rail-

Ch. 6—Technologies for Controlling Work-Related Injury ● 107

Photo credit Department of Labor. Hisforlcal Office

Earthmoving equipment is now built to include protection from falling objects and accidental rollovers

Photo credit: OSHA, Office of Information and Consumer Affairs

Falls from heights are a frequent source of fatal injuries in the construction industry

alive. This can be prevented by shoring trenches or by sloping trench sides as they are dug to prevent collapse. Photo credit OSHA, Office of Information and Consumer Affairs Scaffolding, railings, and blocked holes in floors Improperly shored trenches can be a source of serious under construction are day-to-day steps taken to injury when laying pipes and pipelines prevent injury. These temporary measures must be put up, taken down, and supervised each day ings that are sufficiently strong and high will pre- or as each floor in a new building is completed. vent a worker from slipping through openings Construction workers depend on personal pro- inadvertently. Adequate scaffolds have been de- tective equipment —hard hats, gloves, and steel- scribed for particular construction tasks; bricklay- toed shoes-for immediate protection but consid- ing and stonework require sturdier scaffolding eration of worksite layout to prevent workers than light carpentry. entering areas likely to be filled with flying ob- Another source of injuries and fatalities is jects plays an important role in preventing con- trenching. Collapsing trenches can bury workers struction-related injuries and death.

108 ● Preventing Illness and Injury in the Workplace

In all these cases, the persons responsible for safety must be given authority to require that precautions be taken. The urgency to finish a job and the false confidence that comes with “I’ve done it a thousand times this way” lead both workers and management to ignore hazards. These human tendencies are among the reasons that involvement of management from the top down is necessary for injury control.

Photo credit: OSHA, Office of Information and Consumer Affairs

Mechanical power presses are an important safety hazard in many manufacturing plants

PREVENTING MANUFACTURING= RELATED INJURIES Though the rates are not as high as the con- be related to manufacturing and provide a sum- struction industry, the manufacturing sector still mary of control strategies. has injury and illness rates higher than the all- industry average (see ch. 4 and Working Paper #l). It also offers many opportunities for injury Woodworking Processes prevention because it is a relatively stable work Table 6-4 lists woodworking processes, possi- environment. Workers make products generally ble injuries, and preventive technologies. Devices using the same methods day after day, while man- used to cut, shape, and join wood—including a agement closely controls the entire operation. variety of power saws, planers, shapers, lathes, Manufacturing involves applying energy to and routers—are capable of causing serious in- wood, metals, or other materials to shape them jury, especially to the hand or eye. Preventing into usable products. Even handtools multiply such injuries depends on design of special guards, human muscle forces greatly; the cautionary in- on work practice, and on personal protective structions about hammers, screwdrivers, saws, equipment. For instance, a guard would prevent and snips delivered from parent to child or from contact with a power saw, a lock would prevent teacher to student often include gory descriptions the saw from being turned on while maintenance of injuries known first or second hand. Power work is in progress, and gloves and goggles would tools move faster, generate greater energy, and, protect the worker’s hands and eyes from flying hence, involve a greater degree of hazard. chips. The following selected examples of dangerous Many items of equipment can be purchased operations among the wide variety found in man- with built-in guards (322). There is general agree- ufacturing operations illustrate injuries that may ment that built-in guards are less expensive, more

Ch. 6—Technologies for Controlling Work-Related Injury 109

effective, and less likely to interfere with production than “bolt-ons.” Unfortunately, the extra cost—coming at a time managers are spending money on new machinery-may inhibit purchase of the safer equipment.

Metalworking— Cutting, Welding, and Cold Forming

The most casual sidewalk superintendent can appreciate the hazards inherent in welding and cutting metals. These involve high temperatures, hot metal, and intense visible and ultraviolet light. Table 6-5 shows a summary of the potential hazards of each process and gives examples of the types of control technologies that can be used to prevent injury. The hazards of welding and cutting are similar. Skin may be injured by infrared, visible, or ultraviolet radiation or by fire, hot metal parts, explosions, or electrical shock. Eye injury or burns may result from flying sparks, hot metals, or the immensely strong light emitted in the infrared, visible, and ultraviolet spectrum. Gases including oxygen, acetylene, hydrogen, and others, usually stored under pressure in special cylinders, may explode, or injury may result from physical handling of the cylinders. Shielding can protect ad- Photo credit OSHA, Office of lnformation and Consumer Affairs jacent workers, and face masks can protect the

Improper use of powered woodworking tools, such as welder (322). One type of welding—resistance this saw, can lead to serious injuries, welding— is usually done by a special machine including cuts and amputations that eliminates many sources of injury.

Table 6-4.—Technologies for Preventing Work-Related Injury From Woodworking Machinery

Process, Injury Examples of prevention device, or tool potential technology available Saws Cuts, amputation, eye damage, Kickback and hood guards, jigsa, operating methods, push projectile wounds, hearing loss sticks, maintenance goggles, hearing protectors Jointers, planers Hand and finger cuts Swing guards, hold down clamps, maintenance Shapers Hand and finger cuts, projectile Use of solid cutters instead of knives, maintenance, safety wounds collars, brakes, operating methods, holding jigsa, goggles Power-feed Hearing loss, projectile wounds Isolation, goggles, maintenance, hearing protectors planers Sanders Abrasions, projectile wounds Guards, goggles Lathes Eye and projectile wounds Goggles, face shields Routers Hand cuts Jigsa with handles device for mechanically holding a of work in the correct position while working on SOURCE (322)

110 ● Preventing Illness and Injury in the Workplace

Photo credit” OSHA, Off/cc of Information and Consumer Affairs

The hazards of welding include heat, flying sparks, and fumes, as well as infrared, visible, and ultraviolet radiation

vices to activate power presses could increase productivity as compared with the use of two-handed Photo credit Department of Labor, Historical Office switch activators. Although photoelectric devices for these purposes are ordinarily prohibited by Lack of attention to prevention can result in equipment collapse an OSHA regulation, a variance was granted to one company to evaluate the relative degree of injury control among workers using self-tripping Cold forming involves applying great cutting devices and those using two-handed switches. In- and punching forces to metal. The power of the vestigations found no significant difference in machines for turning, boring, milling, planing, observed injuries or stress as measured by heart grinding, and power pressing is a obvious haz- rate, blood pressure, and subjective responses to ard. Table 6-5 also includes a summary of the a questionnaire among the two groups of work- types of injuries associated with these machines. ers tested (588). Cuts, eye injuries, injuries from being caught in Although the photoelectric devices provide pro- the machine, and foot injuries from dropping tection equal to traditional methods, and at the heavy chucks (devices that hold the cutting tool same time increase productivity, NIOSH studies in place during machine work) are the main prob- have found other innovations less positive. For lems. Control technology ranges from careful de- instance, presence-sensing devices that halt ma- sign that considers safety to personal protective chine operation when a worker’s hand interrupts equipment. a radiofrequency field in the danger zone proved OSHA and NIOSH studied self-tripping power inadequate for reliable injury prevention. These presses that use presence-sensing devices such as studies have also shown that the current stand- photoelectric detectors and light beams to prevent ards for two-handed activator switches for power hands from entering the presses. Use of these de- presses may be inadequate. Under some condi- —

Ch, 6—Technologies for Controlling Work-Related Injury ● 111

Table 6-5.–Technologies for Preventing Work. Related injury From Metalworking Machinery

Process, Injury or illness Examples of prevention device, or tool potential technology available Welding and cutting Shock trauma; eye injury; back injury from Careful layout of equipment, process, and handling heavy gas cylinders; cuts from job; arc and spark-shielding; proper sharp edges on finished work; hearing loss insulation of electrical cable; goggles; welding hoods; welders’ helmets; gloves; ultraviolet-energy-resistant clothing Metalworking machinery; Eye injury (flying chips), scalping (hair Machine design, including shielding, proper turning machines; boring caught in machinery), foot injury from electrical insulation, local exhaust machines (drills, boring dropped chucks; hand, finger, or limb ventilation for grinding wheel discharge and mills); milling machines injury; skin irritation (cutting fluids), for discharge gases from EDMa machines, (saws, gear cutters, electrical shock (EDMa) provision of emergency stop buttons electrical discharge— Safe working practices, including keeping EDMa); planing machines tools sharp, never leaving machine unattended, using handtools instead of hands to work metal Grinding machines Inspection of grinding wheels for integrity and balance; good maintenance; shields to prevent flying chips; use of personal protective devices, including goggles, hearing protectors, closely fitted clothing, haircaps, safety shoes Power presses; hand- or Finger or hand amputation or injury; finger Enclosure of the die during operation; foot-operated presses; punctures; fatigue; eye injury, foot injury; adjustable, fixed, or interlock press barrier metal shears; press brakes lacerations guards; point-of-operation safe-guarding with movable barriers, two-handed tripping devices, presence-sensing devices; semiautomatic feed systems to avoid use of hands; electrical controls such as interlocks, emergency stop switches, and ground fault protection; apparatus for transferring dies; anti-repeat clutches and magnetic brakes to prevent inadvertent triggering of power press or shear; frequent machine inspection; careful maintenance; special handtools for feeding pieces into the machine a Mlll!ng by electric arc b sy~tem~ for Preventing (nadvertanf machrne operation during mafnfenance

SOURCE (322)

tions workers could activate the power press and duction equipment, to non-destructive testing of have time to reach into the press before the press finished products. Injury risk is high to the eye, ram completes its cycle (588). head, foot, trunk, and limbs.

Metalworking—Hot Working Injury prevention depends on proper design of the job, good work practices, and personal pro- Hot metal work is done primarily in foundries. tective equipment (table 6-6). Bums are prevented It is characterized by massive, often older, ma- by spark shielding, special clothing, and insula- chinery. The size of the operations and the weight tion of electrical cables. Electric shock prevention and heat of the materials being worked result in methods include shielding, guarding, good work many injuries. The processes range from handling practices, proper electrical insulation, and main- raw material—ore concentrates, coal, coke—to tenance. Eye protection is needed where sparks working with heating and shaping ovens and pro- and hot metal particles may fly. Fire and ultra-

112 ● Preventing Illness and Injury in the Workplace

violet-energy-resistant clothing is important for worker protection. Gloves are needed to protect hands from bums. Welders’ helmets are required for electric-arc welding.

Preventive Maintenance Careful attention to the plant, its operations, and workers is necessary to maintain controls. Table 6-7 shows a range of maintenance operations required to prevent injury in manufacturing operations. Maintenance of everything from foundations and footing to electric bulbs can prevent injuries.

Photo credit: Department of Labor, Historical OffIce

This machine will not operate unless both of the operator’s hands are on the control switches. The use of “two-handed controls” represents one method of preventing injuries associated with power presses and shears

Table 6-6.—Technologies for Preventing Work. Related Injury During Work with Hot Metal In Foundries

Process, Examples of prevention device, or tool lnjury potential technology available Handling of materials such Injury from electrical shock, heat, being Ladles equipped with automatic safety locks as sand, coke, and coal struck by or against, or being caught or brakes; devices to warn when ladles are between; explosions, vibration, noise; injury being moved; appropriately guarded to hand and foot, eye, trunk, limbs, or other conveyors for transport of sand and molds body parts; death from trauma or asphyxiation Cupolas, crucibles, ovens Mechanical devices for charging cupolas; blast gates to prevent injury from gas explosion Production equipment Venting molds to prevent explosions; two- hand controls for molding and core-blowing machine operators; dust-tight sandblast rooms; guards, two-handed tripping controls for forging hammers, presses, and upsetters; careful and frequent equipment inspection Personal protective equipment, as required, including hard hats, gloves, leather aprons, steel-toed shoes, metatarsal guards, safety glasses, goggles, flame-retardant clothing; devices to prevent doors and other movable parts from hitting workers; material-handling devices to eliminate heavy lifting and possible spills or splashes of solids and liquids SOURCE (322) —

Ch. 6–Technologies for Controlling Work-Related Injury ● 113

Table 6-7.—Technologies for Preventing Work-Related Injury Through Maintenance of Structures

Process, Injury or illness Examples of prevention device, or tool potential technology available Foundations Injuries and fatalities from building Check for and repair settling footings and collapse columns, cracked foundations Structural members, walls, Injuries and fatalities from plant building Check for and repair settling walls, defective floors, roofs, and canopies collapse columns, joists, beams, and girders; cracked building materials such as steel, concrete, wood; cracked walls and windows; sagging or rotted roofs and ceilings; rotted, sagging, cracked floors Stacks, tanks, and towers Injuries and fatalities from structural Place railing around edges of platforms; collapse; asphyxiation and acute poisoning check structural integrity; use proper from tank gases; injuries from explosions procedures for maintenance work during tank cleaning or repair Platforms and loading Injuries from slips and falls Use metal protectors on edges of concrete docks platforming; check surfaces for rotting, holes, or other hazards Sidewalks and driveways Injuries from slips, falls, or collision with Repair damaged sidewalks and motorways; moving vehicles install and maintain warning signs and markings Underground utility repair Injuries and fatalities related to sewer gas Test for gases and oxygen deficiency before and maintenance explosion; asphyxiation from oxygen attempting maintenance work in depletion; injuries or fatalities from underground tunnels and sewers; provide collapsed trenches; drowning or injury from contaminant-free ventilation during failure of underground pipelines maintenance work; slope trenches at angles to prevent collapse Lighting systems Cuts from broken bulbs; injury from electric Provide proper equipment for disposal of shock, fails, or poor illumination burned-out lamps; use properly designed ladders and platforms; clean and replace burned-out bulbs on a regular schedule Stairs and exits Injuries from slips and falls; injuries from Install and maintain adequate handrails, crowding in emergency situations illumination, and walking surfaces on stairs; keep passageways and stairs free for exit; install unobstructed exit doors for fire and emergencies Grounds maintenance Injury from grounds maintenance; cuts Maintain tools and equipment; provide from mower blades and snow throwers; eye guards for belts, chains, and other moving injury from frying objects; collapse from parts quick-acting pesticides SOURCE: (322)

FIRE AND EXPLOSION PREVENTION Fire and explosions levy an enormous cost on Building designers play an important role in fire industry and continue to cause many deaths and prevention. Engineers and architects can antici- injuries among workers and the general public. pate fire problems in the design of buildings and Of fire-related deaths, 75 percent are due to in- facilities and thereby protect workers and the pub- halation of smoke and toxic gases and the other lic from injury. A study of over 25,000 fires be- 25 percent to direct flame. Injury and death from tween 1968 to 1977 resulted in a listing of igni- fires can be prevented by proper design and con- tion sources ranging from the most common— struction of buildings, vigilant inspection, early electrical ignition and arson—to the least com- detection and warning, and appropriate control mon—lightning (table 6-8). These findings point methods. to areas where controls are most needed and on

114 ● Preventing Illness and Injury in the Workplace

Photo credit OSHA, Office of Information and Consumer Affairs

The hazards of hot metalworking are found in steel mills and foundries

Table 6-8.—Common Ignition Sources of Industrial Fires

Frequency Source (percent) Means of prevention Electrical ...... 22 Design and maintenance Incendiarism (arson)...... 10 Security Smoking ...... 9 Supervision, substitution Hot surfaces ...... 9 Design and maintenance Friction...... 7 Design and maintenance Overheated materials ...... 7 Work practice and control Cutting and welding ...... 7 Work practice Burner flames ...... 6 Design and maintenance Spontaneous ignition ...... 5 Housekeeping Exposure (fire from adjacent property) ...... 4 Design Combustion sparks ...... 3 Design Miscellaneous ...... 3 Design and other Mechanical sparks ...... 2 Housekeeping Molten substances ...... 2 Work practice and maintenance Static sparks ...... 2 Design (grounding) Chemical action ...... 1 Work practice Lightning ...... 1 Design (grounding) of that cigarettes to burn for longer times would reduce the number Of cigarette-caused that kill 3,000 Americans annually. SOURCE: (322).

Ch. 6–Technologies for Controlling Work-Related Injury 115

Photo credit OSHA Office of Information and Consumer Affairs

An explosion at this oil refinery killed 3 workers and injured 14 others which building codes and standards tend to con- controlled. Smoking should be prohibited in all centrate for prevention. but approved areas to prevent fire. Friction from improperly maintained machinery can cause fires. Local building codes and OSHA standards, These can occur in workplaces where organic dust generally based on National Fire Protection Asso- may accumulate, such as grain elevators, textile ciation Codes, provide the minimum basis for fire- mills, and woodworking mills. Open flame de- safe construction design. The NFPA Life Safety vices used in industry are frequent causes of fire, Code is an appropriate design guide for warning These include heating torches and welding and and exit requirements. cutting equipment. Many fires occur in plants Since electrically started fires are the most fre- where housekeeping is lax. Accumulations of oily quent type, appropriate sections of the OSHA rags, waste, and other combustible refuse can ig- standards and the National Electrical Code of the nite unless kept in air-tight noncombustible con- NFPA are to be followed for proper installation tainers. Compliance with local ordinances con- and maintenance of electrical equipment. Spark- cerning incineration, sewage, and other means of ing can occur when unwanted metal objects en- waste disposal provides a minimum of safety. ter machinery. Particular attention should be given to preventing heat-generating short circuits A designer, engineer, manager, or worker who and arcs, two of the leading causes of electrical knows the identity of combustible hazards can fires. Temporary and makeshift wiring when ex- choose the proper emergency equipment to deal isting electrical outlets are overextended to accom- with fires. Table 6-9 is the fire classification de- modate a growing office computer system, for ex- veloped by the National Fire Protection Associa- ample, is one potential source of fire. Improperly tion that relates to the decal identifying most fire insulated portable electric tools and extension extinguishers. Clearly, placing the appropriate ex- cords should be avoided. Proper grounding is es- tinguisher in the right place is necessary. sential for preventing electrical shock and fire- Giving early warning of fires, extinguishing a causing arcing. fire at early stages, and quickly clearing buildings Smoking, friction, and open flame devices are are also obviously essential to prevent injury. De- other sources of work-related fires that can be sign should include early warning and alarm sys- 116 ● Preventing Illness and Injury in the Workplace

Table 6.9.—National Fire Protection Association Fire Classification

Class Description Means of control A. Fires occurring in ordinary material such as Cooling and quenching by water to reduce heat is wood, paper, etc. recommended. Special dry chemicals may be used to control flame. B. Fires that occur in the vapor-air mixture over Since water tends to spread such fires, dry chemicals, carbon the surface of flammable liquids such as dioxide, foam, or halogenated agents are preferred. gasoline, oil, etc. c. Fires that occur in or near equipment using Non-conducting agents such as dry chemical or carbon dioxide electrical energy. must be used to extinguish to prevent electric shock. D. Fires that occur in combustible metals such as Since ordinary means of control such as water may increase magnesium, lithium, potassium, etc. the intensity of such fires, special equipment and materials

are required to control this tvpe.r– of fire. SOURCE (322)

terns. Both manual and automatic fire extinguish- produced from phenolic, urea, vinyl, and other ing systems should be strategically located to limit synthetic resins and powders used in the plastics fires in their early stages. Fire-safe stairwells, ade- industry also explode under certain conditions. quate aisle space, and appropriate stairwell width The Explosion Venting Guide of the NFPA pro- are essential for safe and rapid exit. Emergency vides specifications for buildings where the risk stairwell lighting should be provided. of explosion is high. The use of fire-resistant materials, the design Explosion prevention can be accomplished by and construction of building frameworks, ventila- inert gas systems that displace oxygen so that tion systems, concealed spaces, storage areas, ex- gases and dusts are incombustible and by ade- teriors, fuel sources, and building uses—all part quate ventilation, proper process operation, and of the designer’s functions—make a difference. vigilant maintenance. Gas lines and valves should Widespread use of masonry in the interiors of Eu- be inspected frequently for leaks. ropean dwellings is thought to be a factor in their Since three-fourths of fire-related deaths are superior fire record. Plastic interiors used in this from toxic fumes and since industrial fires may country, on the other hand, are the source of involve extremely toxic materials, it is important deadly toxic gases during fires. for firefighters to be able to identify chemical and Explosions are a special preventive concern of material inventories in establishments under their designers. Explosive atmospheres result from the jurisdiction. Several municipalities and States accumulation of organic dusts and from gases and have developed systems for this. The OSHA haz- vapors. The U.S. Department of Agriculture lists ard communication (labeling) standard may also 133 dusts by degree of explosibility (322). Dusts be useful for this purpose.

INJURY CONTROL PROGRAMS IN INDUSTRY

Effective control of work-related injury depends stronger the commitment at the top the greater in part on how well a company is organized to the likelihood of success. Typical management deal with it. Management has the primary respon- programs for preventing work-related injury (and sibility for prevention. Relevant programs include illness) include: training workers, supervisors, and managers to . establishing clearly stated company policies recognize workplace hazards and take steps to for prevention; control them (see ch. 10). • avoiding work-related injury and illness by Successful injury prevention must start with a planning ahead when designing new plant or strong commitment from management. The modifying existing processes; .

Ch. 6–Technologies for Controlling Work-Related Injury ● 117

● analyzing each hazardous operation to de- ● maintaining accurate and complete records termine the steps required to do the job with- of the causes of injuries and illnesses; out causing injury or illness; ● training workers, supervisors, and managers . identifying hazards through workplace sur- to identify hazards and prevent injury and veys and investigating the factors surround- illness that could result from them; and ing an event that caused or nearly caused in- ● placing and maintaining people in jobs suit- juries or illnesses; able to their physical status.

THE DU PONT INJURY CONTROL PROGRAM

Perhaps the best source of information about els of management is partly dependent on success how to run an injury-control program is avail- in preventing injury. Du Pont lists the following able from the successful companies. Determining as incentives for making injury control a manage- one best program is impossible, of course, for ment objective: hazards vary from industry to industry, plant to ● protect workers, plant, and department to department, making ● improve profits, comparisons of overall rates difficult. This sec- ● improve product quality, tion focuses on the Du Pont Company not because . improve productivity, its program is best, but because it is good enough . improve employee-management relations, that Du Pont sells it to other companies and much and information is available about it. Du Pont plants ● comply with OSHA regulations. have very low injury rates, which have dropped dramatically since 1912, falling even more than Du Pont finds that injury control increases prof- the all-industry average according to National its. Using National Safety Council statistics for Safety Council data. A key to the company’s suc- 1976, Du Pont estimated the average cost of each cess in preventing injuries at work is its commit- disabling injury to be $7,182. The company then ment in this area. Its program consists of four compared its estimated costs for 1977 (based on main parts: their rate of 0.2 disabling injuries per million hours worked) with the projected cost if the rate had . establishing injury prevention as a top man- been at the all-industry level of 10.87. Table 6- agement objective; 10 shows Du Pont’s estimated savings were greater ● establishing injury prevention as a line-man- than $15 million, based on the all-industry injury agement responsibility; rate; a more representative comparison, however, ● adopting an injury-control philosophy to would be with large chemical companies, which guide management action; and as a group have lower injury rates than the all- . establishing an organization for injury industry average. Nevertheless, Du Pont’s own control. comparisons were impressive enough to be fea- Du Pent’s success in preventing injuries may tured in a Fortune article about savings from pre- largely be attributed to strong commitment of top venting work-related injury (284). management. The founder of the company was Line-management emphasis on preventing in- personally committed to preventing injury among jury is another critical element of success. Respon- his workers, often family or friends, who were sibility for preventing injury is as much a condi- in the high-risk business of manufacturing gun- tion of employment in Du Pont as getting the job powder. done, High-level staff meetings often begin with Commitment to injury prevention still perme- reports of job injuries, and any fatality has first ates this Fortune 500 corporation from the top priority at any management meeting, including down. For example, performance rating at all lev- the presidents. 118 ● Preventing Illness and Injury in the Workplace

Table 6-10.–Du Pent Calculations of Cost Savings From Injury Control

Du Pont U.S.A. Totals if at all actual (1977) industry rate (1976) Disabling injuries/million worker-hours...... 0.20 10.87 Disabling injuries/year ...... 2,174 Average cost/injury...... $7,182 $7,182 Cost/year (total injuries) ...... $287,280 $15.613.668., SOURCE: (156)

Five basic beliefs underlie Du Pont’s program: tailed checklists of hazardous working conditions. The manager looks for injury potential associated ● all personal injuries can be prevented, with procedures, tools, equipment, orderliness, ● managers and supervisors must personally personal protective equipment, positions of peo- accept their responsibilities to prevent per- ple, and actions of people. A training unit divided sonal injuries, into eight parts is used for instruction and includes ● reasonable safeguards are possible for all con- pocket-size cards to record workplace observa- struction and operating exposures that may tions as practice between teaching units. result in injury, ● efficiency and economy are enhanced when The administrator of the course is key to its ef- personal injuries are prevented, and fectiveness. To show management commitment ● all employees must be trained to understand to the course and to motivate supervisors to be the advantages to them, as well as to the equally committed, the administrator must be company, of taking responsibility to prevent highly placed in the management structure. personal injuries. The desired outcome of the training, which Du Pont’s Safety Training Observation Pro- should take only 12 to 16 hours away from work, gram (STOP) uses programed self-study courses is supervisors who are able to recognize hazard- to enhance injury prevention. It is one of more ous conditions, to assign responsibility for cor- than 200 vocational courses developed for indus- recting it, and to ensure the elimination of the trial craft skill training. The objective of STOP hazard. The success of the program has been is to train line managers to notice workplace con- measured through reduced injury rates in firms ditions that might lead to an injury or illness. that have purchased and used the training program from Du Pont. STOP presents seven broad categories that are suggested for observation rather than having de-

SUMMARY

Perhaps because of the more obvious sources These codes and guides to good practice appear of injury as compared to illness, safety engineer- often to be based on the personal experience and ing appears to be more pragmatic than industrial judgment of the practitioners who happen to write hygiene. Engineers have, to some extent, directed safety textbooks or are members of the commit- their attention to sources of hazard such as elec- tees who write the codes. This field has relied on tricity, falls, and fires, and produced codes and experience and judgment instead of scientific anal- standards for good practice. ysis, systematic data collection, the application Ch, 6–Technologies for Controlling Work-Related Injury . 119 of epidemiologic techniques, and experimental re- include the proper planning of plant sites and the search. It is probably fair to say that up to the design of equipment, as well as the use of guards, present there has been little that might be called personal protective equipment, and appropriate a “science” of safety. In addition, as described in work practices. Adequate attention to preventive chapter 4 on the identification of injury hazards, maintenance, and the training of workers and the attempt to attribute most injuries to so-called managers, also play roles in injury prevention. unsafe acts has often diverted attention away from Successful injury control programs start with a the design of plant and equipment that can pre- strong commitment to injury prevention at all vent “unsafe acts” or minimize the adverse con- levels of management. sequences of “unsafe acts. ” However, it is unclear Controls are known for many hazards. How- what differences a more theory-based rather than ever, they are not always used because they were the pragmatic approach would have made. not built into a plant, are not available at a par- Several specific areas where injury control can ticular worksite, or are pushed aside to get the be applied were discussed in this chapter. A num- job done. All of these reasons for not using con- ber of different methods and technologies are trols emphasize the important role of management available for preventing injuries related to motor in providing controls and seeing that they are vehicles, construction activities, manufacturing used. processes, and fire and explosion hazards. These 7. Ergonomics and Human Factors .

139 ‘

Page. 127

137-

.’ 138 LIST OF FIGURES Figure I No. Page 7-1. Schematic Representation of a Man-Machine System ...... 1 . . . 124 7-2. Control Burner Arrangements of Simulated Stove Used in Experiment About Logical Arrangements ...... , ...... * 125 . 7-3. Major Nerves in the Arm and Hand ...... 7-4* .** 129 The Carpal Tunnel ...... ● ● ● ● ● ● ● ● ● ● ● ● ● ● + ...... , ...... ,, . ● . . 130 7-5. Flexion and Extension of the Wrist ...... ● . ● . . ● ● ● ,1 ● . . . 130 7-6. Job Analysis: Assembly Tasks ...... * ...... ● . . 132 7-7. Good and Bad Designs for Containers and Workbench- ...... 7-8, Good and Bad Designs for Powered Drivers...... 133 #.. 7-9. A Knife Designed to Reduce Cumulative Trauma Disorders in ‘ 134 Poultry Processing ...... 135 “ 7-1o. Potential Ergonomic Risk Factors Associated with VDT Design ... , . . . . . 137

. — — 7 Ergonomics and Human Factors

Neither workers, nor machines, nor workplaces humans” (250). The principle behind ergonomic exist apart from each other. To accomplish work, design is that the machine should fit the worker, all come together, and injuries and illnesses some- rather than forcing the worker to fit the machine. times result from their interactions. The study of To quote from one ergonomics guide (655): these interactions is called “ergonomics,” or “human factors engineering. ” Man’s physical limits for bending, stretching, and/or compressing are such that the machine The term ergonomics was coined in the United must be made to adapt to the man rather than the Kingdom after World War II to describe a dis- converse. Behavior characteristics are somewhat cipline created during the war. It had been noted more flexible. Man can adjust his sensory-motor that “bombs and bullets often missed their mark, behavior to some degree, and . . . can utilize planes crashed, friendly ships fired upon and alternate procedures and make up for certain sunk, and whales were depth charged” (471). In equipment inadequacies. However . . . as oper- response to this situation, research on designing ator load increases due to task complexity, fatigue may reduce operator reliability; system perform- military equipment to match more closely the ca- ance could degrade at a critical time . . . Work- pacities of the users began on both sides of the place layout should favor the man’s physical and Atlantic. Attention was thus devoted to the inter- behavioral capability in all cases in which a likely action between the human operators and their ma- error in human performance could affect . . . chines. safety . . , The designer cannot assume that personnel selection and training will be a panacea for One name for this new discipline, ergonomics, improper workplace considerations. is derived from two Greek words, ergo, meaning work, and nomos, meaning laws. Ergonomics is Figure 7-1 is a schematic representation of a the science devoted to understanding the laws or human-machine system. The person processes in- principles that govern the design of work systems. formation about the environment and acts on it A British professional group—the Ergonomics Re- by using the machine’s controls. The important search Society —was formed by the practitioners features of the machine include the controls used of this new discipline. An organization with simi- by the worker, the operations of the machine, and lar aims, the Human Factors Society, was founded the displays used to feed information back to the in the United States in 1957. The work of the worker. The task of the ergonomist is to analyze original members was termed “human factors how the worker obtains the information needed engineering, ” or engineering psychology. Whether to operate the machine, how that information is they are called human factors engineers or ergo- processed to reach a decision concerning the nomists, the scientists who practice this discipline appropriate way to control the machine, and what draw on a number of other disciplines, including worker actions are appropriate to control the ma- medicine, physiology, psychology, sociology, chine in a fashion that is safe and meets produc- engineering, and physics. tion criteria (380). Ergonomists are concerned with safety, effec- Other considerations, especially improving the tiveness, and efficiency wherever people are part productivity of workers, fit easily into the goals of a system (380). The discipline is “an applied of ergonomics. Opportunities exist for workplace science concerned with the design of facilities, designs that simultaneously improve production equipment, tools, and tasks that are compatible output and reduce the risks of injury. For instance, with the anatomical, physiological, biomechani- C. G. Drury reported that new handtools in the cal, perceptual, and. behavioral characteristics of component-assembly department of a large com-

123

124 . Preventing Illness and Injury in the Workplace puter company and changes in seating, lighting, and savings, and the number of rejected compo- and workbenches led to improved production and nents fell to half the previous level. The workers the elimination of repetitive motion injuries. The involved expressed increased satisfaction with costs of the redesign” were earned back 4.3 times their jobs (340). within one year through increased productivity

Figure 7-1.—Schematic Representation of a Human-Machine System

SOURCE (292)

CLASSIFICATIONS OF ERGONOMICS Ergonomists usually subdivide the field into in- tem failures (critical incidence analysis), detailed formation ergonomics and physical ergonomics. examination of the sequence of actions taken by Information ergonomics is concerned with the col- workers (task analysis), and analysis of situations lection, display, sensing, and processing of infor- that may arise from unprogrammed events or hu- mation. Physical ergonomics is concerned with man errors (contingency analysis) (380). Ergon- worker size, strength, capabilities for motion, and omists also make extensive use of anthropometric working posture. data concerning the physical dimensions and capabilities of the human population. In addition, Ergonomists use a number of techniques that the techniques of biomechanical analysis are used include the evaluation of the transmission of in- to measure expected physical stresses encountered formation between the machine and the worker by parts of the body while performing; work tasks. (link analysis), discovery and evaluation of sys-

Ch. 7—Ergonomics and Human Factors 125

Information Ergonomics sponses, and generally falls into three categories: spatial (the compatibility of physical features or Information ergonomics is generally concerned spatial arrangement for displays and controls), with what the worker senses in the workplace and movement (the direction of movement of displays how that information is processed by the worker, and controls), and conceptual (the associations The two major sources of sensory information are people hold concerning the meaning of signals, visual and auditory. such as in the United States the association of the Visual displays in the workplace include me- color green with “go”) (292). ter scales, control labels, warning signs, cathode- One example of an everyday problem in com- ray tube (CRT) displays, and printed text. To this patibility will perhaps clarify this notion. Figure list can be added video display terminals (VDTs), 7-2 presents four possible arrangements of the which can be CRT displays, liquid crystal dis- burners and burner-controls for a stove. For sev- plays, or other technologies. Three factors impor- eral of these patterns, the relationships between tant to the operator must be considered in design- burners and controls can be difficult to learn and ing a visual display: the size of the critical detail hard to remember because the arrangements lack in the display, its brightness, and the contrast of spatial compatibility. To examine this, research- the display against the background. There are a ers in the 1950s setup an experiment in which they number of techniques available to achieve these told a group of subjects to turn on specific burners. goals in design and to evaluate existing equipment. The subjects’ reaction times were measured and There are also many guidelines for special applica- the number of errors was noted. Design I, asso- tions, including the special needs of groups such as older workers (23). Workers also receive information through Figure 7-2.–Control Burner Arrangements of sound. These can include tones (bells, whistles, Simulated Stove Used in Experiment About Logical Arrangements beepers, etc. ) as well as speech. Designers and ergonomists often face the question of whether to provide information through visual or auditory means. In general, if the message is simple, short, or calls for immediate action, if the worker is already overburdened with visual messages, or if the workplace is too dark or too bright for visual displays, an auditory presentation is rec- I II ommended. If the message is complex, long, and I does not necessitate immediate action, or the I auditory system of the worker is already overburdened, or if the workplace is too noisy, a visual display may be preferred (23). Finally, information ergonomics is concerned with the processing of information by the worker and the design of the workplace, including the design of controls. Research that has measured the Ill Iv I abilities of humans to accurately process infor- I I mation has shown that people often will not be Number of errors in 1,200 trials: able to make quick, accurate responses in com- arrangement plex or unexpected situations, Second, the studies 10 II 76 have also shown that “compatibility” in display- Ill 116 control design is very important. Compatibility IV 129 refers to the relationships between stimuli and re- SOURCE: (292) 126 . Preventing Illness and Injury in the Workplace ciated with both the fewest errors (zero) and the The cherry-picker accident discussed in box C il- best reaction time, was considered to be the most lustrates the importance of controls layout. compatible (292). Physical Ergonomics Many workplaces have dozens of incompati- ble control configurations, which often lead to Physical ergonomics, concerned with the worker operator errors and subsequent serious injury (23). as a physical component of the work process, —.

Ch. 7—Ergonomics and Human Factors ● 127 often uses the techniques and equipment of an- a mechanical system using the fundamental laws thropometry and biomechanics. Anthropometry of Newtonian mechanics. The forces acting on the is the measurement of the physical dimensions of muscles, bones, joints, and spine can be deter- the human body, including both the structure of mined for the lifting of an object, for instance. the body in fixed positions and the extent of move- Through such analysis, it is possible to calculate ment possible, such as reach or lifting capacity. the size and direction of forces acting on the body. These can be compared with expected human tol- Biomechanics is the study of the mechanical erances to judge whether the activity in question operations of the human body. The muscles, will cause harm. bones, and connective tissue can be analyzed as

ERGONOMICS AND PREVENTION OF MUSCULOSKELETAL INJURIES

Ergonomic principles can be applied to prevent Table 7.1.—Occupations and Activities Associated both overt and cumulative traumas. An example With Carpal Tunnel Syndrome in the overt category is the risk of falling from Aircraft assembly Inspecting a ladder, which can be reduced by considering the Automobile assembly Meat processing sizes and mobility of people when deciding how Buffing Metal fabricating Coke making Musicians far apart to place a ladder’s rungs (250). Cumu- Electronic assembly Packaging lative traumas are not the result of single events Fabric cutting/sewing Postal workers or stresses; they stem from the repeated perform- Fruit packing Textile workers Gardening Tire and rubber workers ance of certain tasks. Back problems are by far Hay making Typing the most common cumulative trauma injuries. Waitressing Upholstering Evaluation and redesign of tasks to prevent back Housekeeping injuries is discussed later in this chapter. SOURCE (60) Repetitive motion disorders are a type of cumu- of the hand. Because the musculoskeletal strain lative trauma associated with repeated, often from repeatedly flexing the wrist or applying arm- forceful movements, usually involving the wrist wrist-finger force does not cause observable in- or elbow. Some 20 million workers on assembly juries, it often takes months or years for workers lines and in other jobs that require repetitive, to detect damage. strain-producing motions are at increased risk of The incidence and prevalence of CTS in the developing such disorders. Redesigning work sta- work force is not known. The National Institute tions, equipment, and handtools can significantly for Occupational Safety and Health (567) reports reduce the awkward, forceful movements com- that 15 to 20 percent of workers employed in con- mon to many jobs on assembly lines, in food proc- struction, food preparation, clerical work, pro- essing, in the garment industry, and in offices. duction fabrication, and mining are at risk for Carpal tunnel syndrome, one of this class of cumulative trauma disorders. The Bureau of La- disorders, illustrates the potential for prevention bor Statistics (603) reports 23,000 occupationally offered by the integration of ergonomics, medi- related repetitive motion disorders in 1980, al- cal surveillance, and treatment. though the number of CTS cases is not specified. Carpal Tunnel Syndrome CTS is undoubtedly underreported in aggregate statistics. Research in particular high-risk plants A wide variety of workers (see table 7-I), from provides some insight into the extent of the prob- aircraft assemblers to upholsterers, are among lem. In a study at an athletic products plant, 35,8 those at risk for carpal tunnel syndrome (CTS), percent of workers had a compensable repetitive a progressively disabling and painful condition trauma disorder. In some jobs within the plant,

128 ● Preventing Illness and injury in the Workplace

Photo credit: OSHA, Office of Information and Consumer Affairs

Work on an automobile assembly line can involve cramped working positions. A Volvo assembly plant in Kalmar, Sweden, uses “tipper trolley s.” These trolleys hold the automobile bodies and can be tipped 90 degrees to allow work on the underside of the car

the rate was as high as 44.1 percent, and carpal is formed by the concave arch of the carpal bones tunnel syndrome occurred in 3.4 percent of the and is roofed by the transverse carpal ligament workers (21,23). Many industries claim that the (fig. 7-4). These structures form a rigid compart- incidence of CTS is increasing and is one of their ment through which nine finger tendons and the most disabling and costly medical problems (60). median nerve must pass. Any compromise of this unyielding space usually compresses the median Symptoms nerve.

The onset of symptoms of CTS is usually in- Risk Factors sidious. Frequently, the first complaint is of attacks of painful tingling in one or both hands at Repetitive motions, such as those required in night, sufficient to wake the sufferer after a few many jobs, is one of a number of risk factors for hours of sleep. Accompanying this is a subjective CTS. It is probably the most readily controllable feeling of uselessness in the fingers, which are cause, however. Certain diseases, acute trauma, sometimes described as feeling swollen. Yet little congenital defects, wrist size, pregnancy, oral con- or no swelling is apparent. As symptoms increase, traceptive use, and gynecological surgery all may attacks of tingling may develop during the day, contribute to the likelihood of developing CTS. but the associated pain in the arm is much less Overall, the incidence of CTS is higher in women common than at night. Patients may detect changes than in men, perhaps because of some of these in sensation and power to squeeze things but some risk factors. people suffer severe attacks of pain for many years Occupational tasks responsible for the devel- without developing abnormal neurological signs. opment of CTS include physical exertions with Ultimately, in advanced cases, the thenar muscle certain hand postures or against certain objects, at the base of the thumb atrophies, and strength and exposures to vibration or cold temperatures. is lost. Repeated and forceful up-and-down motions of Compression of the median nerve is the imme- the wrist (flexion and extension) (fig. 7-5), cause diate cause of CTS. The median nerve comes the finger tendons to rub on the structures form- down the arm, through the wrist, then branches ing the carpal tunnel. This constant rubbing can in the hand, supplying the thumb, forefinger, mid- cause the tendons to swell (tenosynovitis), even- dle finger, and half the ring finger with nerves (fig. tually putting pressure on the median nerve in- 7-3). The carpal tunnel itself, located in the wrist, side the carpal tunnel. The nerve itself is stretched

Ch. 7—Ergonomics and Human Factors . 129

Figure 7-3.—Major Nerves in the Arm and Hand

Pal mar (a)

(b)

SOURCE (60)

130 . Preventing Illness and Injury in the Workplace

Figure 7-4.--The Carpal Tunnel

Median nerve

Flexor retinaculum Finger flexor (transverse carpal ligament)

SOURCE (60)

Figure 7-5.— Flexion and Extension of the Wrist by repeated exertions, and compressed between the walls of the carpal tunnel. Forceful movements and the direction of the movement are only two of the underlying causes of tenosynovitis that can lead to CTS. The speed of movements and incorrect posture while working also are important (275). Median nerve com- a. pression also can be caused by tasks that require a sustained or repeated stress over the base of the palm (247). Examples include the use of screwdrivers, scrapers, paint brushes, and buffers. Although the mechanism is not yet understood, low frequency vibration is a recognized risk factor for CTS (405). Vibration exposure may result from air- or motor-powered drills, drivers, saws, sanders, or buffers. Cannon (95) examined medical records at an aircraft company and found a strong association between CTS and use of vibrat- ing tools.

Control of CTS

Bending the wrist causes the finger flexor tendons to rub on Control of CTS requires a two-pronged ap- adjacent surfaces of the carpal tunnel. proach. The primary strategy to prevent cases is the use of ergonomic principles to modify hand- SOURCE’ (60) tools and to improve work-station design and Ch. 7—Ergonomics and Human Factors 131 work practices. Even a successful ergonomic pro- Powered handtools can also be designed and gram will not prevent all cases of CTS, however. used to minimize stress. As illustrated in figure The second important element, therefore, is a 7-8, good designs allow the work to be done with medical surveillance program. This is particularly little or no flexion or extension of the wrist. important now when so little is known about the individual factors that cause some people to de- Armstrong and his colleagues have investigated velop CTS. Thus far, no programs focusing on cumulative trauma disorders in a poultry proc- the medical evaluation of CTS seem to exist (60). essing plant using the procedures described above. They discovered that workers in the “thigh bon- Ways are needed to identify the earliest sign of ing” section had the highest incidence of cumula- CTS, to evaluate progression of the disease, and tive trauma disorders of all departments. Thigh to examine the role of predisposing risk factors. boning involves grabbing the thigh with one hand The purpose of such a medical surveillance pro- on a moving overhead conveyor, then making gram is prevention of advanced disease by in- four cuts with the other to separate the meat from stituting therapy at early stages. the bone. Each worker makes an estimated 15,120 Although medical surveillance for CTS is still cuts per shift. Ergonomic improvements to the in very early stages, ergonomic interventions have process recommended by Armstrong and col- been remarkably successful where they have been leagues include training workers in the “proper instituted. Armstrong (21) describes the steps in- work methods and knife maintenance to minimize volved in developing appropriate controls. First, the time and, hence, the distance that must be plants and specific departments within plants in reached and force that must be exerted on the which there is a documented high rate of CTS thigh.” The work station could be modified to should be identified. Then each job should be minimize the distance to be reached. The work- systematically analyzed. Traditional time-and- ers wear wire mesh gloves with rubber gloves motion studies, in which each movement or act underneath, which increase the force necessary is recorded, can be used. Each element of the job to grasp the thigh and pull the meat away. Gloves can then be checked against factors known to be should fit well, and the addition of barbs on the associated with CTS development. These include palm of the wire mesh glove might facilitate the posture of the hand and wrist, strength, stress con- hand actions. A new knife handle design, to re- centrations over the palm, vibration, cold tem- duce the force required to hold the knife and make perature, and the presence of gloves. the cuts—e.g., that pictured in figure 7-9—is suggested (22). Such a design would also minimize Armstrong presents a typical work task as an wrist flexion. example. Figure 7-6 shows a worker taking parts out of a container and placing them on a con- A high incidence of repetitive trauma disorders, veyor. The six elements involved in this task are including carpal tunnel syndrome, in a telephone reach, grasp, move, position, assemble, and re- assembly plant prompted management to consider lease. Reaching into the container involves wrist how to prevent future cases. McKenzie and col- flexion and pinching, during which the worker’s leagues (299) noted the highest rates in areas using wrist is likely to rub on the edge of the box. The vibratory air screwdrivers, and in jobs requiring forearm is also likely to rub on the edge of the repetitive grasping, squeezing, or clipping mo- work bench while the part is positioned. The re- tions. Ergonomic changes recommended included designed work station should reduce stress on the modifying the screwdrivers with sleeve guards and hand and wrist, and eliminate sharp edges. Good changing work positions to minimize hand and and bad designs for the container and the work- wrist stress. The changes were instituted with bench with jig in this hypothetical case are illus- almost immediate results: from 2.2 percent annual trated in figure 7-7. incidence of repetitive trauma disorders in 1979

132 . Preventing Illness and Injury in the Workplace

Figure 7-6.-Job Analysis: Assembly Tasks

Job analysis 1. Reach for part - 2. Grasp part

4. Position part i 5. Assemble part

• Assembling parts on a moving conveyor can be described by a series of six elements,

SOURCE (21) to 0.79 in 1981. Lost and restricted workdays fell About half the injured workers had received some from 5,471 in 1979 to 1,111 in 1981, and further instruction about proper lifting. reductions were expected in subsequent years. Manual lifting presents a risk of overexertion injuries and cumulative damage to the soft tissues Back Disorders around the spine. Overexertion injuries to the back constitute the largest single category of The Bureau of Labor Statistics completed a sur- workers’ compensation claims, amounting to 25 vey of workers who had incurred back injuries to 30 percent of all disability cases Lower-back while lifting, placing, carrying, holding, or lower- injuries are often extremely painful and signifi- ing objects (605). Of the 900 workers included in cantly diminish the quality of life of the afflicted the tabulated results, more than 75 percent were workers. Many of these can be prevented by job lifting at the time of the injury, Surprisingly, the and equipment redesign. back injuries were concentrated in younger workers; almost 75 percent occurred in 20- to 44-year- The conventional wisdom about how to lift olds. Both the weight and the bulkiness of the something is to squat, pick up the object, and, lifted objects were often associated with injuries. while keeping the back straight, lift straight up

Ch. 7—Ergonomics and Human Factors ● 133

, a. Container .’ .

,. I ,. edges too high Elbow height and too sharp edges too high

Side cut out with rounded edge Elbow height tilted

Containers should be designed so that workers can reach all locations without flexing their wrist. All edges that come into contact with the worker should be well rounded.

b. Workbench and jigs

. . Bad Bad

! Sharp edge of work surface Flat and too high Tilted away

OK \ OK

sOURCE: (21)

I&! . Preventing Illness and Injury in the Workplace

Figure 7-8.—Good and Bad Designs for Powered Drivers

Powered drivers OK OK Bad Pistol handle Pistol handle Pistol handle vertical surface horizontal surface horizontal surface elbow height below waist height

OK Inline handle Bad Inline hand

horizontal surface horizontal surface below elbow height

Wrist posture is determined by the elevation and orientation of the work surface with respect to the workers and the shape .- of the tool.

SOURCE: (21) with the legs. This procedure is thought to pre- method incorporates workers’ perceptions and vent injury to the back. In many cases this pro- sensations into the assessment of work load for cedure is justified, but there are many other situ- both static and dynamic strength. Epidemiologic ations where it is not. criteria are derived from aggregate data concerning the incidence, severity, and distribution of low back pain. These four approaches can be inte- Work-Load Evaluation grated and guidelines established with input from each. The National Institute for Occupational There are at least four basic sets of criteria for Safety and Health has developed recommenda- determining acceptable work loads: biomechani- tions in this way. cal, physiological, psychophysical, and epidemiological (453). Biomechanical criteria are based Maximum acceptable work loads are often ex- on pressures and stress exerted on the body, par- pressed in terms of the weight and frequency of ticularly on the spinal column. Limits of tolerance lifting. For instance, 75 percent of the industrial have been developed by observing damage to male population can lift a 13-kilogram box of cadavers when pressure is applied. Physiological fixed proportions every 5 seconds, and a 34- criteria are primarily metabolic, e.g. oxygen con- kilogram box every 30 minutes without trigger- sumption and heart rate. The psychophysical ing or aggravating low-back injury symptoms.

Ch. 7—Ergonomics and Human Factors ● 135

Figure 7-9.—A Knife Designed to Reduce Cumulative Trauma Disorders in Poultry Processing

One possible knife handle with three blades for reduced wrist deviations. The handle is designed to reduce the tendency for the knife to fall out of the hand in thigh boning.

SOURCE (22)

There is great practical significance to having compensable back problems. For case histories of very specific work-load recommendations. They task evaluation and redesign carried out by the can be used both for determining that a task is Liberty Mutual Insurance Co., see boxes D and unacceptable and as guidelines for redesigning the E. According to Snook (453), “most industrial task. Recognition of a problem often occurs only tasks with unacceptable work loads can be mod- after a number of compensation claims have been ified for less than the average cost of a single low awarded, triggering investigation. Insurance com- back compensation claim. ” panics have a direct incentive to cut down on

APPLYING ERGONOMICS TO VDT DESIGN

The first reports from Europe concerning po- cluded that there were no known radiation tential adverse health effects associated with video hazards and that the real hazards were ergonomic display terminals included accounts of many re- problems: musculoskeletal problems, visual prob- ported musculoskeletal and visual problems. Early lems, and fatigue. Poor design of equipment or U.S. studies of potential VDT hazards also con- poor job design may produce such problems. 136 . Preventing Illness and Injury in the Workplace

Box E.—Ergonomics-Task Evaluation and Redesign This case history involves back injuries to materiaI handlers in the packing department of a metal office equipment and desk accessory manufacturer. The six female material handlers are required to manually lift boxes of various sizes and weights from a conveyor and stack them on wooden pallets. There were six compensable back injuries reported between Feb- ruary 1980 and September 1982, with a cost of $131,415. Boxed products are transported by belt conveyors to the position of the material handler, where they are taken off the line and lowered and/or lifted onto pallets. Due to the wide variety of Sire, shapes, and weights, the boxes handled vary in weight from a pound to large bulky boxes weighing 50 pounds. Full pallets are taken by forklift to a warehouse.

Musculoskeletal problems among office workers range from discomfort to pain and medical Source: (675}. disability. The back, neck, and shoulders are the most frequent sites of problems. Table 7-2 summarizes the results from several studies of VDT VDTs, including equipment design (VDTs, work workers. There is general agreement that muscu- stations, and chairs) and job design (constrained loskeletal problems are associated with poor working positions, repetitive work, and inade- working positions, repetitive motions, and the quate rest breaks). Any of these can be changed. length of work time without a break. The keyboards and screens of many early VDT Figure 7-IO illustrates risk factors contributing work stations were fixed relative to each other and to musculoskeletal problems associated with the height of the work station was not adjustable.

Ch. 7—Ergonomics and Human Factors . 137

Table 7-2.—Frequency of Reported Musculoskeletal Problems From Various Studies of VDT Workers

Percent reporting frequent or daily problemsa Study/type of worker Neck-shoulder Back Arm-hand Arndt, 1980: Telephone operatorb ...... 8-17 15 12 Service assistant ...... 9-14 14 6 Accounting clerkb ...... 7-1o 10 5 Laubli, et al., 1980: Data-entry VDT operator...... 11-15 N.A. 6-15 Accounting machine operator ...... 3-4 N.A. 8 Conversational VDT operator...... 4-5 N.A. 7-11 Typist...... 5 N.A. 4-5 Traditional office worker ...... 1 N.A. 1 Hunting, et al., 1981: VDT operator ...... 6-12 6-10 6-12 Typist , ...... , 6-12 6-10 6-10 Traditional office worker , ...... 2-5 2-5 2-5 Canadian Labour Congress, 1982: VDT operator ...... 10-12 10-15 2-3 Non-VDT operator ...... 7-8 9-14 4 Sauter, et al., 1983: VDT operator ...... 24-34 27-35 3-4 Non-VDT operator ...... 19-28 22-29 4-6 reporting almost frequent, often constant Ranges represent Questions 33 percent use N A = SOURCE (25)

Figure 7-10.— Potential Ergonomic Risk Factors Associated with VDT Design

Display height

keyboard Iocat ion

Chair height

Foot

SOURCE (25) 138 ● Preventing Illness and Injury in the Workplace

Equipment can be redesigned and made to fit the The problems reported are related to the visual various sizes of people. Many manufacturers now demands of VDT work. The intensity and dura- offer, usually at a higher price, office furniture tion of visual demands, the quality of the VDT and equipment that has been “ergonomically de- image, and illumination are important. VDTs signed. ” Although this equipment has been adver- have often been manufactured and introduced in tised as being vastly superior to ordinary office workplaces without sufficient attention to the equipment, its biggest advantage from an ergono- these factors. The National Research Council (321) mic perspective is the simplest—it is adjustable. report notes that “[i]n many instances . . . VDTs Job redesign to reduce repetitiveness and to have been designed without attention to existing alleviate constrained postures is also possible. scientific data on image quality, and many VDTs Finally, rest breaks can be instituted to reduce on the market do not provide the legibility of fatigue from working in one position for long peri- high-quality printed material.” Snyder (454) has ods of time. Rest breaks often result in increased estimated that nearly half of all VDTs now in use productivity as well as fewer musculoskeletal are improperly designed for intensive office use. complaints (25). VDTs have usually been placed into workplaces Visual problems reported among VDT users in- without consideration of the available lighting. clude eyestrain, burning/itching eyes, blurred vi- There has been a tendency to believe that more sion, double vision, color fringes, and deteriora- light creates a better environment for office work- tion of acuity (table 7-3). A recent review by a ers. The result is that many offices are overlit or National Research Council panel (321) concluded, too bright for certain activities. This is a particu- however, that there was no evidence that the use lar problem in offices with VDTs, Increasing the of VDTs led to visual system problems different light level, up to a point, improves contrast and from symptoms reported by workers engaged in visibility for most office tasks. VDTs, however, other tasks involving intensive close work. emit their own light and increasing illumination

Table 7-3.--Frequency of Reported Visual Problems From Various Studies of VDT Workers

Percent reporting frequent problemsa Burning, itching, Blurring, Study/type of worker Eyestrain irritation double vision Holler, et al., 1975: VDT user ...... 21 N.A. 7 Arndt, 1980: Telephone operatorb ...... 8 N.A. 1-5 Service assistant ...... 9 N.A. 1-4 Accounting clerkb ...... 6 N.A. 1-3 Laubli, et al., 1980: Data-entry VDT operator...... 20 N.A. N.A. Conversational VDT operator...... 28 N.A. N.A. Typist...... 18 N.A. N.A. Traditional office worker ...... 5 N.A. N.A. Canadian Labour Congress, 1982: VDT operator ... , ...... N.A. 9-17 13-16 Non-VDT operator ...... N.A. 7-8 8-9 Sauter, et al., 1982: VDT newspaper editor ...... 40 43-46 N.A. Non-VDT newspaper editor ...... 17 17-22 N.A. Sauter, et al., 1983: VDT operator ...... 24 19-27 5 Non-VDT operator ...... 20 21-23 5 aFreqUenCY of problems reported daily, constantly, often, vew often, freWJentlY. bApproximately 33 percent used ‘fDTs. N.A. = Not available. SOURCE: (25). Ch. 7—Ergonomics and Human Factors ● 139 levels actually reduces the contrast and visibility nation problems that must be addressed concern of the screen image. reflections off the VDT screen, glare, and contrast. There are techniques for reducing and eliminat- It is frequently suggested that the general lighting most of these problems and for enhancing the ing level in a room be somewhat reduced for ability of each worker to adjust the environment screen-based work. An alternative recommenda- of the VDT to fit his or her needs. Unfortunately, tion is to provide even lower levels, with sup- these techniques have very often been ignored by plementary task lighting for reading text from manufacturers, vendors, and employers (25). paper and for other non-VDT work. Other illumi-

SUMMARY

Ergonomics —techniques for examining worker- tive trauma injuries. Changing work practices, machine interactions and design of machines to tools, and machine design and redesigning plant fit workers better—can be important in reducing layout are among the principal ergonomic solu- injuries and discomfort from repetitive motions tions. There are many examples of the incidence and other work activities. Information ergonomics of carpal tunnel syndrome and back disorders, involves workers’ receipt of sensory information two common repetitive trauma injuries among in- and how it is processed; physical ergonomics deals dustrial and office workers, being reduced through with worker body size, strength, and stresses the use of ergonomic designs. In addition to pre- while working. Generally, when faced with a venting overt and cumulative trauma injuries, problem in the workplace, ergonomists systemat- ergonomic design has been shown to increase ically analyze the job using biomechanical, phy- worker productivity and efficiency by making the siological, psychophysical, and epidemiological machine or tool more closely fit the worker and analyses; each job element is compared with fac- the worker’s capabilities. tors known to be associated with overt or repeti-

8. Personal Protective Equipment 8-1. Comparison of Extrapolated and Measured Respirator Protection Factors ...... ● ...... 148 8-2. Comparison of Noise Reduction Achieved by Insert-Type Hearing , Protectors in the Laboratory and in Field Use ...... 158 8-3. NIOSH Testing of Various Types of Personal Protection Equipment.. ,,.. 160 8-4. Comparison of Various Protective Garment Materials’ Capacity to Resist Penetration by Chlorinated Ethanes and PCB ...... 167 8’-5. Selected Information From BLS Survey of Head Injuries ...... 168 8-6. Selected Information From BLS Survey of Eye Injuries ...... l69

LIST OF FIGURES Figure No. Page 8-1. Respiratory Protection Devices ...... 144 8-2. Effect of “Non-wearing” on the Effectiveness of Respiratory Protection . . . 151’ &3. Protective Eyewear ...... 163

8 Personal Protective Equipment

Based on an analysis of the literature about personal protective equipment, OTA concludes that the effectiveness of many of these devices, especially under conditions of use in the workplace, has not been demonstrated. Instead, many devices have been tested only in laboratory situations that do not duplicate and may not even approximate workplace conditions. The overall impression is that test results tend to exaggerate the effectiveness of personal protective devices. Additionally, the few in-workplace evaluations reveal that continual maintenance and supervision, which are seldom provided, are necessary for acceptable performance. Photo credit E.I. du Pont de Nemours & Co. This discussion is divided into three parts. Greater emphasis is placed on respirators and Signs are frequently used to indicate when the use of hearing protectors than on all other personal pro- protective equipment is mandatory tective devices. Respirators and hearing protection devices are primarily intended to protect tinue to be suggested as less expensive and equally workers’ health, while most other personal pro- effective alternatives to engineering controls for tective devices are important for safety. Further- health risks. Knowledge of the effectiveness of more, both respirators and hearing protectors are these devices is thus important for making com- frequently mentioned in arguments about the costs parisons of the costs and benefits of personal proof workplace health. They have been and con- tective equipment versus engineering controls.

RESPIRATORS

Dust masks, gas masks, and devices that supply respirators shown in the figure, all air-supply de- clean air to workers through hoses or from tanks vices, are positive-pressure respirators, which are all called respirators. The most common are supply clean air to the wearer from a tank, from “dust masks” that employ fiber filters to prevent a hose that originates in an area of clean air, or particles from being inhaled (see fig. 8-l). The sec- from a hose that supplies air that has been purified ond general class of respirators are “gas masks, ” by passage through a filter or sorbent. familiar to most veterans of the armed forces. They purify air contaminated with fumes and va- Federal involvement in testing and certifying pors by passing it through a chemical “sorbent. ” respirators originated in congressional and pub- Less common are air-supply devices. Figure 8-1 lic concern about coal mine safety. Beginning in illustrates a variety of respirators. The disposable the Civil War period, a number of bills were in- face mask “dust mask” and the “reusable air- troduced and passed by one or the other House purifying respirator” are negative-pressure respi- of Congress to set up an agency with responsibil- rators and rely on the wearer’s breathing to pull ity over mineral industries. In 1910, a series of “outside” air across the filter or sorbent. The other dramatic mine disasters led Congress to establish

143

144 ● Preventing Illness and Injury in the Workplace

Figure 8-1 .- Respiratory Protection Devices

Abrasive bias.tlng helmet Disposable respirator by 3M by 3M

Reusable air-purifying respirator by Cesco

Powered alr purtfying helmet by Racal

Airline respirator by Willson

Self-contained breathing apparatus by Globe

Ch. 8—Persona/ Protective Equipment 145

In 1969, Congress brought the Department of Health, Education, and Welfare into respiratory protection certification. The Federal Coal Mine Health and Safety Act of that year included the Department’s Bureau of Occupational Safety and Health in a joint certification and testing program with the Bureau of Mines. A year later, the Bu- reau of Occupational Safety and Health was replaced by the newly created National Institute for Occupational Safety and Health (NIOSH), and responsibility for testing and certifying respirators was assigned jointly to NIOSH and the Bureau of Mines by the Occupational Safety and Health (OSH) Act. (The mine safety function was moved to the Department of Labor in 1977 and is now called the Mine Safety and Health Administration (MSHA).) More importantly, Federal regulations require that industrial users select federally approved respirators, if they are available. Under authority of the Federal Mine Safety and Health Act of 1977, NIOSH established a laboratory to certify respiratory protective devices, Al- though the Occupational Safety and Health Ad- ministration (OSHA) can accept testing results for certification from any laboratory, it has chosen to accept only NIOSH certifications. Therefore, there is one laboratory responsible for respiratory protection certification (68). Photo credit Department of Labor, Historical Office

An example of a respirator from earlier in this century Qualitative and Quantitative Fit Testing of Respirators the Bureau of Mines in the Department of the In- Most tests of respirator effectiveness are con- terior. The Bureau was interested in techniques ducted in laboratories and usually maximize the and equipment for use in mine rescue operations, apparent effectiveness of the devices. For instance, and, because the air in mines following cave-ins most air-purifying (negative-pressure) respirators and fires was often unfit to breathe, some effort depend on the user’s inhalation to create a nega- was devoted to development of respiratory pro- tive pressure inside the mask. In theory, the only tection. source of air to equalize the pressure is air that passes through the air-purifying system (i.e., a fil- The use of poison gases during World War I ter or sorbent). The practical realization of the spurred on the work of the Bureau of Mines, theory requires that the seal between the edges which cooperated with the military to develop of the mask and the wearer’s face be sufficiently masks for use by American soldiers in France. In tight to prevent contaminated air leaking in from 1919, following the war, the Bureau published the sides. procedures by which manufacturers could apply for certification of gas masks and breathing appa- Many factors may reduce the security of the ratus for use in mine rescues (438). seal-changes in the tension of the headstraps that 146 . Preventing Illness and Injury in the Workplace secure the mask, daily growth of facial hair or Wilmes (673,674) provides a readable and in- the presence or absence of a beard, perspiration, formative discussion of the two types of fit tests. head movements, and talking. Although labora- He maintains that quantitative measurements, al- tory tests attempt to make allowances for such though providing greater precision, are not worth factors, no test duplicates “field situations.” Two the additional costs. Instead, the money saved general methods are available to test whether the from not doing the quantitative tests would be respirator is properly worn, and, indeed, if a par- better spent by increasing efforts directed at in- ticular respirator can protect an individual. struction of workers and reinforcement of good respirator habits, careful maintenance, and edu- The “qualitative fit test” relies on the sensory cation of workers, supervisors, and managers perception of the worker wearing the mask. A about the importance of respirators. Additionally, chemical that has a distinctive smell or taste (iso- he states that provision of different models of amyl acetate, which smells like bananas, or sac- respirators, which costs money, is a better invest- charin) or that is an irritant (stannic chloride or ment because it allows workers to choose respira- titanium tetrachloride) is introduced into a cham- tors on the basis of comfort. A respirator that is ber where someone is wearing a respirator. If the not worn provides no protection; having a single wearer detects the smell or irritation, the fit of model of respirator that, in tests, provides ex- the mask is judged to be inadequate. In a program cellent protection but is uncomfortable does not designed to match appropriate respirators to peo- provide good workplace protection. ple with different facial shapes, other masks would be tried until one was found that passed the qual- Wilmes’s position is not shared by others. Both itative fit test. NIOSH and the Los Alamos Scientific Laboratory (LASL) endorse quantitative fit procedures. In The “quantitative fit test,” on the other hand, their opinion, a worker’s sense of smell and taste uses instruments to measure concentrations of the may be insensitive under certain conditions— contaminant inside and outside the mask. For allowing mistakes to be made. Mistakes may be instance, the respirator-wearing worker can be especially common if the worker tries several subjected to an atmosphere containing dioctyl masks and his or her senses become jaded to the phthalate (DOP), and instruments can be used to test substance. A worker who is in a hurry may measure concentrations of DOP. This method has say that the respirator is “okay” too quickly. two obvious advantages: It does not depend on Alternatively, as demonstrated by a study dis- human sensory perceptions, which may vary be- cussed later in this chapter, the worker may tween workers and for a single worker depend- “smell” the test substance even though it is not ing on a number of factors, and it provides a detectable by instruments. Finally, only a quanti- quantitative measure of how well the respirator tative fit test provides information about the de- works. gree of protection. The quantitative measure is generally expressed Disagreements about the merits of quantitative as a “protection factor” or “PF,” the ratio of the fit tests have been going on for years, with OSHA, concentration of the test substance outside and for instance, taking different positions at different inside the mask. One disadvantage is cost. The times. The lead standard that became final in 1978 testing requires highly trained personnel and the required quantitative fit testing. However, revi- necessary equipment costs up to $10,000, accord- sions made in 1982 allow qualitative testing. If, ing to a 1978 NIOSH estimate. Another disadvan- as is expected, NIOSH formally requires quanti- tage is that DOP, as is the case with many phthal- tative tests in its revised respirator testing regu- ates, is a suspect carcinogen. Substitution of lations, the argument about certification tests will another test agent is possible, and aerosols of be settled. Arguments will probably continue sodium chloride (table salt) have been used in about what methods employers should use to fit some tests. respirators. .—. — —.—

Ch. 8—Personal Protective Equipment ● 147

Use of Qualitative and Quantitative Fit Tests in titative fit test in the laboratory). Therefore, Certification Programs according to these study results and assuming that a PF of 10 is sufficient and that a test-measured Currently, under the provisions of 30 CFR Part PF of 10 is reflective of the protection to be ex- 11, NIOSH certification of respirators employs pected on the shop floor, an employer can quali- the isoamyl acetate (IAA) fit test. A panel of peo- tatively test various respirators and select one that ple chosen to represent a cross section of facial will provide adequate protection. sizes and shapes tests each mask. Each subject is first checked to ensure that he or she can smell Unfortunately, 16 of the 22 workers, who—ac- IAA; in general, people chosen to participate in cording to the quantitative fit test-were protected the tests can detect it at concentrations of 1 to 3 to a PF of at least 10, reported smelling MA. This parts per million. A subject enters a chamber in suggests that an employer who relies upon quali- which the concentration of isoamyl acetate is be- tative fit tests might also have to provide other tween 100 and 1,000 parts per million and is asked respirators for workers who detect with a particu- whether the respirator prevented him or her from lar mask IAA even though a quantitative test smelling the chemical. The minimum requirement would show that they would be adequately pro- is tected. In other words, although the quantitative fit test costs more, for instruments and operators, six persons will each wear the apparatus in the test concentration. . . for 2 minutes and none its absence may also generate costs because work- shall detect the odor or taste of the test vapor (30 ers’ senses lead them to report a poorer fit than CFR 11.85-19). has been achieved. Evidently, NIOSH is moving away from qualitative tests and toward requiring quantitative fit tests. According to Wilmes (673), NIOSH objected The Los Alamos Scientific Laboratory to the 1980 American National Standards Insti- Tests of Respirators tute (ANSI) consensus standard on respirators that The Los Alamos Scientific Laboratory studies included reliance on the qualitative fit test because are the major work on respirator effectiveness, the agency concluded that such a test was unable Begun in 1969, they are gererally regarded as to predict fit. The ANSI Board of Standard Re- thorough and definitive studies. Nevertheless, view considered NIOSH’s objection and decided their importance does not appear to be based on that it was without merit, a decision NIOSH ap- their superiority to other studies. Rather, they appealed to the ANSI Board of Directors. But in pear to be the only studies available for many res- 1980 the agency agreed to the consensus stand- pirator types. ard with the provision that a statement be added to it saying that NIOSH takes the position that LASL used the DOP quantitative tests to meas- only quantitative fit tests should be used. ure and assign PFs to classes, or categories, of respirators. The standard for assigning a PF to a Comparing Results of Qualitative and class was to determine the highest PF obtained by Quantitative Fit Tests 95 percent of all subjects using each respirator in the category. In practice, that means the “class” A former Chairman of the ANSI Respiratory PF was determined by the poorest performing res- Practices Standard Subcommittee has reported a pirator in the class. For instance, the testing of comparison of results obtained in the two types “full-facepiece air-purifying respirators” involved of fit tests (386). Four models of NIOSH/MSHA six different masks. The one that provided the approved respirators were tested by 22 people (19 lowest protection was shown to protect 97 per- males and 3 females) using qualitative and quan- cent of the tested men to a PF of SO, so that cate- titative methods. gory was assigned a PF of sO. Other masks in the Importantly, all respirator wearers who did not same class provided better protection; the best smell IAA in the qualitative fit test were protected provided a PF of at least 2,000 for 97 percent of to a PF of at least 10 (as measured by the quan- tested men. 148 . Preventing Illness and Injury in the Workplace

At the end of the first phase of testing, LASL Also shown in table 8-1 is an exception to the assigned PFs to the classes of respirators it tested general trend of extrapolated values for positive- and extrapolated PF values for other classes (219). pressure respirators underestimating their effec- As expected, positive-pressure masks provided tiveness. NIOSH (580), after a complaint,, earned higher PFs, from 1,000 to 10,000. Positive pres- out field studies of one model of “high-efficiency sure tends to blow contaminants away from any Powered Air-Purifying Respirator (PAPR)” and leaks or ventings around the mask, minimizing found that measured PFs in the workplace were the influence of differences in personal anatomy. significantly below the PF of 1,000 extrapolated from LASL tests. By contrast, negative-pressure respirators tend to suck contaminants into the protective mask Before 1969, when the LASL studies began, no through any openings, which makes proper fit- one had conducted laboratory tests of respirators ting critical. The highest PF assigned to any class in a systematic fashion. Furthermore, the first field of negative-pressure respirator was 50, although studies of respirator effectiveness were not pub- some specific respirators in some classes achieved lished until 1973 (see section on field testing later much higher PFs. The PF assigned to a class was in this chapter) and few have been reported. Two judged to be applicable for all respirators of that million workers who wear respirators rely on this class that are used in a respirator program that limited research for assurance of protection. includes routine equipment maintenance and qualitative fit testing. “Dust Masks" In a subsequent phase of testing, LASL measured PFs for other classes of respirators and com- The progenitor of the most widely used type pared them with the extrapolated PFs. In the ex- of respiratory protection was a defective bra cup amples of this work given in table 8-1, measured that fell off an assembly line in 1972. “Someone PFs were found to be higher for positive-pressure with a bright idea at 3M Corporation clipped an respirators than the earlier extrapolated values, elastic band to the fiber cup and produced the first and the measured PFs for a class of negative-pres- disposable dust mask, which, with some modifi- sure masks was lower (196). cations, now claims the largest share of the multi-

Table 8-1.—Comparison of Extrapolated and Measured Respirator Protection Factors

Extrapolated Pressure protection Class of respiratora in mask factor Measured protection factor Loose-fitting supplied air hood . . positive 2,000 b,c 2 tests <1,000 9 tests 1,000-10,000 36 tests >10,000 47 tests totald Continuous-flow supplied air: Half-mask ., ...... positive 1,000 10,000 (combined results, Full-facepiece ... , ...... positive 2,000 both half- and full-facepiece)e Pressure demand: Half-mask ...... positive 1$300 20,000 (combined results, Full-facepiece ...... positive 2,000 both half- and full-facepiece)e Demand: Half-mask ...... negative 10 <5 (93 percent attained PF Full-facepiece ...... negative 50 of 5)* Powered air-purifying respirator (PAPR) ...... positive 1,000 <15 percent achieved PF of 1,000f ~he respirators are descrlbad in Working Paper #9. bprotectlon factor (pF) reached by at least 95 percent of subjects wearing any respirator of that class cHyatt (219). dbu~la~l Hesch, and LOWV (152) ‘Hack, et al. (196) ‘NIOSH (580) SOURCE Office of Technology Assessment from cited references.

Ch. 8—Personal Protectlve Equipment ● 149

There is often little warning that a dust mask has failed to provide protection. In very dusty workplaces, the first indication may be streaks of dust on the worker’s face. Short of such drastic failures, tasting or feeling the dusty material on the mouth or nose may alert the worker to the absence of sufficient protection. On the other hand, a person’s senses may become acclimated to the dust, rendering the worker unable to detect the mask’s failure. Field Testing of Respirators Duplicating a worker’s movements and activities in a laboratory is difficult. In the course of a day, a worker walks or rides or runs from place to place, reaches and lifts, bends and stretches, gestures and converses, eats and drinks. Each of Photo credit: NIOSH those actions may affect the fit of a respirator, and some of them require its removal. Even if the A worker wears a powered air-purifying respirator (PAPR) during a NIOSH field study respirator, at the start of the workday, provided protection equivalent to that measured in the laboratory, it is unlikely that those conditions would million dollar market for industrial respirators. . .“ prevail at the end of a shift. (178a). A few reports describe field testing of respira- Lowry, Hesch, and Revoir (272) tested six tors. The more sophisticated tests involve at- masks on a panel of 5 males and 5 females. Quan- taching one sampling device to the worker’s cloth- titative measurements of the leakage of sodium ing and one inside the respirator and measuring chloide aerosols into the masks showed that only concentrations of airborne dusts, fumes, and two of the six masks provided a PF of at least 5 vapors inside and outside the respirator. Other for all 10 subjects. These findings suggest that the evaluations depend on collecting “pencil and pa- LASL study (219), which had earlier assigned a per” data about whether government-approved PF of 5, overestimated the effectiveness of dust respirators are used in the workplace, whether the masks. mask is the appropriate type for the hazard, whether respirators are worn correctly and at the Interestingly, the dust masks failed to provide right times, and whether the equipment is prop- a PF of 5 far more often for females than for erly maintained. males. Gillen (181) pointed out that two models that, as tested, provided a PF of at least 5 for men would fail to provide that level of protection for Protective Factors Realized in Field Studies 17 percent of women because of differences in face of Respirators size. Another model that, as tested, provided a PF of at least 5 for 94 percent of men would fail NIOSH (59) has reported on the use of respira- to provide that level of protection for 56 percent tors in the abrasive blasting industry during 1971- of women. Given the large number of women em- 73. The results were not encouraging. PFs (which ployed in some dusty trades, such as the textile according to current NIOSH usage would be pro- industry, and the increased participation in the la- gram protective factors; see box F) ranged from bor force by women, the poor level of protection less than 2 to 3,750. Lower PFs were associated accorded women is especially noteworthy. with poor maintenance, poor training, poor fit-

150 . Preventing Illness and Injury in the Workplace

16(9 percent), less than 1.0. The EPFs of less than 1.0—which mean that the concentration inside the mask was higher than the concentration outside— are thought to have resulted from dust collecting in respirators while hanging around miners’ necks. One general conclusion reached from the study was that miners did not wear their masks enough (201). The WPF was measured during a half-shift timespan after the respirator was donned according to the manufacturer’s instruction. As expected, WPFs tended to be higher because of the attention given to fitting the respirators. Two respirators produced mean average WPFs of greater than 11 on all nine of the miners who participated in the study; all five tested respirators produced WPFs greater than 5 on all nine men. Therefore, the observed WPFs show that PFs approaching those measured in the laboratory can be achieved, at least for relatively short periods of time under conditions of close observation and continuous use. However, the EPFs show that those high levels of protection were seldom reached under conditions of normal use. Smith, et al. (450) observed a wide range of EPFs (from 1.12 to 146) in a study of cadmium workers, which resulted partly from workers deciding whether or not to wear respirators in different situations. The paucity of information on how effective ting of respirators, and inadequate supervision of respirators are in the workplace is well illustrated use. Although in many cases inappropriate res- by the observation of these authors that “only one pirators were used, low PFs (2 or less) were found other published study was found on the effective in at least some cases with all types of respirators, protection provided by intermittent respirator pointing to problems of fit, maintenance, and use. usage.” Similar findings of poor respiratory protection The major—and, in retrospect, perhaps obvi- were also reported in NIOSH-contracted studies ous—conclusions reached from the papers by of paint-spraying operations (481) and coal min- Hams, et al. (201) and Smith, et al. (450) are that ing (200,201,202). The coal mine study measured respirators do not afford protection unless they the effective protection factor (EPF) and the are worn and that the degree to which they af- workplace protection factor (WPF). ford protection depends on how well they fit and how well they are maintained. The EPF was determined by measuring dust concentrations inside and outside the respirator Figure 8-2 illustrates the rapid decrease in EPF during a complete workshift, whether the respira- as the time the respirator is not worn increases. tor was being used or was hanging from the wear- The uppermost curve shows that not wearing a er’s neck. Twenty tests (11 percent) produced EPFs respirator with a WPF of 10 for 1 minute each greater than 10; 54 (29 percent), less than 2.0; and hour reduces the EPF to 90 percent of the WPF;

Ch. 8—Personal Protective Equipment 151

Figure 8-2.—Effect of “Non-wearing” on the Effectiveness of Respiratory Protection

100 J

0 1 2 3 4 5 6 Minutes respirator is not worn per hour SOURCE Adapted from (403) for 6 minutes, it drops to less than 70 percent. The results of field testing. The LASL studies (219) EPFs, then, if a respirator with a WPF of 10 is assigned a PF of 10 to half-mask, air-purifying not worn for 1 minute or 6 minutes each hour respirators. In the coal miners’ study (201), an become 9 and less than 7 respectively. The per- overall median EPF of 3.2 was observed; in the centage decrease increases with higher WPFs; a cadmium workers’ study (450), a geometric mean respirator with a WPF of 100 that is not worn for of 3.9 was noted, “which compares favorably with 1 minute per hour provides an EPF of only about the median 3.2 found by Hams, et al.” Overall, 40; if not worn for 6 minutes, the EPF is only then, the protection factors realized by workers about 10 (403). in those two studies were about one-third those predicted in the laboratory. Before turning from the limited information about effectiveness of respirators in the work- The mean and median (or average) effective place, the protective factors assigned on the basis protective factors obscure the high and low EPFs of laboratory tests can be compared with the obtained by some workers. In the study of coal 152 ● Preventing Illness and Injury in the Workplace miners, 11 percent of EPFs exceeded 10; some were Several reports of studies of respirator effective- as high as 20. In the cadmium workers’ study, 22 ness in the workplace are soon to appear or have percent of EPFs exceeded 10. Furthermore, in the just been published (674a). The founding in 1982 latter study, one “fastidious worker,” who was of a new professional society, the International exposed to the highest levels of cadmium also Society for Respiratory Protection, and its pub- achieved the highest EPF. On the other hand, lication of an international journal are expected about 9 percent of miners achieved EPFs of less to increase the availability of information about than 1.0 and 29 percent had less than 2.0 (450). testing. Miners who realized EPFs of less than 1.0 were worse off than if they had not worn a mask at Deficiencies in Respirator Programs all. Those achieving EPFs of less than 2.0 were little better off although they might have thought Nicas (329) prepared a working paper about they were protected from airborne hazards. respiratory protection programs for the use of unions affected by the cotton dust and lead stand- Dixon and Nelson (145) and Myers (315) re- ards. The paper drew attention to deficiencies in ported EPFs that stand in marked contrast to those many respirator programs because of poor main- reported above. The first authors found WPFs tenance and supervision, to the limited testing of ranging from 40 to greater than 27,000 for work- respirators in the workplace, and to the costs of ers who had been instructed in the proper use of maintaining an effective respirator program. respirators and who wore masks during the 30 minutes to 2 hours necessary to complete specific The sometimes high cost of a proper respira- tasks. One reason offered for the greater efficiency tory protection program (111) was suggested as was the use of a new model of respirator (146). a bargaining chip in labor-management negotia- Use of silicon rubber for the facepiece produces tions about whether to install engineering controls a much more comfortable fit with the face, and or depend on respirator programs. The paper by the mask can be worn for longer periods without Nicas (329) is especially interesting because it pro- causing discomfort that leads to easing the respi- vides a concise, readable review of the respirator rator off the face. testing literature. As it is available only in a photocopied form, it has a limited distribution, By another measure, Dixon and Nelson (145, however. 146) determined the Program Protection Factor (PPF) for a respirator program directed against Dixon and Nelson (146) point out the impor- lead. In six of seven groups of workers who wore tance of good respirator programs in achieving respirators, the PPF was about 36. Technical high protection factors. Rosenthal and Paull (402) problems prevented accurate measurement of the examined OSHA inspection records to determine PPF in the seventh group, but those workers also how frequently respirator programs were cited for achieved significant protection. falling below OSHA standards. From 1977 through 1982, at least one respirator program violation Myers (315) measured WPFs in a primary lead was found in about 10 percent of all OSHA health smelter and a blast furnace area. The median inspections. “All” health inspections includes those WPFs were 450 and 130, respectively, with ranges of establishments that have no respirator pro- from 110 to 2,200 and from 10 to greater than grams. They (402) estimate that 27 percent of all 1,700. (The respirators that were used in those respirator programs were in violation of all workplaces had rated protective factors of 10 OSHA respirator program standard. This percent- based on the LASL studies.) age was constant over the 6 years from 1977 to Dixon and Nelson (146) state that higher WPFs 1982. The importance of respirator programs is were seen in their own work and in Myers’ studies underlined by their finding that respirator pro- because the research was carried out in work- gram deficiencies were found in 40 to 70 percent places with good respirator programs and “ade- of inspections in which overexposure was docu- quate fit testing.” mented. — -.

Ch. 8—Personal Protective Equipment 153

Adequate respirator programs include provid- of approvals does not show that innovative res- ing, maintaining, and insisting on the proper pirators have been introduced, for each time an wearing of respirators. Given that more than a “old” design is cleared for a new use, it receives quarter of all OSHA-inspected respirator pro- an additional approval. grams are cited for noncompliance, simply pro- In general, the resources of NIOSH are seen as viding good respirators in a workplace is not being woefully inadequate to carry out all the enough. Occupational health professionals insist desired activities in the area of respiratory pro- that careful attention to the respirator program tection research and development, testing, and is essential, and they also point out that the costs certification (305,393). of such a complete program must be figured into any comparison made between engineering con- OTA staff heard many complaints about NIOSH’s trols and respirators (111). testing and certification program—primarily that it was slow, bureaucratic, restricted innovation, NIOSH Certification of Respirators and depended on outdated criteria for an acceptable respirator. While all those complaints evi- NIOSH tests complete respirator systems con- dently have some basis in fact, NIOSH is far from sisting of the mask and any air-purifying filters, satisfied with all the companies in the respirator cartridges, and canisters or air-supply apparatus. industry. For instance, the first test that NIOSH The agency certifies the respirator for specified applies to a respirator when it arrives for testing, uses (for instance, against a particular chemical) “the shake test, ” reflects that quality control is not and if the manufacturer subsequently decides that always good. In that test, a NIOSH employee certification of the same respirator against another takes the respirator out of the box in which it ar- hazard is needed, the new use must be submitted rives, shakes it vigorously, and if some piece falls for NIOSH approval. off, the mask is sent back to the manufacturer. As part of this program, NIOSH purchases respirators from suppliers and checks that they meet Possible Changes in NIOSH certification standards. Additionally, a program Certification Procedures that was initiated to investigate complaints about NIOSH is considering changes in its respirator inadequate respirator performance has been ex- testing regulations that would shift responsibility panded to include field studies. The latter program for testing from the NIOSH laboratory to the has been under way since 1981, and reports from manufacturers (344). This approach follows sug- it are now appearing (315 and see below). gestions by a group of five consultants to NIOSH In large part, NIOSH’s testing specifications (30 (73) who concluded that routine testing consumed CFR Part 11) were developed, published, and re- too many resources and that the current proce- vised by the Bureau of Mines from 1920 through dures lacked a feedback loop to alert NIOSH to 1970. Those somewhat dated specifications adopted failures in respirators that were already on the by NIOSH in the early 1970s have been criticized market. Evidently, the revisions forwarded to Dr. by industry (305,393) and Government officials Donald Millar, the Director of NIOSH, in early (360,572) because they are believed to restrict in- summer 1983 would allow self-certification, but novation in respirator design and to be inappro- he was not satisfied that NIOSH retained enough priate for testing devices that will be used in the authority to assure itself of the validity of the workplace (360, 610). To date, however, oppo- manufacturers’ tests. The revisions are being nents of the current regulation have been unsuc- redrafted. cessful in getting a new version adopted. These changes would significantly alter the Increasing emphasis on workplace health has process of testing respirators. NIOSH would no been accompanied by a greatly increased work longer receive samples of respirators and evaluate load; in 53 years, the Bureau of Mines approved them in its own laboratory. Instead, manufac- 340 devices, or about 6 each year. In 1981, NIOSH turers would test respirators against standards to issued 99 new approvals (67). This large number be developed by NIOSH, and, when they are 154 ● Preventing illness and Injury in the Workplace satisfied that the devices meet the requirements, Third= Party Testing of Respirators they would be allowed to represent the respirator as up to NIOSH standards. NIOSH’s role An alternative both to NIOSH testing and cer- would be essentially twofold: It would write the tification and to self-certification would be to have standards, and it would carry out “spot checks” the testing done by a third party, such as the of respirators on the market. Safety Equipment Institute (SEI), which springs from the Industrial Safety Equipment Association To play a more active role in self-certification, (334). SEI would undertake certification only if NIOSH may, as part of the revised 30 CFR Part NIOSH leaves the field of routine respirator 11, reserve the right to see any data produced by testing, for there is no future in being a competi- manufacturers in the self-certification process. In tor with Government testing and certification. (A addition, it could require notification by manu- description of SEI testing of other personal pro- facturers of any major modifications in their tective equipment is provided at the end of this products. chapter. ) The reported information (344) did not mention SEI argues that a third-party laboratory would a feedback loop for manufacturers or users to alert free the Government from routine testing and pro- NIOSH about failures in respirators. Such a proc- vide public assurance of higher quality testing ess would seem to be necessary to ensure that than is possible under self-certification. If ar- NIOSH could alert users to possible difficulties. rangements were made for a third-party certifica- Union representatives, while generally support- tion program, SEI estimates that it could estab- ing changes in the regulations about testing and lish a program within a year by contracting with certification that would make the tests more pre- existing laboratories (672). Other estimates are dictive of workplace performance, favor NIOSH that 2 to 3 years would be needed to equip and continuing its testing and certification. One area staff an adequate testing laboratory (610). they single out for attention in any revision is The idea of a third-party testing and certifica- changes in the recertification process, by which tion program is supported by some manufacturers respirators are removed from commerce and use but not others. Those who support it see an op- when NIOSH finds they are deficient (610). portunity to have a more timely and responsive There are many rather vague suggestions that program than NIOSH has been able to provide the testing requirements of 30 CFR Part 11 be and, at the same time, to increase acceptance over changed to encourage innovations in respirator what would be expected from self-certification. design. John Moran, then chairman of the respi- Supporters draw an analogy between the sug- rator Research Subcommittee and now director gested function and the Underwriters Laboratory of Safety Research at NIOSH, made a specific sug- that certifies many electrical devices. Those who gestion (310): that a minimum standard be set for oppose third-party certification see it as essentially approval of all respirators of each class, as is now a self-certification program because safety equip- done, and that higher standards also be indicated. ment manufacturers are involved in SEI’s direc- In this way, a manufacturer who produced a res- tion; although these critics do not fault self-certi- pirator that was significantly better than others fication, they are uneasy about a third-party on the market would be rewarded with a higher testing program that they see as under control of degree of certification, which would be useful in the manufacturers. marketing.

HEARING PROTECTORS

Exposure to continuous noise at levels greater cling at 40 miles per hour so feet away, may be than 80 decibels (dB), about the noise level of a associated with progressive loss of hearing. Noise garbage disposal at 3 feet or of a diesel truck trav- above 90 dB is definitely associated with hearing Ch. 8—Personal Protective Equipment ● 155 loss. Reduced noise exposures can be accom- various frequencies was established at the begin- plished by engineering controls (redesigning, muf- ning of the 5-year exposure period and tested fling, or enclosing machinery or providing work- again at the end. ers with soundproof areas), by administrative Evidently the hearing conservation program controls (reducing the time workers spend in noisy was quite successful over a period of at least 5 areas), and by use of hearing protectors that re- years. Changes in the hearing levels of workers duce the level of noise reaching the workers’ ears. in the three noise levels were essentially the same OSHA requires employers to reduce workpace and did not vary depending on the hearing level noise levels to 90 dB (about the noise level en- observed in the first measurement. The latter point countered in a newspaper press room) by the use is important because there is the possibility that of engineering or administrative controls. When workers who are already hearing-impaired might such controls are infeasible or cannot achieve 90 be more sensitive to continued noise. Results from dB, employers must issue and require the use of the Du Pont study (364) showed that hearing-im- hearing protectors to reduce the noise exposure paired workers, as well as workers with normal to less than 90 dB (29 CFR 1910.95). In addition, hearing, were protected from further loss of hear- any worker who has lost a certain amount of ing by a program that used hearing protectors in hearing ability, as defined by OSHA, must wear noisy areas. hearing protectors if exposed to noise greater than Despite those results, some reservations must 85 dB. The hearing protectors must work well be attached to the conclusions drawn from the enough to reduce the noise that reaches the work- study. Hearing loss is known to increase with time ers’ ears either to no more than 90 dB or, if some of exposure, and studies conducted over periods hearing loss has already occurred, to no more than longer than 5 years are necessary to be certain of 85 dB. the continued success of hearing conservation programs. The importance of longer-term studies is apparent when it is remembered that many peo- Hearing Conservation Programs ple work 40 to 45 years, and their hearing should be conserved throughout that time. Programs designed to reduce hearing loss generally have three components: identification of Temporary Threshold Shift noisy areas; implementation of engineering, administrative, and personal protection controls to Immediately following exposure to a suffi- reduce exposure to noise; and audiometric testing ciently loud noise, the ability of a person to hear of workers to check that the controls are protec- quiet noises is reduced. Over time, unless there ting hearing. Hearing protectors are an essential has been permanent hearing impairment, the abil- part of hearing-conservation programs. ity to hear quiet noises will return. Temporary loss of hearing acuity is called temporary [hear- Economic considerations make it unlikely that noise can be reduced at the source to acceptable ing] threshold shift (TTS). levels in the immediate future and isolation of In hearing conservation programs, hearing tests workers in sound-attenuating enclosures or reduc- are administered 14 hours after the last exposure tion of an individual’s exposure time is not always to loud noise so that TTS will not interfere in the (389). practical person’s ability to hear. Richman (387), however, The Du Pont Company has maintained a pro- argues that hearing loss involves a lengthening of gram to protect workers’ hearing since the 1940s the period of TTS. He suggests that audiometric (364). A study was undertaken of the hearing acu- testing within 4 or 6 hours after exposure would ity among workers in three different situations: pick up cases of TTS, and altered TTS readings quiet, office-like areas; work areas with noise would alert the responsible authorities that hear- levels in the general range of 85 to 94 dBs; and ing protection was insufficient. (Of course, in or- the noisiest work areas. Each of the men in the der to have a baseline for comparison, the initial study worked in one of the three noise levels for hearing tests should be made 14 or more hours 5 years. The ability of each to hear noise at after the last noise exposure. ) 156 . Preventing Illness and Injury in the Workplace

By changing the time of routine audiometric Objective. —This method also uses micro- testing to within 4 to 6 hours after exposure, phones to detect noise levels, but it uses “an arti- Richman (387) believes the tests would be more ficial head or ear” instead of a human subject. Al- useful for prevention of hearing loss. Testing only though it is attractive to manufacture of hearing at 14 hours after exposure, he feels, leaves the tests protection devices because it would be convenient unable to predict permanent hearing loss before for quality control, it has proved to be very dif- it can be prevented. ficult to simulate a human head.

Methods of Audiometric Testing Acceptance of Hearing Protectors Riko and Alberti (389) briefly describe the three classes of audiometric testing, which differ in the OSHA’s noise standard and its Hearing Conser- amount of reliance they place on the human sense vation Amendment (625, 640) require employers of hearing as opposed to the ability of instruments to provide hearing protection to their employees to detect noise. and ensure that the protectors are worn. Compli- ance with requirements to wear hearing protec- Real-Ear Attenuation at Threshold.—This tors is undermined by dissatisfaction with the de- method depends on a person’s report of being able vices. Workers often object to wearing earmuffs to hear sounds. It resembles the classic “hearing in hot, humid conditions, and earplugs are so un- test” of a nurse or doctor whispering across the comfortable that some workers reject them. Ad- room and asking the patient if the whisper can ditionally, earplugs can contribute to infections be heard. In the workplace and in clinics, the real- in dusty, dirty environments, and earmuffs can- ear attenuation at threshold (REAL) test involves not be worn with glasses. placing someone in an acoustically quiet room and using earphones to generate defined levels of noise Furthermore, although some hearing protectors at various frequencies. The worker then puts on are more comfortable than others, they all work hearing protection, and the machine generates by creating a physical barrier to the passage of noises at the same frequencies. The difference in sound, The seal between the protector and the the intensity of noise that is heard with and with- worker’s head or ear is of great importance to the out the protectors is a measure of the protection effectiveness of the protector. In general the seal afforded. is created by pressure. The pressure creates discomfort and, for some workers, pain. The REAL test is used more commonly than the other two tests (“semiobjective” and “objective”) An interesting study (686) demonstrated that described here. The audiometric tests used in hear- quick feedback about the value of hearing pro- ing conservation programs are similar; the lowest tectors promoted their use. Hearing tests were intensity of sound that a worker can hear is deter- administered to some members of a metal fabrica- mined and recorded at the time of first employ- tion department at the beginning and end of their ment. Subsequently—at the time of an annual workshifts on 2 days over a l-month period. Ex- physical examination, perhaps—another REAL posure to workplace noise reduced aural acuity test is administered. sufficiently that pre- and post-work hearing tests differed when hearing protection was not worn. —This method eliminates the Semiobjective. The people providing the tests discussed the im- complete dependence of the REAL method on a plications of the results with workers, who con- person’s hearing, which is, to some extent, sub- gregated around the results that were posted in jective. In the semiobjective test, a tiny micro- the department, until everyone understood the phone is positioned in the ear and the noise energy meaning of the test results. is measured with and without a protector in place. The major drawback of this test is that it is vir- According to the authors, workers’ apprecia- tually impossible to evaluate any protectors that tion of the value of the protectors resulted in a are inserted into the ears. change in accepted behavior; 5 months after the Ch. 8—Personal Protective Equipment ● 157 program began, 85 to 90 percent of the workers in comparison with the others, regardless of which in the department wore their hearing protectors. device was being tested, The authors felt “this fact In contrast, only 10 percent of workers in another, suggests the likelihood of a systematic bias in the equally noisy, department who received a “stand- testing procedure. ” The paper explores the influ- ard lecture” about hearing conservation protec- ence of proper or improper fitting of the hearing tion wore hearing protection. A subsequent at- protector, subject selection and training, and data tempt to discipline workers in that department reduction techniques in producing the variability who did not wear hearing protectors failed be- among the laboratory measurements. cause of both union and management resistance. Perhaps the most striking finding was that Union members objected because the disciplinary action (removing workers from the noisy depart- The labeled NRRs for all four devices were ment) reduced the workers’ earnings, while man- based upon data from Laboratory 8, and even agement did not like losing the services of experi- though in some cases they were derated by the enced workers. manufacturers, all devices would have failed an EPA compliance audit test conducted at any one of the other seven facilities. This would have re- Noise Reduction Ratings quired relabeling. Since there now will be no enforcement of the labeling regulation [because of As a result of the Noise Control Act of 1972, EPA budget cutbacks], it is likely that manufac- the Environmental Protection Agency (EPA) be- turer’s data will continue to reflect the highest ginning in 1979 required that all hearing protec- measurable values found today. tors be labeled with a noise reduction rating (NRR). The NRR is a single number that describes the attenuation of noise that can be expected from OSHA Use of Noise Reduction Ratings wearing the protectors. NRRs are used in connec- One of the authors of the paper that compared tion with measured noise levels in the workplace NRRs measured in different laboratories presented (640) to select appropriate hearing protectors. similar evidence to OSHA during a hearing on As discussed in the beginning of this chapter, the Hearing Conservation Amendment (48). In all models of each class of respirator receive the addition, he said that the laboratory that reported same protection factor, and only extrapolated PFs the highest NRRs was responsible for “85 percent are available for some classes of respirators of manufacturers’ reported NRRs. ” He also ex- (219,580). In contrast, EPA required that each pressed the opinion of his company (E.A.R. Divi- model of hearing protector be labeled with its own sion of the Cabot Corp. ) that the NRRs should NRR. be “de-rated” for two reasons: the generally high NRRs determined by laboratory 8, and the sub- Problems With Noise Reduction Ratings stantial difference between results of laboratory and field testing of hearing protectors. The importance of accurately determining NRRs was recognized by EPA, and that agency con- OSHA mentions this testimony in its latest pre- ducted a study to determine if NRRs were in amble to the Hearing Conservation Amendment agreement by having four models of hearing pro- (640), and says that it will consider it in any tectors tested in seven laboratories. The NRRs re- modification of the noise standard. In a 1983 ported on the labels of the four models tested had directive, OSHA has instructed its inspectors not all been determined in a single laboratory, which to cite exposures between 90 and 100 dB as viola- was referred to as laboratory 8 in the study. Al- tions of the hearing conservation standard pro- though EPA was unable to analyze the results be- vided that the employees are wearing adequate cause of budget cutbacks, three employees of two hearing protection. “Adequate” was described as private-sector companies did analyze the study (50). the NRR de-rated by 50 percent. The various laboratories differed quite consist- The situation regarding noise reduction ratings ently in measuring NRRs. That is, each labora- needs correction, but there appears to be little tory tended to report low, medium, or high NRRs, chance of that happening. A published report has 158 . Preventing Illness and Injury in the Workplace

presented evidence that the NRRs that appear on of “earplug” hearing protectors received only 3 package labels are higher than the NRRs deter- dB protection. mined in seven of the eight laboratories equipped These results are consistent with the findings to carry out the appropriate tests (so). A com- of Riko and Alberti (389), who measured the at- pany that manufactures hearing protectors has tenuation of noise achieved by 400 workers using written OSHA that the effectiveness of its prod- their own protective devices. Although they do uct is overstated and that the protection offered not present numerical results, they state by the protectors in the field is lower than that obtained in the laboratory. The Agency (EPA) re- . . . for both muffs and plugs, the average at- sponsible for ensuring that NRRs are assigned to tenuation was less than that expected from man- hearing protectors and that they are accurate has ufacturer specifications and was considerably less discontinued its NRR program because of bud- than their theoretical potential. Mean attenuation get cutbacks and evidently will not change existing was not as revealing, however, as the scatter of NRRs. OSHA has instructed inspectors that the attenuation scores, which was wide. For given frequencies, values ranging from O to 50 dB between NRRs are in error, but OSHA’s instructions to individuals were obtained. The use of standard employers about selecting appropriate hearing deviation measurement was avoided deliberately protection (640) continue to rely on NRRs that because the distribution of attenuation was not are almost certainly too high. sufficiently uniform and was skewed in the direction of poorer attenuation values. Field Testing of Hearing Protectors Riko and Alberti’s comments about the scatter of attenuation are often repeated in discussing Poor agreement between laboratory and field hearing protection. Although some workers ob- tests was demonstrated by a NIOSH test of 420 tain satisfactory protection, others obtain none workers at 15 industrial plants (267). Fifty per- or hardly any. In some cases, the hearing protec- cent of the workers achieved considerably less tion simply doesn’t fit or doesn’t work. The solu- than half the potential protection demonstrated tion generally suggested for that problem is to of- in the laboratory (table 8-2). Also shown in the fer the worker a variety of protectors so that table is the level of protection achieved by the 10 comfortable ones can be selected; the worker’s percent of workers who received the least protec- selection should be checked by an audiometric test tion. At least 10 percent of the users of preformed to be certain that it provides hearing protection. types of hearing protectors received no protection; Another solution is to reduce noise exposures at least 10 percent of the users of the other types through engineering controls.

Table 8-2.—Comparison of Noise Reduction Achieved by Insert-Type Hearing Protector In the Laboratory and in Field Use

a Median noise reduction ratings Protection obtained by Type of protector laboratory field lowest 10°/0 in fieldb All earplugs...... 28 13 — Preformed types...... 29 7 0 Acoustic Wool ...... 26 10 3 Custom-molded c ...... 20 3 Acoustic foam ...... 36 : 3 %Aeaeurementa in dB. bAvarage noise reduction mhlevm by tha 1(I percent of workera who obtained tha pooreet nOISe ProtOCtlon. cln one plent, the Custorn.rnolded earpiuga wera fabricated by the plant nurse; In the other PlOnt, thay were fabricated by the manufacturer. SOURCE: (267).

Ch. 8—Persona/ Protective Equipment . 159

OTHER TYPES OF PERSONAL PROTECTIVE EQUIPMENT

NIOSH tests and certifies respirators, but no The principal finding from the NIOSH tests was other personal protective equipment. Formerly, that many items of personal protective equipment NIOSH also tested and certified several types of did not function as expected, given the ANSI measuring equipment, such as gas detector tubes, specifications (see table 8-3 for a summary of the coal-mine-dust personal sampler units, and indus- data in this section). trial sound-level meters, but these programs were discontinued in 1983. Head Protection According to NIOSH, some equipment is ad- Hard Hats.—Protective headwear, all of which vertised as meeting “Federal” or “OSHA” stand- is designed to protect against impacts, is classified ards. In fact, there are no general Federal or as A, B, or C depending on its resistance to trans- OSHA standards for personal protective equip- mitting electricity. Class A requires limited elec- ment. In some regulations OSHA requires that trical protection, Class B requires higher electri- safety equipment be worn and stipulates that the cal protection, and Class C requires no electrical equipment meet certain requirements, generally protection. (Reviewers of drafts of this report re- ANSI standards. NIOSH concluded that the “Fed- marked on the classification that placed the most eral” or “OSHA” standard statements arise from protective units in the middle of an A-B-C classi- those requirements. fication scheme. ) Personal protective equipment other than res- NIOSH tested Class B industrial helmets “be- pirators is “self-certified” by the manufacturers, cause they, as a class, offer the most comprehen- who may test it to determine if it meets ANSI sive head protection available to the industrial standards. Equipment that conforms to the stand- worker.” The tests revealed a distressingly high ard can be so advertised. Since few purchasers failure rate. Only 4 of the 21 tested models passed have the facilities, professional staff, and other all the ANSI performance tests. Only 7 passed the resources to test safety equipment, knowing that impact resistance test; 16, the electrical resistance a device meets ANSI standards should provide test. Hats that failed the impact test were found assurance to the purchaser and user that the equip- to transmit too much force, and those that failed ment will provide a specified level of protection. the electrical test did not insulate as well as claimed. NIOSH Tests

In the mid-1970s NIOSH purchased samples of personal protective equipment that was advertised and sold as meeting ANSI standards. NIOSH tested the samples to determine if they did, in fact, meet the appropriate standard. In several cases NIOSH noted some ambiguity in the ANSI standards. Because of the possibility of various interpretations of some standards, NIOSH called together representatives of manufacturers, labor unions, and Federal agencies to discuss the test procedures to be used in evaluating the particular classes of equipment. In all cases, NIOSH tested equipment against the ANSI standard, sometimes including modifications made after Photo credit OSHA, Office of Information and Consumer Affairs the meeting; the agency did not draw up its own Many employees are now required to wear hardhats requirements. during the workday 160 . Preventing illness and Injury in the Workplace

Table 8-3.--NIOSH Testing of Various Types of Personal Protection Equipment

Number (percent) Equipment type (reference) Number of models tested meeting ANSI performance standards Head protection: Class B industrial helmets (“hard hats” with 21 (“randomly selected”) 4 (19°/0) met all performance standards highest electrical resistance) (Cook and 7 (33%) passed impact resistance test Groce (1 16)) 16 (76%) electrical resistance test 20 (95%) passed crown clearance test all passed penetration resistance test

Miners’ safety caps (Cook and Love (120)) 16 (all available models) 13 (81 ‘/0) met all performance standards 14 (88°/0) passed electrical resistance test 15 (94°/0) passed impact resistance test Firefighters’ helmets (Cook (117)) 8 (6 advertised as meeting for the 6 advertised as meeting ANSI ANSI standard) standard, 4 met aII performance standards and passed penetration resistance, electrical resistance, and self-extinguishing tests Eye protection: Glass piano safety spectacles (Campbell and 22 (1 model from each 21 (95°/0) passed Impact resistance test Collins (92)) manufacturer) (only 1 of 24 samples fractured of the model that failed) 21 (95°/0) passed frame impact test 18 (82°/0) passed flammability tests (failures were by small margin, judged not to be major) all passed optical quality tests Plastic piano safety spectacles (Collins and 17 (1 model from each all passed impact resistance test (16 of 17 Wolfe (113)) manufacturer) passed test at 5 times impact energy required by ANSI) all passed frame and lens penetration resistance tests 7 (41 ?/0) passed flammability tests (failures judged to present little danger in workplace) all passed optical quality tests Flexible fitting safety goggles (Campbell and 50 (all available clear-lens all passed impact resistance tests Collins (92)) models) all passed penetration resistance tests 48 (96%) passed “design test” that estimated risk from particles entering inside goggles through ventilation opening 32 (64°/0) passed flammability test 33 (66%) did not meet ANSI optical standards Eyecup goggles (Campbell, Collins, and 24 (1 model from each 13 (54%) passed impact resistance test Wolfe (93)) manufacturer) 16 (67°/0) passed frame impact test all passed flammability test all passed optical transmittance test Welding filter plates (Campbell (91)) 94 (all available shade 19 (20%) met all performance standards models) more than 90% passed ultraviolet, infrared, and impact tests (when appropriate) 76 (80°/0) passed visible-light transmittance tests Face protection: Industrial face shieids (Campbell (90)) 37 (“representative 36 (97°/0) passed impact resistance test sample”) 36 (97°/0) passed penetration resistance test ail passed flammability test Hand protection: Linemen’s rubber insulating gloves (Cook 12 (randomly selected and 11 (92%) passed electrical resistance test and Fletcher (119)) representative of 155 (model that failed electrical test was available models) withdrawn from market by manufacturer) 10 (84%) passed tensile strength

Ch. 8—Personal Protective Equipment ● 161

Table 8-3.—NIOSH Testing of Various Types of Personal Protection Equipment—Continued

Number (percent) Equipment type (reference) Number of models tested meeting ANSI performance standards Foot protection Men’s safety-toe footwear (Cook (118)) 76 (random samples of ANSI analysis:a types in general use (71), 49 (55°/0) passed impact resistance tests plus styles not 60 (79°/0) passed compression tests represented in general 43 (37°/0) passed overall tests samples (5)) Statistical analysis: 28 (37°/0) passed impact resistance tests 50 (60°/0) passed compression tests 36-50 (48°/0-600/0) passed overall tests confidence level that 9 out of 10 shoes would ANSI test (see text) SOURCE Office of Technology Assessment from cited references

The impact resistance test is conducted for vertical impacts only. A member of the advisory panel to this OTA assessment reports that NIOSH also conducted nonvertical impact tests—with disastrous results. Those results were not reported by NIOSH. Miners’ Helmets. —Miners are required to wear helmets in underground mines and in surface mines where falling objects may create hazards. NIOSH characterized 15 of the 16 models of miners’ helmets as industrial helmets to which a lamp and a cord bracket for mounting it have been added. Miners’ helmets, like “hard hats, ” can be Class A, B, or C, depending upon their resis- Photo credit: E. /. du Pont de Nemours & Co. tance to electricity. These firefighters are using protective clothing, firefighters’ helmets, and self-contained The miners’ safety caps had better test results breathing apparatus than the Class B hard hats: 15 of 16 models passed providing protection against impacts and electri- the impact resistance test; 14, the electrical resis- city, firefighters’ helmets must also be self- tance test. Manufacturers had drilled holes for use extinguishing. NIOSH used two different meth- in attaching lamps to the 2 models that failed the ods to measure the transmission of impact forces electrical tests; electrical shorts across the holes in testing firefighters’ helmets. An analysis showed caused those models to fail. Overall, 13 models that the method they favored produced more con- met all the ANSI performance standards. sistent results; using that test, four of the six Oddly, of the five companies making both models advertised as meeting the ANSI standard miners’ helmets and Class B hard hats, four pro- passed the impact resistance test. All six models duced a miners’ helmet that passed the the impact passed the tests for electrical resistance and resistance test and at least one model of Class B penetration resistance, and all were self-extin- hard hat that failed it. By contrast, the producer guishing. NIOSH concluded that the two helmets of the miners’ helmet that failed the impact resis- advertised as meeting ANSI standards that failed tance test made a Class B hard hat that passed. the impact resistance test suffered from poor quality control in the manufacture of the suspen- Firefighters’ Helmets. -Firefighters’ helmets dif- sion systems. fer in shape from hard hats and miners’ helmets, and they have a broad brim to carry water away The agency suggested that the tests for these from the wearer’s face and neck. In addition to devices would more accurately reflect firefighters’ 162 ● Preventing Illness and Injury in the Workplace needs if an impact resistance test at 300o Cen- Protective Eyewear 0 tigrade (572 Fahrenheit) were added to the ANSI Protective eyewear (fig. 8-3) is the most com- standards. The severity of such a test reflects the monly used personal equipment. In the work- conditions that firefighters and their equipment place, this equipment provides protection against face. particles, sparks, and chemicals that might hit the Comments on Head Protection.—The tests eye and/or protection against harmful ultraviolet, showed significant performance variations from infrared, and too-intense visible light. ANSI ANSI standards. The highest failure rates were specifies that protective eyewear must not affect found in Class B hard hats, which are used in the visual acuity to the point where a worker’s per- greatest numbers and manufactured by the largest formance is impaired and certainly not to the number of firms. The miners’ helmets, as a class, point where the worker’s safety is affected by re- performed very well. The single failure in the im- duced field or clarity of vision. pact resistance test was associated with quality An obvious concern in safety eyewear is that control during manufacture, and the electrical test the eyewear itself can present a hazard if it fails. failures were related to holes being drilled in the A particle splintering the lens of spectacles may helmets. be the cause of an accident, but the injury may Overall, the performance of protective head- stem from fragments of the lens entering the eye. wear in tests completed in 1976 and 1977 does not Since 1972, all spectacles sold in the United States produce confidence that the devices work as must have impact-resistant lenses, but the level claimed. Furthermore, NIOSH drew attention to of impact resistance in “street-wear” spectacles is the absence of tests of impact resistance for the far below that required of safety eyewear used in front, sides, and rear of helmets, highlighting what the workplace. To facilitate making a distinction appears to be a deficiency in the tests. In the re- between lenses intended for street wear and those port on miners’ helmets, NIOSH noted that the for industrial use, all the latter spectacles are present ANSI standards do not consider the ef- marked with the manufacturer’s name. fects of projections into the interior of some Piano Safety Spectacles.—Glass (or plastic) helmet models, which would tend to concentrate piano safety spectacles (“piano” means flat, non- the force of any impact from the front, sides, or corrective lens) are the “safety glasses” with which rear. After recommending that models with pro- most people are familiar. They are intended to trusions be avoided, NIOSH suggests a crude test be a barrier between the eye and foreign objects to identify them: and not to interfere with the wearer’s vision. With the helmet in place, a cautious series of blows to the perimeter of the helmet is sufficient NIOSH tested 22 models of glass safety spec- to identify models with these protrusions. tacles, one from each manufacturer. All but one model passed the impact resistance test prescribed The report on Class B helmets acknowledged by ANSI. Twenty-four samples of each model that “It should be recognized that many lives have were tested, and even in the model that failed only been saved through use of industrial helmets and one lens of the 24 fractured. All the models passed such devices are a valuable adjunct to the over- the frame impact test, which requires the lens to all protection of workers. ” However, apparently remain in the frame after impacts on the side or in recognition of the deficiencies found in those top. helmets, the agency added, “NIOSH intends to promulgate regulations in the immediate future The spectacles were also tested for resistance to establish the legal basis for a testing and cer- to higher energy impacts, although the results of tification program for industrial helmets. ” That this test did not affect NIOSH’S decision on wheth- was 8 years ago. er a lens passed or failed the basic impact test.

Ch. 8–Personal Protective Equipment ● 163

Figure 8.3.—Protective Eyewear

Service 164 . preventing Illness and Injury in the Workplace

NIOSH found that many lenses provided impact pect many glass lenses to fail that test, and that resistance in excess of the ANSI requirements. For they are unaware of why it is not required of glass 5 models, none of the tested samples fractured at lenses. impacts 2.5 times greater than the ANSI stand- Flexible Fitting Safety Goggles.—All the ard, and 467 of the 528 sample lenses exposed to NIOSH-tested flexible fitting safety goggles passed that much impact survived. the impact and penetration tests. Two of the 50 All models passed the optical quality tests, and models suffered from design defects that would NIOSH considered overall optical quality to be allow particles to enter the eye area from the side, above the ANSI requirements. and, as was found with the safety spectacles, the Sideshields, constructed of wire mesh or plastic, plastic used in some of the goggle frames burned that cover the opening between the outside edge at a rate slightly greater than allowed by ANSI. The excess burn rate was not considered to be a of the glasses and the wearer’s face are necessary problem in most workplace situations. The op- to prevent particles from reaching the eye from tical quality of the goggles was poorer than that the side. NIOSH found that four models with of spectacles. plastic sideshields did not pass the ANSI flammability test, but they considered these failures Eyecup Goggles. —Eyecup goggles performed to be of minor importance because they were poorly. Eleven of the 24 models tested failed the barely outside the acceptable range. impact test. All the models that failed used flat The plastic piano safety spectacles tested also lenses, and no models with curved lenses failed. met ANSI standards. All 17 models, one from In addition, the frames of eight models failed (all each manufacturer, passed the lens and frame im- the failures occurred in models that had also failed pact tests. Sixteen of the 17 passed a lens impact the lens impact test). NIOSH concluded that test at 2.5 times the energy level specified by “many models of eyecup goggles were found to ANSI, and NIOSH reported that the manufac- be seriously defective and are considered to rep- turer of the one other model planned to change resent a significant hazard to the user”; unfortu- its lenses to the more energy-resistant plastic. nately, the defects in eyecup goggles “are not de- Overall, plastic lenses provided more impact re- tectable by the user. ” Overall, although some sistance than glass ones. eyecup goggles offered good protection, NIOSH pointed out that safety spectacles and flexible fit- The optical quality of plastic lenses was suffi- ting goggles provided better impact protection. cient to pass the ANSI standard, but below that of glass lenses. An important tradeoff in selection Welding Filter Plates. —The primary purpose of safety spectacles is that glass lenses are more of welding filter plates is to protect the welder’s resistant to abrasion, and they are generally cho- eyes from intense ultraviolet, visible, and infrared sen for use in situations where abrasion is a radiation. Different shades of plates are available, problem. depending on the type of welding and the radiation encountered on the job. NIOSH tested a total Ten models of plastic safety spectacles did not of 94 different shade-models; only 20 percent met pass the flammability test because of sideshields all ANSI performance standards. burning at rates greater than allowed by ANSI. As with the glass models, however, the bum rate Comments on Protective Eyewear. -Testing of was so close to the ANSI limit that NIOSH did protective eyewear showed that spectacles and not consider these failures to be serious. flexible fitting goggles performed well. Eyecup goggles did not, as a class, provide the level of One difference between the ANSI tests for glass protection specified by ANSI. Few welding plates and plastic spectacles is that glass lenses are not met all ANSI standards. tested for resistance to penetration. All plastic lenses passed the lens penetration test specified by It is impossible to tell from the results reported ANSI. NIOSH commented that they would ex- by NIOSH whether the failures were due to de-

Ch. 8—Personal Protective Equipment . 165 sign defects or poor quality control. An excep- Linemen’s Rubber Gloves tion to that generalization is the two models of Only three manufacturers make the 155 models flexible fitting goggles with design defects that of rubber gloves worn by electrical linemen to allowed particles to strike the eye from the side. provide protection against electrical shock. NIOSH tested 12 models considered to be repre- Face Protection sentative of all those available. Eleven models NIOSH tested faceshields that were attached to passed the ANSI specified electrical resistance test. hard hats. All but 1 of 37 models (a “representa- The model that failed was, according to ANSI, tive sample” of those available) passed both the withdrawn from the market by the manufacturer. impact and penetration tests. The same model One of the two models that failed the tensile failed both tests. NIOSH concluded that the other strength test was advertised as meeting the ANSI shields performed in accordance with ANSI stand- standard; the other was not. These two are the ards, but warned, as do several manufacturers, only models not made from rubber and are in- that shields are not a substitute for eye protec- tended for use in areas with very high voltages. tion. Particles can pass around or under the shield NIOSH apparently sees the use of plastic in place and cause injury if eye protection is not worn in of rubber in these gloves as acceptable because conjunction with the face protection. they offer superior service qualities in the specific uses for which they are sold.

Men’s Safety-Toe Footwear Safety-toe footwear (more generally called “steel-toed shoes”) are subjected to two performance tests. The impact resistance test measures the deformation of the toe of the shoe when it is subjected to a single hit with a falling object. The compression test measures the deformation during the application of a steady squeezing force. Safety-toe shoes are rated by the manufacturers on the basis of the shoes being able to withstand an impact of 30, 50, or 75 foot-pounds and an average compressive force of 1,000, 1,750, or 2,500 pounds. The shoes are rated as Class 30, SO, and 75 respectively. NIOSH tested 76 different models, a random sample of those available. The ANSI standard is based on a pass/fail test. NIOSH analyzed its results on this basis as well as on another statistical basis. The statistical analysis considered how close each model came to failing the test, and resulted in an estimate of the 95 percent probability that at least 90 percent of all shoes of that model would pass the test. Using the ANSI standard, 49 of 76 models passed the impact tests; 60, the compression tests; Photo credit OSHA, Office of /n format/on and Consumer Affairs only 43, both tests. The statistical test resulted in Faceshields and safety glasses protect these workers 28 of 76 models passing the impact tests; 50, the from flying particles compression tests; 36-50 the overall tests.

166 . Preventing Illness and Injury in the Workplace

Under both methods of analysis, a significant they were manufactured. Others—safety-toe proportion of the tested shoes did not conform shoes, eyecup goggles, industrial safety helmets, to the ANSI standard. NIOSH concluded that firefighters’ helmets, and welding filter plates— manufacturers overrate the shoes. For instance, often did not measure up to the ANSI standards. a shoe marketed as Class 75 provided only the Furthermore, some deficiencies were found in the protection required of a Class 50 shoe. NIOSH ANSI test standards, which do not measure some also noted, however, that some cases of failure, important properties such as resistance of hard in which the shoe deformed just past the fail mark, hats to off-center impacts or of glass safety glasses might not be associated with an injury in actual to penetration. use, and that safety shoes have afforded a sub- NIOSH has not tested these safety devices since stantial degree of protection to workers. the publication of these reports in the mid-to late- 1970s. Furthermore, several items of personal Comments on NIOSH Testing of safety equipment, notably clothing and chemical- Personal Protective Equipment resistant gloves, were not tested in the NIOSH Some types of safety equipment—notably spec- program. Whether design and quality control now tacles, flexible goggles, miners’ helmets, linemen’s are better, about the same, or worse than when rubber gloves, and face protection—were found NIOSH performed these tests is not known. to conform to the ANSI standard against which Tests of Gloves Against Chemical Hazards Manufacturers, in their literature, rate various types of gloves as providing “excellent, good, fair, or poor” protection against workplace chemicals. There seems to be little reason for the general assignments of resistance to chemicals. As is shown on table 8-4, many chemicals penetrate “chemical-resistant” glove materials quite quickly. Some workplace solvents, for instance halogenated ethanes (dichloroethanes and trichloro- ethanes) and polychlorinated biphenyls (PCBs), have presented special problems because they penetrate most gloves in minutes or seconds. A relatively new material, Vitron, developed by Du Pont, provides excellent resistance to the halogenated ethanes, however, and better resistance to PCBs than any other tested material. This example illustrates how new developments in technology find application in the protective devices industry. Yet Vitron gloves may not be used widely because of the high cost—10 times as much as any other glove. Involvement of Personal Protective Equipment in Injuries

Photo credit: E.I. du Pont de Nemours & Co There are no field tests of personal protective Protective clothing is frequently required for exposures equipment used for preventing injuries. It is prob- to toxic substances ably impossible to design such tests, and only —

Ch. 8—Personal Protective Equipment 167

Table 8-4.—Comparison of Various Protective Garment Materials’ Capacity to Resist Penetration by Chlorinated Ethanes and PCB

Number of minutes required for solvent to penetrate 1/1,000 inch of the protective material 1,2-di- 1,1,1-tri. 1,1 ,2-tri - polychlorinated Protective material chloroethane chloroethane chloroethane biphenyl (PCB) Butyl rubber ...... 6.4 2.7 2.3 0.1 Neoprene rubber latex ...... 0.9 2.0 0.3 0.02 Nitrile rubber latex ...... 0.3 3.8 0.3 0.1 Polyethylene ...... 1.2 1.5 1.8 0.4 Surgical rubber latex...... 0.2 0.05 0.1 0.04 Vitron ...... 82.0 >144.0 >144.0 6.0 SOURCE Adapted from (456) workers with a great deal of confidence in their Selected Data From the BLS equipment would submit to having their hard hats Survey of Head Injuries or safety-toe shoes struck by a heavy weight while The head injuries suffered by workers wearing their heads or feet were inside. In the absence of hard hats were divided about equally among im- such data, it is useful to inspect the Bureau of La- bor Statistics (BLS) reports of accidents involv- pacts sustained on the hard hat, on an unprotected area, and both (600) (table 8-5). The shells of 37 ing different parts of the body. percent of the helmets broke, and the suspensions BLS warns that its data do not allow any con- of 17 percent failed as a result of the accident. An clusions to be drawn about the incidence of in- injury can result, of course, from an impact that juries suffered by workers wearing and not wear- does not damage the helmet. For instance, too ing protective equipment, for no information much force can be transferred if the shell is pushed about exposure is available. For instance, the BLS forcibly onto the worker’s head. The high failure data do not reveal if workers wearing protective rate reflected in shell and suspension breakage eyewear are exposed to more airborne particles parallels NIOSH’s observations that only 33 per- with the potential of harming eyes than are work- cent of tested Class B industrial helmets passed ers who do not wear protective eyewear. Al- the ANSI impact resistance test. though the BLS data cannot reveal how many injuries personal protection devices prevented, they Table 8-5 presents data on the reasons that hard can be used to learn about conditions that caused hats were not worn. The majority of workers who failure of such devices and why personal protec- were unprotected were not supplied with protective devices were not worn. tive equipment, not required to wear it, or thought it was unnecessary. Less than 20 percent The BLS data were collected over time periods of the injured workers who were not wearing head ranging from 2 to 5 months in 19 or 20 States (de- protection said that hard hats were uncomfor- pending on the injury studied). Employers’ reports table, impossible to wear, or interfered with work. of injuries to State workers’ compensation agencies were reviewed, and questionnaires were The percentage of workers usually wearing mailed to workers in selected occupations in all hard hats in their work is very close to the per- industries except mining. In general, the survey centage required to do so. Twenty-one percent of period ended when a certain number of question- all workers who responded to the BLS questionnaires were returned, and the results cannot be naire are required to wear hard hats; 20 percent taken as representative of all injuries affecting the wear hard hats all or most of the time. Ninety- specified part of the body. five percent of the workers wearing hard hats at

168 . Preventing Illness and Injury in the Workplace

Photo credit: Department of Labor, Historical Office

Although materials have changed since this photo was taken, the basic purpose of protective clothing is still to prevent worker contact with harmful substances

Table 8=5.—Selected Information From BLS Survey of Head injuries

Number Percent Workers’ reports of head Injuries while wearing hard hats: Total ...... , ...... 169a Struck on hard hat area only ., . . 53 31 Struck on unprotected part of head only , ...... 60 36 Struck on hard hat area and unprotected part ...... 55 33 Don’t know ...... 1 1 Helmet shell broken or Photo credit: Chemical Manufacturers Association damaged , ...... 40 37 Helmet suspension broken or Worker dons full protective suit damaged ...... 18 17 Workers’ reports of masons for not wearing hard hats: a Total ...... , ...... 852 the time of their accident were required to wear Thought it was not needed . . . . . 216 25 them. Not available from employer . . . . 176 21 Not normally used or not practical ...... 471 55 Selected Data From BLS Survey of Eye Injuries Uncomfortable, did not fit with other equipment, hard to work The most important information for assessing with it on, or in bad the effectiveness of eye protection would be to condition ...... 163 19 know how many injuries were prevented by them. Other ...... 42 5 responses number of Injuries and sum of As with head injuries, however, no data are avail- 100 because multiple responses could be given by single individual. able to calculate that. SOURCE: .

Ch. 8—Personal Protective Equipment . 169

Table 8-6 lists the reasons workers thought pro- eye protection was necessary or that the eye protective eyewear failed to protect them (598). The tection interfered with the worker’s vision. most frequently cited reason was that the object There is no way to be certain that wearing eye or chemical that caused the injury went around protection would have reduced the number of in- or under the protection. NIOSH (94,133) has drawn attention to this possibility in reports on juries to unprotected workers, but it is a safe assumption that at least some injuries would have piano safety spectacles (which should be equipped been prevented and that the severity of others with side shields) and on face protection (which would have been reduced. Apparently greater use should not be used without eye protection). of protective eyewear would result from greater Despite all the caveats that must be attached supervisor attention and improved designs to re- to conclusions about the effectiveness of protec- duce interference with workers’ vision. Seventy- tive eyewear, it can be concluded that few devices nine percent of those wearing eye protection were failed because of characteristics for which ANSI required to; only 52 percent of all workers were has testing standards. Only 4 percent of the in- required to. juries involved lens failures, and only 1 percent Just over half (56 percent) of the injured work- were related to frame breakage. ers who were wearing eye protection thought that The reasons given by injured unprotected work- it reduced the severity of their injuries. Five per- ers for not wearing eye protection are tabulated cent thought that the protection contributed to in table 8-6. Twenty-two percent reported that pro- the injury. The remaining 39 percent did not have tective eyewear was unavailable at the work site. an opinion about the effect of the eye protection The other reasons given were either that the or thought that it had had no effect. worker or the worker’s supervisor did not think Selected Data From the BLS Table 8-6.—Selected Information From BLS Survey of Face Injuries Survey of Eye injuries Only 9 of 770 workers who were included in Number Percent the survey of face injuries reported that they had Workers’ reports of reasons Injury occurred when eye been wearing face shields when they were hurt protection was worn: Total ...... 401a (599). However, about a third of the injured work- Object or chemical went under ers had been wearing eye protection; about 20 per- or around protection ...... 376 94 cent of them thought the eye protection minimized Object went through lens or the their injuries, and about 10 percent reported facial shattered lens hit eye; lens injuries from broken frames or lenses. was knocked out of frame . . . 15 4 Frame broke and injured Most workers who did not wear face protec- worker . . . . , ...... 4 1 Eye or face protection slid or fell tion reported that it was not required or was not out of place ...... 16 4 considered necessary. Only about 10 percent re- Other ...... 28 7 ported that it was uncomfortable or interfered - Workers’ reports of reasons for not wearing eye protection. with vision. Total ...... 612 Eye or face protection lifted up; The one face shield that failed was split by not in place ...... 37 6 None available, not required, or fragments of an exploding cutting wheel, which worker thought none needed or caused multiple fractures. In five cases, the ob- not normally used or 136 22 ject or chemical that caused injury went around impractical ...... , . . . 402 66 Protection device reduced vision or under the face shield. or device fogged up, or device was uncomfortable or in bad Selected Data From the BLS condition ...... 293 39 Survey of Foot Injuries Other ...... 63 10 asorne resporlses exceed total number of Injuries and Sum Of Percentages exceeds 100 because mult Iple responses could be given by single individual Workers only infrequently wear safety foot- SOURCE (598) wear unless required by employers; “fewer than 170 Preventing Illness and Injury in the Workplace a tenth of those not required to use foot protec- The Industrial Safety Equipment Association tion were wearing safety shoes” (601). Eighty-five is the trade association of manufacturers of per- percent of the workers wearing safety shoes were sonal protective equipment. According to its presi- injured in an unprotected part of the foot. Impor- dent (672), the association had supported the ex- tantly, given the ANSI emphasis on testing the tension of NIOSH certification programs from strength of the safety toe, only 7 percent of foot respirators to other personal protective equip- injury accidents reported to BLS involved safety ment. In 1980, the association, upon deciding that toe failures. NIOSH was unlikely to be able to expand its certification program, established as a separate en- As noted earlier, NIOSH (117) observed that tity the Safety Equipment Institute. manufacturers sometimes exaggerated the impact or compression resistance of the safety toe. Evi- SEI is a testing and certification organization. dently users of safety shoes are seldom aware that It does not develop its own test standards; instead, the shoes are rated on these features; 82 percent as NIOSH did in the late 1970s, it tests equipment of workers wearing safety shoes did not know against the ANSI standards. The tests are carried what class of protection the shoes provided. Fifty- out in laboratories under contract to SEI. The In- seven percent of workers wearing safety shoes stitute so far has tested hard hats made by 95 per- thought that the severity of their injuries had been cent of the manufacturers and eye and face pro- reduced; only 1 percent thought that the shoes tection made by 65 percent of the producers, and contributed to injury. it makes available lists of hard hats and eye and face protection that passed the ANSI standards Possible Conclusions From the BLS Surveys and were certified. Currently it is testing emergency eyewash and shower facilities. The results of the BLS Surveys are consistent with the conclusion that the equipment works-in Following the testing phase of the SEI certifica- that the failures the ANSI standards are to guard tion procedure, the Institute, through an inde- against occurred infrequently except in the case pendent consulting firm, arranges for quality as- of hard hats, which failed relatively often. That surance audits of the manufactures of certified conclusion is necessarily limited. For instance, if equipment on a biannual basis. In addition, at 6- the 7 percent of safety-toe shoes that failed on the or 12-month intervals, SEI retests a number of each job did so because they received an impact of certified model of personal protective equipment. much greater force than the shoes were designed Upon retesting hard hats, SEI found that all the to withstand, then the shoes performed up to the tested green-colored hats of one model made by standard. On the other hand, if some fraction of one manufacturer did not meet the ANSI standards. the 7 percent failed at an impact less than they Upon being notified of this fact, the manufacturer were designed to resist, then the 7 percent failure withdrew that lot of green hats from the market rate is an underestimate. and informed all distributors that had purchased them that the hats were below ANSI standards. Third= Party Laboratory Testing of SEI has not been greeted by everyone as an in- Personal Protective Equipment dependent source of information about the effectiveness of personal protective equipment. Spring- Manufacturers of personal protection devices ing as it did from the trade association, it is seen are not required to test their products. They may by some as under the control of the manufactur- do tests to assure themselves that the products ers. The Institute has partially addressed this crit- meet ANSI standards. However, the results of the icism by establishing a board of directors that has (now dated) NIOSH evaluation of personal pro- only one -member from the trade association. Its tective equipment against ANSI standards pro- bylaws also separate it from the trade association. vides no assurance that such tests are always done or that quality control is sufficient to guarantee Although time will have to pass before the suc- that products coming off the assembly line meet cess of SEI’s program can be evaluated, the pro- the standards. gram is now well under way, and it offers third- Ch. 8—Personal Protective Equipment 171

party certification of personal protective equip- purchaser of certified personal protective equipment. SEI points out that it offers both an alter- ment that the equipment meets national consen- native to self-certification and assurance to the sus standards for performance.

SUMMARY

The Federal Government does not certify that ing workers’ hearing periodically to determine if any types of personal protective equipment, ex- hearing acuity has dropped below prescribed lev- cept respirators, work. In the case of some other els. All hearing protectors are labeled with a Noise items of personal protective equipment, the Amer- Reduction Rating, which is supposed to give the ican National Standards Institute has drawn up purchaser information about the amount of pro- standards, and manufacturers advertise that their tection provided by the protector. products meet those standards. There is almost universal agreement, and no The procedures used by NIOSH to certify res- evidence to the contrary, that the NRRs overstate pirators are dated, and there is a great deal of dis- the amount of protection afforded by hearing pro- satisfaction with them. Even devices that pass the tectors. NRRs are required by a law administered tests may not work well in the field, and since by the Environmental Protection Agency; because there is only a pass/fail evaluation, better respi- of budgetary cutbacks, however, EPA is no longer rators receive the same grade as ones that barely monitoring the accuracy of NRRs. Therefore, meet the standards. products bearing a Government-approved effi- Few studies have measured the effectiveness of ciency rating are known to be overrated, and that respirators in the workplace. High effectiveness situation is unlikely to change. has been found only in workplaces that have well- developed respirator programs with careful main- Field testing of hearing protection has yielded evidence that noise attenuation achieved in the tenance, education, and supervision. workplace is much less than that expected from NIOSH is currently drafting a revision of the the NRRs. Some workers, even though they wore certification regulations, which reportedly will rehearing protectors, received no benefit in terms quire that manufacturers test and certify their own of noise reduction. On the other hand, a limited products. Not everyone supports self-certification. study of hearing acuity in a major chemical com- Labor unions are in favor of the government pany showed that hearing losses among its work- maintaining a role in certification. An alternative ers over a 5-year period were not related to noise offered by some manufacturers is the establish- levels. Additional, longer studies are necessary, ment of a testing laboratory, supported by indus- however, before the success of these programs can try fees, to carry out the certification tests. In- be fairly judged. dustry is not uniformly behind the idea of a third-party laboratory, however, because the fa- Analogous to the situation with respirators, the cility, as proposed, would have ties to some man- usefulness of hearing protectors depends on how ufacturers. To opponents of the third-party lab- they are chosen and used. A continuous program oratory, those ties mean that it too would be a of instruction, supervision, and maintenance is method of self-certification, and they favor necessary. straightforward self-certification. In the late 1970s, NIOSH tested several types Hearing conservation programs depend on of personal protection equipment against the measuring the level of noise in the workplace, ANSI standards. Since almost all the equipment lowering it when possible by engineering controls, was manufactured by companies that claimed ad- providing hearing protectors to workers exposed herence to ANSI standards, the NIOSH tests were to noise above certain specified levels, and check- a measure of the quality assurance programs of 172 ● Preventing Illness and Injury in the Workplace the manufacturers or of their ability to carry out The Bureau of Labor Statistics has collected the ANSI tests. questionnaire data about some industrial accidents. Their survey data about head injuries Some types of safety equipment, notably spec- showed that hard hats frequently failed in acci- tacles, flexible goggles, miners’ helmets, linemen’s dents, which fits with NIOSH’s findings that many rubber gloves, and face protection, were found hard hats did not meet ANSI standards. Other to conform to the ANSI standard against which Bureau reports found that other types of personal they were manufactured. Others—safety-toe protective equipment also failed under certain, shoes, eyecup goggles, industrial safety helmets sometimes very severe, conditions. However, pro- (hard hats), firefighters’ helmets, and welding filter tective equipment was seldom identified as the plates—often did not measure up to the ANSI cause of an accident. Although it is impossible to standards. Furthermore, some deficiencies were know how many injuries were prevented by per- found in the ANSI test standards, which do not sonal protective equipment, it is clear that its use assess some important properties such as resis- depends on good design and supervision of its use. tance of hard hats to off-center impacts or of glass safety glasses to penetration. 9 Hierarchy of Controls Contents

. Page 175 175 176 177 177 178 179 ● ☛☛✎ ✎☛ Q.. *.** **. ,.*, ****C 179

. . * *..,.*... .*. .., ... * * ec 181 ● * *...*.... ● . . . . ● **.*,... 1$2 ● ☛☛✎ *O. *.. ..ee*c*, ceo . ● ● ● ~ 183 ...... ,.*,.* - “ “ “ “ ● ● ● ● ● ● ● ● ● 285

b. . . Hierarchy of Controls

As explained in chapter 5, a generalized model tive equipment, work practices, and administra- of the workplace environment looks at sources tive procedures (see discussions in chs. 5 and 8). of hazards, transmission of the hazard, and work- This model can also be applied to injury haz- ers (see fig. 5-2 in ch. 5). Health and safety hazards are controlled by interrupting the transmission of ards, although much of the safety research published to date has followed other approaches. For hazardous agents to workers. Controls can be intro- example, control of electrical energy at the source duced at several points: the source of the hazard, might involve grounding to prevent the buildup the workplace atmosphere (or transmission points), and inappropriate release of hazardous amounts and the workers’ location(s). For health hazards, of energy. Control of transmission could include control at the source can include substitution of separating workers from the hazard by, for exam- less toxic agents or substances, as well as process ple, placing physical barriers or guards between and engineering changes to reduce emissions of the worker and the hazard, Personal protective the hazardous agents or substances. Control of equipment, such as insulated gloves, represents transmission of the agent can be accomplished by a control applied on the worker. ventilation, isolation, dilution, or enclosure. Con- trol at the worker can include personal protec-

DESCRIPTION OF HIERARCHY OF CONTROLS Health and safety professionals have tradi- cluded and generally constitute the “third” line of tionally ranked controls according to their relia- defense, falling after engineering controls and bility and efficacy in removing or controlling haz- work practice controls and ahead of protective ards. Put simply, the principle of the hierarchy equipment. For some hazards, however, the Oc- of controls is to control the hazard as close to the cupational Safety and Health Administration source as possible. (This hierarchical approach is (OSHA) places administrative controls on an most commonly discussed in relation to health equal basis with engineering controls. In addition, hazards and the methods of industrial hygiene. OSHA now usually groups engineering controls Although the principle can also be applied to safe- and work practices together and assigns them the ty hazards, most of the discussion in this chapter same priority. Nevertheless, in all cases, personal will focus on the control of health hazards. ) protective equipment is listed as the control of last resort. Expressed differently, the hierarchy of controls describes the order that either should be followed or must be followed when choosing among op- Views of Health Professionals tions for controlling health and safety hazards. In its simplest form, the hierarchy of controls The hierarchy of controls is widely supported specifies that engineering controls (including sub- in the professional community. Every current institution, enclosure, isolation and ventilation) are dustrial hygiene textbook endorses the idea of preferred to the use of personal protective equip- such a hierarchy and lists engineering controls as ment. Work practices are frequently added to this the first priority and personal protective equip- list between engineering controls and personal ment as a last resort (455). It is often expressed protective equipment. Administrative controls, in the context of controlling exposures to airborne such as worker rotation, are also oftentimes in- contaminants-fumes, dusts, and vapors-that

175 176 ● Preventing Illness and Injury in the Workplace

may enter the worker’s respiratory system. Elim- Consensus Standards ination of the contaminants by substitution of materials, enclosure of operations that generate The hierarchy of controls is also found in the fumes and vapors, dust suppression methods, or “consensus” standards written by committees dilution of the contaminants by ventilation are meeting under the auspices of the American Na- all preferred over reliance on respirators. Leaders tional Standards Institute (ANSI) and its prede- in this field and industrial hygiene texts all agree cessor, the American Standards Association. The on this point (129,173,199,362,369,423). Association’s 1938 standard for protective equipment did not require any particular “hierarchy of Excerpts from a 1947 textbook on industrial hy- controls, ” but the attached commentary noted: giene and from a 1980s text underline the un- changing preference for engineering controls. It is obviously better to remove the hazard, when this is possible, than to protect the worker Obviously, the most successful approach to the against it [using personal protective equipment]. problem of industrial atmospheric sanitation lies Thus, in granite cutting it is preferable to remove in the design or alteration of plant and equipment the dust by an exhaust system rather than to allow so that the control features are engineered into the it to contaminate the air, and then to protect the structure and machinery (71, emphasis in original). worker against breathing the dust. . . . This code In many instances, however, the reduction of ex- does not attempt to specify how various indus- posure [through engineering methods] is not suffi- tries shall be conducted, with respect to avoiding cient to eliminate the hazard, and other control hazards, but points out the method of protecting measures are needed. . , . Respiratory protective de- the worker where the hazard has not been elimi- vices have a distinct place in the field of industrial nated by other means (15). health engineering. That they are a last Zinc of de- In the 1959 revision of this consensus standard, fense can hardly be denied (71, emphasis added). respirators were assigned a supplemental function The newer text states: in the actual text of the standard: If confinement or removal by adequate proper- General Considerations. Respirators are used ly engineered and operated ventilation systems is to supplement other methods of control of air- not possible, personal protective equipment on a borne contaminants rather than to substitute for temporary basis should be considered. We stress them. Every effort should be made to prevent the the word “temporary” since respiratory protec- dissemination of contaminants into the breathing tion can seldom be relied on for long periods of zones of the workers. In some instances, it is nec- time in hazardous exposures, unless highly unu- essary to use respirators only until these control sual control procedures are established and rig- measures have been taken; in others, such meas- orously enforced (172). ures are impracticable, and the continued use of respirators is necessary (16). In 1963, the American Conference of Govern- mental Industrial Hygienists (ACGIH) and the In 1969, ANSI issued a separate standard for American Industrial Hygiene Association jointly respiratory protection. The heading of the para- issued a comprehensive guide to respiratory pro- graph that describes the applicable principles was tection in the United States. The opening para- changed from “General Considerations” to “Per- graph is clear on the preferred methods of con- missible Practice,” adding weight to the impor- trolling occupational hazards: tance of the control hierarchy. In addition, emphasis was given to what was now viewed as the In the control of those occupational diseases “primary objective” of workplace controls: pre- caused by breathing air contaminated with harm- venting the contamination of the workplace at- ful dusts, fumes, mists, gases, or vapors, the pri- mosphere. In addition the word “feasible” was mary objective should be to prevent the air from used in connection with controls. becoming contaminated. This is accomplished as far as possible by accepted engineering control Permissible Practice. In the control of those oc- measures; . . . (9). cupational diseases caused by breathing air con- . .——. ...—

Ch. 9—Hierarchy of Controls ● 177

taminated with harmful dusts, fogs, fumes, mists, other protective measures shall be used to keep gases, smokes, sprays, or vapors, the primary ob- the exposure of employees to air contaminants jective shall be to prevent atmospheric contamina- within the limits prescribed . . . Any equipment tion. This shall be accomplished as far as feasible and/or technical measures used for this purpose by accepted engineering control measures (for ex- must be approved for each particular use by a ample, enclosure or confinement of the operation, competent industrial hygienist or other technically general and local ventilation, and substitution of qualified person. Whenever respirators are used, less toxic materials). When effective engineering their use shall comply with section 1910.134 [the controls are not feasible, or while they are being OSHA respirator standard] (29 CFR 1910.1000 instituted, appropriate respirators shall be used (e), emphasis added). . . . (11, emphasis in original). In addition, in all of its substance-specific pro- The 1980 revision of this ANSI standard is vir- ceedings that resulted in new or revised permissi- tually identical in wording, although the phrase ble exposure limits, OSHA has maintained a pol- beginning with “the primary objective” is no icy of first requiring employers to comply by longer underscored. In addition, respirators are implementing “feasible” engineering and work now permitted “[w]hen effective engineering con- practice controls. Although the precise wording trols are not feasible, or while they are being in- of these requirements has differed slightly among stituted or evaluated . . . “ (12, emphasis added). these health standards, the basic outline is clear: First, an employer must institute feasible engi- Government Standards neering and work practice controls to reduce em- The conclusions and practices of the profession- loyee exposures to or below the permissible exposure limit. al community and the consensus standards organizations now have regulatory force. The startup Second, even if the feasible engineering and standards adopted by OSHA in 1971 under the work practice controls are insufficient to achieve compliance with the permissible limit, the em- authority of section 6(a) of the Occupational ployer is required to use them in order to reduce Safety and Health (OSH) Act (see ch. 12) included exposures as low as possible before issuing per- three specific provisions concerning the hierarchy sonal protective equipment. of controls. Personal protective equipment, such as dust The applicable paragraph in OSHA’s standard masks, gas masks, and other types of respirators, concerning respiratory protection (29 CFR 1910.134 is to be used against airborne contaminants in (a)(l)) was taken, word for word, from the paragraph only four circumstances: in the 1969 ANSI standard quoted above, although the ● one phrase underscored in the ANSI standard is not during the time period necessary to install underscored in the OSHA standard. Second, the feasible engineering and work practice OSHA noise standard requires employers to com- controls, ● ply with the permissible exposure limit through when engineering and work practice controls the use of “feasible administrative or engineering are not feasible (including many repair and controls” (29 CFR 1910.95(b)(l)). maintenance activities), ● when engineering and work practice controls Finally, OSHA’s main requirement for limiting ex- are not sufficient to achieve compliance with posures to toxic substances (the Air Contaminants the permissible limits, and standard) reads: ● in emergencies. To achieve compliance with . , , this section, administrative or engineering control must first Controls for Safety Hazards be determined and implemented whenever feasible. When such controls are not feasible to achieve A similar priority system has been suggested full compliance, protective equipment or any for the control of injury hazards. A clear exam-

178 ● Preventing Illness and Injury in the Workplace

ple can be found in the National Safety Council’s Accident Prevention Manual (322), which lists the following “hierarchy”: The basic measures for preventing accidental injury, in order of effectiveness and preference are: 1. Eliminate the hazard from the machine, method, material, or plant structure. 2. Control the hazard by enclosing or guarding it at its source. 3. Train personnel to be aware of the hazard and to follow safe job procedures to avoid it. 4. Prescribe Personal protective equipment for personnel to shield-them against the hazard. Discussions about the hierarchy of controls in these situations is often concerned with the relative priorities to be placed on workplace and machinery design as opposed to worker training. Em- ployee training and education will always be an important adjunct to any control technique. But ergonomics and safety research now stress the importance of design over primary reliance on the inculcation of certain “safe” work routines through training. (See also discussions of safety hazard identification, injury controls, ergonomics, and worker training in chs. 4,6,7, and 10).

Photo credit: NIOSH (photo provided by OSHA, Office of Information and Consumer Affairs Because respirators increase the effort required for breathing and are uncomfortable, workers often do not wear them or wear them only intermittently. Engineering cent rots can often eliminate the need to use respirators

ADVANTAGES AND DISADVANTAGES OF THE HIERARCHY OF CONTROLS The preference for engineering methods for con- professionals, the National Institute for Occupa- trolling workplace hazards is sometimes ques- tional Safety and Health (NIOSH), and OSHA tioned. Why are engineering controls preferred assigned personal protective equipment to the and why is personal protective equipment ranked position of being only a ‘last line of defense”? The lower than other methods? Why have nearly all reasons can be divided into those that deal with .

Ch. 9—Hierarchy of Controls ● 179 specific, often technical, problems of currently available personal protective equipment and those that address more general advantages of engineering controls.

Problems with Personal Protective Equipment

Many types of existing personal protective equipment do not provide reliable, consistent, and adequate levels of protection. Indeed, research conducted on the real-world or workplace, as opposed to laboratory, effectiveness of such equipment shows that the protection provided by these devices is unequal, highly variable, and substantially lower than that predicted from laboratory measurements (see ch. 8). For example, it has been estimated that the mean real-world attenuation of hearing protectors is only about one-third their laboratory attenuation (49). These shortcomings often make personal protective equipment inade- those at the source of a hazard, can simultane- quate for worker protection. ously control several exposure routes, such as respiratory and dermal exposure, while protective Advantages of Engineering Controls equipment is generally limited to protecting only one exposure route. The problems specific to personal protective In addition, OTA has identified five other areas equipment are sufficient to cause professionals to for which engineering controls have advantages rank engineering controls above them in the hier- over personal protective equipment. First, many archy of controls. Additionally, engineering controls have a number of advantages ranging from employer-provided personal protective programs the relatively mundane and pragmatic to the more are currently inadequate. Most discussions of the philosophical. relative merits of engineering controls and personal protective equipment center on respiratory In brief, engineering controls are inherently hazards. For those hazards, it is argued that res- more reliable and provide more effective protec- pirator programs can be designed and imple- tion than personal protective equipment and are mented to provide protection equal to that af- more likely to result in successful hazard control forded by engineering controls. Successful control (see box G). Once installed and adequately main- through the use of respirators has sometimes been tained, engineering controls provide reliable and achieved by a few, usually large, employers who consistent performance, and, if designed correctly, have sophisticated health and safety programs. adequate levels of protection. But not all employers have implemented such programs or are capable of doing so. Engineering controls work day after day, hour after hour, without depending on human super- Indeed, failures in employer respirator provision or intervention. Because they do not de- grams are the most frequent cause of OSHA cita- pend on a “good fit” with workers, they provide tions in health inspections, reinforcing the experi- the same protection to all, and monitoring devices ence of many industrial hygienists and OSHA can measure the protection afforded. Separated inspectors that most existing respirator programs from contact with the worker, engineering con- are inadequate. More than one-third of all OSHA trols do not create additional health or safety citations for violations of health standards in- problems, Moreover, these controls, especially volved the respirator program requirements, 180 . Preventing Illness and Injury in the Workplace which were directly adopted from the 1969 ANSI ogy-forcing” controls (see ch. 14). New control consensus standard (180). technology can be, and has been, accompanied by other changes in plant and equipment that im- Former OSHA inspectors have written that, in prove productivity, as well as protecting worker their experience, for most cases of employee over- health and safety (see ch. 16). Relying on personal exposure to air contaminants the employers’ res- protective equipment would reduce technologi- pirator programs were inadequate (329,363). This cal development—including the development of is supported by analysis of OSHA inspection data. controls that may find application to other OSHA inspectors cite violations of the respira- hazards, as well as improvements in equipment tor program requirements in 40 to 70 percent of that reduce hazards and simultaneously raise pro- the inspections in which employee overexposure ductivity. For example, if OSHA had permitted was measured (402). A leading consultant in the the use of dust masks to comply with the cotton field has summarized his experience: dust standard, the installation of new technology having reviewed the respirator programs of in the cotton textile industry would probably not hundreds of private companies, I can state that have taken place as rapidly as it did. Modernized I have not, with the exception of the nation’s very equipment has both dramatically improved pro- largest corporations, ever observed a proper and ductivity and lowered worker exposures to cot- adequate use of respirators in the occupational ton dust (413). setting (309). Second, it is generally impossible to make peri- Finally, personal protective equipment is bur- odic measurements of respirator efficiency, and densome on employees. It is usually uncomfortable, it is very difficult to monitor each individual em- decreases mobility, and the weight of many types ployee as he/she cleans, dons, and uses his/her of personal protective equipment increases em- respirator. Moreover, it is probably human nature ployees’ physical work loads. Negative-pressure to be on one’s “best behavior” when someone is respirators, in addition to their discomfort, also “watching. ” It is therefore difficult to be certain significantly increase the effort needed for breath- that the use of equipment observed by a respira- ing. This can create special difficulties for work- tor program manager is representative of regular ers who have already suffered lung impairments. use. Similarly, it is much more difficult for OSHA Moreover, this equipment often impairs produc- to monitor the use of masks and respirators dur- tivity. In many cases workers are paid on a “piece- ing an inspection, which are conducted infre- work” basis or are evaluated on how well they meet quently, and to be certain that the use observed employer-set productivity standards. If allowance then is representative of use on other days. is not made for the decrease in productivity due to Third, in a hierarchical approach, with control the use of protective equipment, an economic bur- achieved close to the source of the hazard, there den may be borne by the employee. is room for backup or supplemental controls. For The use of such equipment, especially respirators, example, if the primary control involves the use also impairs communication and worker-to-worker of process containment (“control at the source”), contact. This may lead to additional safety prob- then this control can be supplemented by having lems because employees wearing respirators will be personal protective equipment available (“control unable to warn each other of potential safety haz- at the worker”) for emergency use. If the primary ards, while those wearing hearing protection may control is protective equipment, the opportunity be unable to hear instructions, warning signals, or to provide for supplemental control is eliminated changes in the operations of plant equipment. (How- because the control selected is already “at the ever, one study has found decreases in accidents worker”; that is, the last line of defense has after implementation of a hearing conservation pro- already been used. gram (622).) In addition, reducing worker-to-worker Fourth, primary reliance on engineering con- communication can increase a person’s sense of trols can spur the advance of technology. The isolation and detract from the important social func- OSH Act allows the agency to mandate “technol- tions of work.

Ch. 9—Hierarchy of Controls ● 181

Lastly, workers required to use personal protective equipment may feel that they are no longer being treated as persons, especially if they are not allowed to participate in any decisions concerning the use of equipment. People in this position may feel that they are merely cogs in the workplace, similar to other machines that require hoods, filters, and masks.

Problems with Engineering Controls

Notwithstanding the advantages of engineering controls over personal protective equipment, there are some drawbacks. First, although they are less subject to human error and inherently more reliable, engineering controls can fail. No matter how well designed, a ventilation system will fail to remove an airborne contaminant if the fan stops. Although these failures can be kept to a minimum with adequate maintenance, respirators must be available for use in such emergencies. Even substitution does not eliminate the possibility of a failure in hazard control. For example, carbon tetrachloride, which had been used as a substitute for petroleum naphtha, is now widely recognized as toxic itself. Some of the substitutes for carbon tetrachloride (tricholoroethylene, per- Photo credit OSHA Office of Information and Consumer Affairs chlorethylene) are now suspected of having vari- Engineering controls for occupational noise exposures ous previously unrecognized toxic effects. include the use of sound dampening enclosures Second, there are a number of situations for control techniques or by facilitating dissemina- which feasible control methods have not yet been tion of existing techniques. fully developed. In these situations, the best that can be done is to require the use of personal pro- Cost is the principal objection to requiring engi- tective equipment while attempting to develop neering controls. Although a well-designed and feasible engineering controls. In addition, such well-conducted respirator program can be expen- equipment will always be needed when engineer- sive, such programs are in many cases cheaper ing controls are insufficient to reduce exposures than engineering controls and certainly capital to permissible levels and during the interim period costs are lower. Sometimes the concern about the while engineering controls are being designed and costs of engineering controls arises because of the implemented. belief that these costs would be infeasible for a particular firm (leading to a plant closing) or for These situations are widely recognized and are an industry as a whole. Sometimes arguments are provided for in OSHA’s hierarchy of controls. based on the relative “cost-effectiveness” of There are, and will be, however, disputes about controls. what kinds of controls are “technologically feasible” in any given situation. This is particularly Some employers and economists have suggested true when OSHA issues a regulation that “forces” that employers should be allowed the flexibility technology, either by speeding development of to choose among the available control methods, 182 ● Preventing Illness and Injury in the Workplace

rather than being required first to install feasible ployee protection at reduced cost. This argument engineering controls. This flexibility would allow is used with regard to both full 8-hour exposures employers to choose the least-costly control method. and for brief and intermittent exposures. These industry representatives argue that if the degree For example, in its reconsideration of the lead of protection offered by the equipment is equiva- standard, OSHA estimated the compliance costs lent to that provided by engineering controls, it for a group of industries. If these industries were would be sensible to choose the least costly means. required to use engineering controls, total annual Unfortunately, as already noted, many kinds of Costs were estimated to be nearly $130 million. protective equipment have not been demonstrated But if respirators were allowed, the costs would to be effective in actual workplace conditions. In be about $78 million (626). addition, there may be additional benefits from A second example concerns compliance costs requiring engineering controls, such as increases for reducing asbestos exposures from 2 fibers/ in productivity or relative decreases in absentee- cubic centimeter to 0.5 fiber/cubic centimeter. The ism. These economic benefits need to be consid- total annual costs for all industries, in 1982 ered when judging the “cost-effectiveness” of dif- dollars, are estimated to be about $134 million ferent control techniques. if engineering controls are required and $54 mil- The cost-effectiveness argument is also raised lion if respirators are allowed. If the permissible in connection with the requirement that feasible limit is set still lower, at 0.2 fiber/cubic centi- engineering controls be installed even when engi- meter, the annual costs would be about $170 mil- neering and work practice controls are insufficient lion for engineering controls and $56 million for to achieve compliance. In these situations personal compliance through the use of respirators (647). protective equipment is needed for workers in or- For both examples, there are limitations to the der to meet permitted exposure levels. It is argued accuracy of these estimates. Moreover, in neither that in such cases expenditures for engineering case was OSHA able to estimate the potential re- controls are wasted because personal protective duction in worker productivity due to the in- equipment will still be required. But because per- terference caused by respirators. But, as esti- sonal protective equipment often fails, reducing mated, the cost differences between these control contaminant levels through the use of engineer- methods appear, in many cases, to be substantial. ing controls minimizes the likely degree of worker overexposure. Many employers believe that the use of personal protective equipment can provide adequate em-

OSHA’S POLICY TOWARD THE HIERARCHY OF CONTROLS Since its creation in 1971, OSHA has enforced sonal protective equipment. Many of these same the provisions of the startup standards requiring professionals, and other representatives of em- the use of engineering controls. In addition, OSHA ployers’ and trade associations, would then ex- has developed, through rulemaking proceedings, plain why this general policy did not apply to their new or revised standards covering a number of own workplaces or industries. Representatives toxic substances or hazardous physical agents. from labor unions, public-interest groups, and In these proceedings, nearly every health and Government would carefully argue in reply that the preference for engineering controls was nec- safety professional, irrespective of whether the person worked for an employer, a trade associa- essary to protect employee safety and health. tion, a labor union, a public-interest group, a The requirement that regulatory agencies per- university, or the Government, agreed that engi- form “economic impact assessments” (see ch. 14) neering controls are preferable to the use of per- has resulted in economists becoming directly in- Ch. 9—Hierarchy of Controls ● 183 volved in discussions concerning health standards. NIOSH, in its comments, supported the hierar- The Council on Wage and Price Stability advised chy of controls: OSHA, on a number of occasions, to change its Each element of the hierarchy: 1) preventing policy and allow the use of personal protective emissions at their source, 2) removing the emis- equipment in place of engineering controls. In sions from the pathway between the source and each proceeding on health standards before 1981, the worker, and 3) control at the recipient, should OSHA rejected that advice and continued to re- be applied in descending order to the extent fea- quire engineering controls as the first line of de- sible before the next lower element is applied fense for controlling exposures to air contaminants. (585). Health and safety professionals working for Actions by the Current Administration universities and government agencies supported the preference for engineering controls. The two In February 1983, OSHA published an Advance associations of professional industrial hygienists Notice of Proposed Rulemaking announcing that (the American Conference of Governmental In- it was conducting a review of the hierarchy of con- dustrial Hygienists and the American Industrial trols and requesting information and comments Hygiene Association) supported the concept of from the public. OSHA stated that its objectives first using engineering controls: were to: The elected Board of Directors of the [ACGIH] ● Explore whether a revised policy will allow . . . unanimously endorses continuation of the employers to institute more cost-effective current [OSHA] policy to require employers to compliance strategies. use feasible engineering controls, work practices, ● Investigate whether advances in respirator and administrative controls to prevent employee design, technology, and applications may exposures above permissible levels. Personal protective equipment, including respirators, may be permit increased reliance on respirators. ● used as alternatives only when other methods are Attempt to identify processes, operations, not adequate, are not feasible, or have not yet and circumstances appropriate for particu- been installed. Furthermore, we endorse engineer- lar compliance strategies. ing controls as the preferred approach (125). ● Assess actual workplace conditions and em- The AIHA would like to go on record as stating ployee health in industries and operations that the elimination of workplace hazards is employing different compliance strategies superior to the use of engineering controls, per- (639). sonal protective equipment, and other control strategies. Where elimination is not feasible, engi- During the comment period that followed this neering and other control strategies should be the announcement, OSHA received 132 separate com- primary methods for reducing or eliminating ex- ments from the public, including employers, trade posures in the workplace. However, personal pro- associations, labor unions, and individuals. Em- tective equipment may be necessary pending more ployers and their representatives supported a gen- long-term solutions, We recognize that there are eral change in OSHA’s policy, and asked that the times where personal protective equipment is ul- agency allow the use of respirators to substitute timately the only feasible control. The decision for engineering controls. There were some differ- to recommend engineering controls, personal pro- ences among employers and trade associations tective equipment or other control strategies de- concerning the precise circumstances under which pends on the nature of the hazard in question and such a substitution should be permitted, but should be based upon the professional judgment nearly all were in favor of allowing employers of an industrial hygienist (375). flexibility to choose between engineering controls At this time, OSHA has not yet publicly an- and respirators. Labor unions, on the other hand, nounced what course it will follow concerning the voiced support for the existing policy and objected general hierarchy of controls policy. in light of strongly to any changes. OSHA's reconsideration of the hierarchy of con- 184 . preventing Illness and Injury in the Workplace

trols, the advisory panel for this OTA assessment also concluded that most operations that gener- asked to be recorded as endorsing the hierarchy ated short-term exposures to ethylene oxide could of controls. To turn away from the hierarchy of be controlled with the use of engineering controls. controls without careful verification of the level Thus, the agency did not, as it proposed for ethyl- of protection afforded by personal protective ene dibromide, allow the general use of respira- equipment is likely to increase exposures to health tors for all short-term operations with ethylene hazards. oxide exposures (649). (However, because of objections from OMB, the agency did not issue any In several substance-specific standards, the cur- short-term exposure limit for ethylene oxide, but rent administration has both continued the tradi- requested public comments on the need for such tional policy and proposed changes. In its recon- a limit and the feasibility of a short-term exposure sideration of the cotton dust standard, OSHA limit without the use of respirators.) decided to continue to require engineering controls, even in the face of objections from the Of- In addition, OSHA concluded that for some sit- fice of Management and Budget (OMB). This dis- uations, engineering controls to meet the 8-hour pute between OSHA and OMB was, at least time-weighted average for ethylene oxide were not temporarily, resolved in OSHA’s favor and the feasible and, for the first time in any health stand- published proposal reiterates the hierarchy of con- ard, specifically lists those operations in the text trols (641). of the regulation. For those situations, employers However, in a proposal concerning a new are allowed to issue respirators as the primary standard for ethylene dibromide (EDB), OSHA means of control (649). proposed for the first time to make an important Moreover, the current administration has also exception to the traditional hierarchy of controls, issued an administrative directive to OSHA in- although this exception would be limited to cases spectors, ordering that no citations concerning of intermittent exposure. The agency’s proposal OSHA’s permissible exposure limit for noise be would allow employers to use respirators as the issued to employers who had issued and required primary means of control “where exposure to EDB the use of hearing protectors by employees ex- . . . is intermittent [defined as an operation that posed to levels up to 10 decibels (dB) above the results in exposures occurring for 1 or 2 days at permissible limit (that is, exposures between 90 any one time] and occurs less than a total of 30 dB to 100 dB) (646). Because of the logarithmic days per year” (642). nature of the decibel scale, this difference of 10 In its 1984 proposal for a new asbestos stand- decibels is equal to a tenfold increase in permissi- ard, the agency proposed that employers be al- ble exposures. One researcher has estimated that lowed to use respirators on a continuous basis as using hearing protectors instead of engineering the primary means to comply with a new, reduced controls for noise exposures between 90 dB and exposure limit. Employers would still be required 100 dB will double the probability that an exposed to use feasible engineering and work practice con- worker will incur an occupational hearing loss trols to meet the current existing limit (2 fibers/ (165). cubic centimeter). But they would be allowed to Finally, OSHA first granted and then withdrew use personal protective equipment on a continu- an experimental variance from the medical remov- ous basis to reduce exposures from the 2 fibers/ al protection provisions of the lead standard. The cubic centimeter limit to the new permissible ex- lead standard requires that employees whose posure limit (either 0.5 or 0.2 fibers/cubic centi- blood lead levels exceed certain specified limits meter) (647). must be transferred to jobs with little or no ex- In its 1984 final rule reducing the permissible posure to lead. The variance would have allowed exposure limit (measured as an 8-hour time-weighted one employer to issue respirators to several em- average) for ethylene oxide exposures, OSHA re- ployees with blood lead levels above the permitted quires employers to comply with the new permis- limits, who would then continue in lead-exposed sible exposure limit through the use of feasible jobs instead of being transferred to positions with- engineering and work practice controls. OSHA out lead exposures (243,338,349). . —

Ch. 9—Hierarchy of Controls ● 185

CONCLUSION

In summary, OTA finds the hierarchy of con- controls allow easier monitoring of performance trols to be a well-founded and protective concept. by employers, employees, and OSHA. In addi- Applicable to both health and safety hazards, the tion, these controls are inherently more reliable hierarchy is derived from the experience of health and do not create employee burdens, and requir- and safety professionals and has been embodied ing them enhances the development of new con- for years in consensus standards, professional trols and production technology. practice, and OSHA regulations. Engineering con- OSHA’s reevaluation of its longstanding pol- trols are more likely to meet the essential require- icy favoring engineering controls for airborne con- ments for hazard control, and personal protec- taminants may indicate a shift in the agency’s ap- tive equipment is a last line of defense to be used proach to the hierarchy of controls. That policy when engineering controls are infeasible, insuffi- has led to improvements in the health of U.S. ciently protective, or not yet installed. The prob- workers and has spurred, at least to some extent, lems of personal protective equipment arise out the development of control technologies. Policy of 1) limitations in performance, 2) difficulties in changes that allow greater use of personal pro- evaluating their performance, 3) problems asso- tective equipment may endanger the health and ciated with their use, and 4) the physical and other well-being of many American workers and reduce burdens they create. Moreover, engineering con- the regulatory imperative to develop new and bet- trols are preferred on more general grounds be- ter production and control technologies. cause most personal protective equipment programs are inadequate and because engineering 10. Training and Education for Preventing Work-Related Injury and Illness Contents

Page Worker Training and Education ...... 189 Fundamentals Programs...... 190

Recognition Program$ ...... 191 Problem-Solving Programs ...... 191 Empowerment Programs ...... 191 Evaluation of Worker Training and Education Programs ...... 194 Occupational Safety and Health Professionals . . 195 Continuing Education ...... 195 Curriculum Development and Dissemination . 195

Planning...... ,..,, ...... 6... < ...... 197 Training Grants...... 198 Educational Resource Centers ...... 198 Other Training Programs for OSH Professionals ...... 198 Engineers and Managers ...... 6 199 Educating Engineers ...... 199 Training Business Managers ...... 199 Computer Networks as an Educational Tool...... 200 NIOSH Use of Computer Networks ...... 200 Privacy andTrade Secret Concerns ...... 201 The Future of Computer Exchanges ...... ,., 201 Summary ...... 201

Figure No. 1o-1. Effect of Funding Changes on Numbers of Graduate Degrees Granted and Continuing Education Courses Provided by Both NIOSH and Grant Supported Universities ...... 196

. . ., . . ——

10 . Training and Education for Preventing Work-Related Injury and Illness

Worker training in injury and illness preven- Responsibilities of NIOSH and OSHA overlap. tion is provided by many employers; it can range For instance, NIOSH publications are used in from rudimentary instruction about “safety rules” worker training and the OSHA Training Institute to sophisticated instruction about potential haz- trains professionals in short courses about OSHA ards and technologies for their control. Training regulations. was at management’s discretion until the Occupa- The importance of managers and engineers in tional Safety and Health Administration (OSHA) identifying and controlling workplace hazards has included training requirements as part of several led some professional organizations to discuss regulations. Unions have rarely had the resources educational needs of their members. Both NIOSH to train members in preventing work-related in- and OSHA are participating in developing short jury and illness. However, some unions now have courses for these professionals. (or had) training programs for their members supported by funds from the OSHA New Directions It is difficult to measure the effects of training program (see ch. 12). OSHA is the Federal agency and education programs. Ideally, decreases in ill- primarily responsible for worker training. ness and injury rates would be associated with the programs, but the rates are influenced by so many The National Institute for Occupational Safety factors that clear conclusions are difficult. Cur- and Health (NIOSH) is primarily responsible for rently most measures of success are “process meas- educating occupational safety and health profes- ures” that count the number of students taught sionals. NIOSH, especially through its funding of or hours of instruction provided. Educational Resource Centers (ERCs), has emphasized interdisciplinary education of occupational physicians and nurses, industrial hygienists, and safety engineers,

WORKER TRAINING AND EDUCATION

Workers trained to recognize and avoid hazards and unforeseen problems. It prepares the worker are an indispensable part of creating and sustain- to identify potential hazards, to request and col- ing an illness- and injury-free workplace. In this lect information about hazards, and to seek ways assessment, worker training is defined as instruc- to eliminate or control hazards. tion in recognizing known hazards and in using Although the boundary between training and available methods of worker protection. Such education programs is often vague, the various training, for example, emphasizes physical pro- programs may be described in four major and cedures and skills, such as keeping a shop floor logically related categories: fundamentals, recog- free of debris, proper attention to the locking of nition, enforcement, and empowerment (227). In machinery during maintenance, or proper use and general, the narrower the role the worker is ex- cleaning of respirators. pected to assume, the more instruction is “train- Worker education is defined as instruction in ing.” The broader the role, the more instruction analyzing and responding to new circumstances is “education .“

189

Since a comprehensive analysis of worker train- potentially hazardous tools, equipment, and ing and education programs, their effectiveness, materials; and the resources devoted to them is lacking, ● emergency procedures; OTA contracted with INFORM to survey 40 ● personal hygiene as related to use of hazard- worker training and education programs. Re- ous materials; spondents included 8 business firms and trade ● the need for medical checkups or examina- associations, 10 unions, 4 hospital-based pro- tions; and grams, 8 university-based programs, 6 Commit- .• use of protective devices such as masks, tees (or Coalitions) for Occupational Safety and respirators, safety goggles, or gloves during Health (COSH groups), and 4 miscellaneous edu- nonroutine maintenance, where protective cational programs. engineering methods have not yet been implemented, or during emergencies (227 and see box H). Fundamentals Programs Of the five “fundamentals” programs in the INFORM survey, three were operated by business “Fundamentals” worker training programs instruct about known hazards in order to prevent firms and two by trade associations. Four teach work-related illness and injury. Such training may supervisors, occupational safety and health instruct workers in: specialists, or other “keypersons” to convey information and skills to workers. In the fifth case, * prevention of work-related injury and illness the health and safety staff of a company instructs through the proper use and maintenance of both workers and supervisors. The program ob- Ch. 10—Training and Education for Preventing Work. Related Injury and Illness ● 191

jective most frequently cited was to reduce costs ● judge the effectiveness of alternative meth- of work-related injury and illness. ods of control (227). Of the 14 “problem-solving programs” included Recognition Programs in the survey, 5 are administered by unions, 5 by “Recognition” programs prepare workers to universities, 1 by a hospital, and 3 by independ- participate in a broader range of worksite safet y ent educational programs. The major program ob- and health activities through: jectives emphasize teaching workers to use particular mechanisms to address workplace problems. ● emphasizing awareness of the hazards present in the workplace; ● understanding different methods available for Empowerment Programs hazard elimination or control; “Empowerment” programs aim to teach work- ● understanding rights and responsibilities ers the broadest range of skills so as to involve under the law, e.g. right-to-know laws; them in defending and expanding their rights to ● collecting information about workplace haz- an illness- and injury-free workplace. The major ards, e.g. chemical identity of workplace sub- assumption underlying “empowerment” programs stances, symptoms associated with ex- is that the goals of cleaning up hazardous work- posures, exposure levels; sites and ensuring the health and safety of work- ● observing or informally inspecting the workers requires a substantial transfer of political and place for potential hazards; and economic power to workers and their unions. To ● reporting hazards or potential hazards to help educate workers to play a broad social and appropriate individuals or committees (227 political role, “empowerment” programs must ei- and see box I). ther instruct workers in the entire range of skills Of the 14 “recognition” programs in the sur- and knowledge offered by “fundamentals,” “rec- vey, 3 are conducted by businesses, 5 by unions, ognition, ” and “problem-solving programs’’—or 3 by universities, and 3 by hospitals; their major they must work to ensure that such training is pro- objective is to teach workers to recognize hazards. vided to workers by unions and/or management (See box K). Problem-Solving Programs Although INFORM (227) did not collect information about employers’ responses to “empower- The general objective of “problem-solving” pro- ment” programs, it is expected that those programs is to provide workers with the information grams would be opposed. The transfer of political and skills necessary to participate in hazard rec- and economic power from employers and manag- ognition and control. These programs prepare ers in any area, including safety and health, would workers to: be a dramatic shift. . help solve problems that may arise on the Six of the seven “empowerment” programs in shop floor, in an ongoing, regular way, by the survey are conducted by Committees or Coali- use of union and management resources and tions for Occupational Safety and Health. The mechanisms; and other is conducted by a nonprofit educational pro- ● exercise legal rights when necessary and prac- gram. The political focus of “empowerment” pro- tical, to bring outside agencies, especially grams is apparent in the following descriptions OSHA and NIOSH, into the workplace to of program objectives listed by four groups from help address hazards (see box J). the survey: In order to be effective, workers must learn to: ● empower workers to deal with problems at ● question work processes and materials to work; assess whether hazards have been adequately ● build on hazard recognition and problem- identified; and solving skills to give safety committee mem- 192 ● Preventing Illness and Injury in the Workplace —— —.

Ch 10—Train/ng and Education for Preventing Work-Related Injury and Illness ● 193 — .

IBPAT reports that its program has had a significant impact on collective bargaining. According to an IBPAT evaluation of its fiscal year 1982 activities, 5 out of 10 locals or district councils that completed the IBPAT program prior to negotiations of their contracts secured provisions for health and safety committees. Among affiliates that did not receive IBPAT instruction, only 3 out of 10 secured such committees. The IBPAT program evaluation also showed that 27 percent of those affiliates that have completed the program won provisions for training programs in their new contracts. This compares with only 10 percent of affiliates that had not undertaken the program. (Both measures of success are complicated because the affiliates who participated in the program may have already made significant commitments to safety and health. If that were so, participation and success in negotiations might be better characterized as two successes stemming from overall commitment rather than any cause-effect relationship. )

Source: (227).

Box J.—Example of a Problem-Solving Program: Labor Occupational Health Program (LOHP) at the University of California, Berkeley . LOHP offers a variety of programs and services, including: 1. formal education and training courses; 2. workshops and conferences; 3. plant walk-throughs to assist in hazard identification; 4. off-site technical advice; 5. legal advice; 6. assistance in collective bargaining; and 7. maintenance of a resource center for public use. In 1982, LOHP instructed more than 800 workers and union officials through courses, workshops, and conferences and provided educational and training materials to 4,000 workers and union officials. In addition to providing instruction typical of “fundamentals” and “recognition” programs, LOHP places special emphasis on teaching the problem-solving skills listed in the chart below: Problem-solving programs convey information LOHP offers the following programs that and skills to workers that enable them to: instruct workers and union officials in problem- solving skills: 1. take an active role in seeking solutions to 1. training for participation in joint committees problems on the shop floor (such as an annual course for the Oil, Chem- ical and Atomic Workers International Union , and petrochemical companies in the Berkeley area) a conference on workers’ compensation for workers and trade unionists workshops on “Collective Bargaining on Health and Safety” steward training sessions in health and safety ‘ for unions Z. exercise legal and contractual rights and 2. a conference on California right-to-know ensure management compliance with laws law, attended by 150 union and management and contractual agreements representatives, professionals, and students small workshops to aid in the implementation of the state right-to-know law

Source: (227). 194 . preventing illness and Injury in the Workplace

bers and local union officials and stewards skills in observing and interpreting air monitoring and medical tests; track trade names of potentially hazardous chemicals; devise engineering control methods; negotiate effective health and safety contract language; ● empower workers and their unions to win illness- and injury-free working conditions; have more control over their working conditions; and ● empower workers with the skills and the con- fidences they need to prevent work-related illness and injury in their workplaces through collective action (227). Evaluation of Worker Training and Education Programs Ideally, effectiveness of worker education and training programs would be measured by reduced job-related injuries and illnesses. However, the data currently available are often limited, especially for illnesses (see ch. 2). In addition, even with the appropriate data, attribution of improvements to one factor requires knowledge of all other factors that might have an effect. In practice, effectiveness measures are generally indirect and include counting the number of workers trained or educated and surveying workers’ and management’s perceptions of the value of the programs. Ch. 10—Training and Education for Preventing Work-Related Injury and Illness . 195

OCCUPATIONAL SAFETY AND HEALTH PROFESSIONALS

Congress recognized the need for training a the recognition of workplace hazards; to the Mine cadre of occupational safety and health profes- Safety and Health Administration Training Acad- sionals to reach the objectives of the Occupational emy to provide information about industrial hy- Safety and Health (OSH) Act of 1970. Section 20 giene sampling and analysis for mine inspectors; of the Act designates NIOSH as the agency to plan and to Case Western University to train science for and carry out programs to assure an adequate department faculty in preventing injuries in lab- supply of professionals. NIOSH was given author- oratories. it y to conduct programs for this purpose directly, Between 1977 and 1981, continuing education through grants, and through contracts. enrollment increased from approximately 1,600 Since 1971, NIOSH has conducted short-term to over 10,000 per year (fig. 10-1). At this rate, training programs for occupational health and enrollment could eventually increase to 50,000 per safety professionals, developed curricula and year if each specialist in the field attended one con- training materials, and provided grants to univer- tinuing education program at least every other sities. Through a grant program, it has established year. As figure 10-1 illustrates, the number of Educational Resource Centers to train occupa- attendees in continuing education courses in- tional safety and health professionals through creased with funding from 1976 to 1980. interdisciplinary programs, to provide consulta- However, a substantial reduction in the num- tion and training for workers and employers, and ber of attendees was projected, based on an to conduct research related to worker health and assumption of a lack of Federal funding of ERCs safety. NIOSH has also provided financial sup- in the 1984 Presidential Budget (102). In fact, port to students through training grants to qual- funding was continued by Congress and the num- ified universities (588). ber of trainees is now estimated to be 14,000 per In addition to NIOSH courses, large companies year. The demand for continuing education in- sometimes train their own professionals. A coop- creased dramatically as OSHA and NIOSH re- erative agreement between General Motors and sources for disseminating information to profes- the United Auto Workers to train union safety sionals were cut. This is partially because stewards is described in chapter 15. graduated professionals sought to update their Continuing Education knowledge and skills. Funds have been available to meet this demand due to the flexibility of ERCs NIOSH conducts technical training courses for in rebudgeting funds intended for graduate train- private-sector and Government professionals and ing, where enrollment has been declining. technicians. The cost of the training courses is paid Curriculum Development and by reimbursements to NIOSH. Short-course top- Dissemination ics include: . industrial hygiene; Refinement and modernization of curricula for short-term and continuing education can improve . occupational safety; the effectiveness of occupational safety and health . industrial toxicology; training. One way NIOSH disseminates recent re- ● occupational health nursing; and search findings that may help prevent work- ● occupational medicine. related illness and injury among high-risk groups Special custom courses are produced to meet is through the development of curricula. Gener- specific training needs identified by Government ally this approach consists of identifying groups and private sector groups. For example, in fiscal who have access to high-risk workers, develop- year 1982, courses were presented to the U.S. ing materials that will train these target groups Navy about analyzing asbestos samples taken dur- in recognition and means of prevention of harm, ing ship refitting; to the Health Departments of and conducting “training-the-trainer” programs New Jersey and Arizona to train state officials in so these groups can transfer the knowledge.

196 . preventing Illness and Injury in the Workplace

Figure 10.1 .—Effect of Funding Changes on Numbers of Graduate Degrees Granted and Continuing Education Courses Provided by Both NIOSH and Grant Supported Universities

781 800 - 747 743 Based on ERC directors’ forecast and funds reinstated by 600 Annual congress graduate 432 degrees 400 granted 217 Based on ERC 200 directors’ forecast and no federal funds

1976 1977 1978 1979 1980 1981 1982 1983 1984 (est)

14,0 14 r Estimated following

12 reinstatement of funds by congress 10.5 9.8 Annual number 10 - of attendees in continuing 7.9 education 8 “ courses (thousands) 6 “ 4.4 4 Based on ERC 2 “ 1.5 directors’ forecast and no federal funds

1976 1977 1978 1979 1980 1981 1982 1983 1984 (est) 12.9 12-

10 9.4 8.7 Annual 8 - Reinstated training 7.0 by congress funds ($ million) 6 “

4 2.9

President’s 1984 budget

0’ 1976 1978 1979 1980 1981 1982 1983 1984

Ch. 10—Training and Education for Preventing Work-Related Injury and Illness ● 197 —

An example of curriculum development is the centrated in jobs that “inspect, interpret, in- joint NIOSH-OSHA program for instructing vo- vestigate, and plan, ” such as industrial hygienists cational education teachers how to teach skills in and safety engineers (570). recognition and control of workplace hazards. In- The future supply of occupational safety and troducing such concepts to new workers before health personnel was estimated in another NIOSH they are employed is especially important since survey by asking 112 programs in occupational young workers are at the highest risk of traumatic safety and health education how many graduates injury. Pilot programs were conducted with over they expected in the future. These results, com- 100 vocational education administrators and bined with estimated numbers of personnel teachers in Ohio, California, Minnesota, Arizona, trained by insurance companies and by the Fed- Florida, and New York. The program was de- eral Government, showed that in 1977 the designed for flexibility so that its 17 units can be mand exceeded supply of occupational safety taught over 3 days or over longer periods, depend- specialists, occupational health nurses, and oc- ing on individual needs. cupational health physicians, but that the demand Another example of training-the-trainer was the equalled supply for industrial hygienists. Due to NIOSH program to increase the awareness of high survey limitations, data collected at that time were school science teachers about hazards in the work- insufficient to predict adequacy of supply of oc- ing environment. In a three-year period, over cupational safety and health personnel by 1990 100,000 teachers were trained through this pro- (570). A report that updates work force supply gram at a cost of little more than $1 per trainee. and demand information has been completed This program also resulted in changes that de- (584); however, publication of the report awaits scribed chemical hazards to workers in a recently OMB approval. published high school science text. The NIOSH surveys of demand and supply, Specific teaching modules have also been devel- like all such projections into the future, were based oped and made available to appropriate groups. on assumptions about conditions in the years For example, audiovisual presentations on safe ahead. Because of the sagging economy in the removal of asbestos from school buildings were early 1980s, and probably also because of a re- developed jointly by NIOSH, OSHA, the Envi- duced Federal presence in occupational safety and ronmental Protection Agency, and the National health, there have been few positions available Cancer Institute. Over 1,000 copies of these pro- in this field. The Advisory Panel to this assess- grams were distributed to train asbestos removal ment lamented the fact that occupational safety contractors. In another case, videotape programs and health professionals are often among the first were developed to instruct firefighters and other workers in the techniques of maintaining and don- ning self-contained breathing apparatus. These tapes featured the latest findings from the testing of these devices. Planning

In 1977, NIOSH conducted a nationwide survey of governmental agencies and nonagricultural firms with more than 100 workers to determine the current number of employed occupational safety and health professionals and to predict future demand. Based on nearly 3,300 survey responses, NIOSH estimated the number of full-time professionals to be between 104,360 and 110,840 Photo credit E.I. du Pont de Nemours & Co and forecast 3,100 new positions a year would be Measuring the concentration of potentially harmful available by 1990. Growth in demand was con- gases is one aspect of an industrial hygienist’s job 198 ● Preventing Illness and Injury in the Workplace

to be laid off from company and union jobs, but Recent budget cuts have reduced the current could offer no suggestions to alter that condition. level of funding for Educational Resource Centers to $5.8 million, 55 per cent below the level in fiscal Training Grants year 1980. According to projections by the Asso- ciation of University Programs for Occupational NIOSH’s findings of shortages of trained oc- Health and Safety, if Federal funding for ERCs cupational safety and health graduates were cited had been eliminated as proposed in the President’s in successful efforts to expand training grants pro- fiscal year 1984 budget, the number of graduates grams. One part of this expansion was to intro- completing their programs would have decreased duce the concept of multidisaplinary educational to approximately 338 (compared with about 781 resource centers. The other part was growth of in 1981) and only about 25 degree programs of single-discipline training grants. the 112 programs currently in existence could have Single-discipline training project grants have been expected to survive. Furthermore, there was been established in 28 universities, and over 100 concern that without Federal funds, multidisci- different academic degree programs were in place plinary programs would revert to more narrow in 1980. Because of budget cutbacks, the number and limited single-discipline programs. of programs was reduced to 60 in 1982. The num- While the President’s 1984 budget for NIOSH ber of professionals graduating from these and contained no request for ERC funding, the Con- ERC programs each year increased from 217 in gress added $8.8 million for the centers. 1976 to 747 in 1980. As shown in figure 10-1, the number of graduates increased as the funding in- Other Training Programs for creased. OSH Professionals Educational Resource Centers Other Federal agencies and private companies train occupational safety and health professionals Congress authorized creation of up to 20 Educa- in special cases. OSHA maintains a training insti- tional Resource Centers for occupational safety tute in Des Plaines, IL, primarily to give short- and health in 1976. Funding increased from $2.9 term training to State and Federal compliance of- million in 1977 to $12.9 million in 1980, and the ficers, and the institute also provides short courses ERCs now number 15. These centers: 1) provide to some private sector groups (636). The Mine continuing education to occupational health and Safety and Health Administration maintains a safety professionals; 2) combine medical, indus- Training Academy in Beckley, WV, to give short- trial hygiene, safety, and nursing training so that term training to its inspectors (452). graduates are better able to work effectively in Many short-term training courses for workers complex and diverse conditions; 3) conduct re- and occupational safety and health professionals search; and 4) conduct regional consultation serv- are sponsored in the private sector. The courses, ices. All ERCs but one are located in universities. including a wide variety of specialized short-term The centers are distributed as far as possible to training in both illness and injury prevention, are give regional representation and to meet training announced weekly in journals such as the Occupa- needs for all areas of the Nation. The Federal cost tional Safety and Health Reporter published by of ERC education is approximately $7,000 for the Bureau of National Affairs. The number of each degree graduate and $70 for each attendee private sector courses has grown substantially at continuing education courses (102). since the passage of the OSH Act. Ch. 10—Training and Education for Preventing Work-Related Injury and illness ● 199

ENGINEERS AND MANAGERS

Although there appears to be a growing aware- ticularly important group because the greatest op- ness in the business community of the costs of portunity for change is in the planning stages of work-related illness and injury (284), little infor- industrial processes. “All undergraduate engineer- mation is provided about prevention in business ing curricula (particularly design courses) should schools (579). Furthermore, there is little evidence include instruction on the responsibilities of engi- that engineering schools treat the subject at any neers for occupational safety and health engineer- level, but there are some attempts to change this. ing problems and solutions” (587). The report also Professional engineering societies and trade asso- recommends that academic programs in occupa- ciations have formed committees on occupational tional safety and health should be offered to prac- safety and health, governmental agencies are con- ticing engineers as continuing education. The Amer- ducting training and information programs, and ican Board of Engineering and Technology, which educators are slowly becoming concerned. accredits programs in engineering, finds it “desirable” for engineering schools to teach safety (4). Engineers and business managers are often unaware of the potential for reducing work-related Specific legal responsibilities require profes- injuries and illnesses either because the hazard re- sional engineers to protect workers and the pub- mains to be identified or because information lic as well as their employers. Professional engi- about it is new and inadequately disseminated. neers may be required to act according to the Even after a hazard is recognized, these business prevailing practice of the profession by State law, decisionmakers may lack access to information but they may lack knowledge about local or Fed- on which to base plans for prevention. eral regulations or penalties for noncompliance. Educating Engineers Indeed, engineers have been held accountable for the actions of untrained subordinates in some The engineering curriculum is one of the few cases. In the case of workplace health and safety that prepares baccalaureate degree holders for a regulations, training may be absent since few professional license. Because it provides both an schools of engineering have courses bearing on undergraduate and a professional education, the responsibilities under the Occupational Safety and curriculum is a crowded one and under constant Health Act. pressure to add new material. Occupational safety Complicating problems for engineering schools and health would be a useful addition or augmen- are current pressures to update curricula to meet tation to currently scheduled lectures, but it is dif- rapidly changing technology, to respond to the ficult to squeeze it in. sudden upsurge in enrollment, and to upgrade Few people more fully recognize the difficulties faculty and overcome critical staff shortages. of the engineering curriculum than professional Undergraduate enrollment exceeded 387,500 in the educators, yet they see some opportunities to in- academic year 1981-82. This was an all-time high crease safety and health instruction. For instance, for the Nation’s 286 engineering schools (419). industrial-hygiene-engineering educators recommend, in a recent report, presenting students with Training Business Managers information on recognition and control of work- related injury and illness hazards and making such According to a national survey of 217 schools information available to practicing engineers of business management accredited by the Amer- (587). They single out design engineers as a par- ican Assembly of Collegiate Schools of Business 200 ● Preventing Illness and Injury In the Workplace

(a group representing a majority of business man- ing programs in business schools. The project will agement schools) that was conducted by the Na- be publicized through a series of briefings given tional Safety Management Society, few specify to administrators and faculty at selected business safety management studies as a degree require- schools. Audiovisual materials will be prepared ment (311). It is not difficult to imagine that such and a case-study book assembled containing 50 studies would be of low priority in management related occupational safety and health articles. schools, but some companies argue that commit- These materials will be designed so that they can ment of top management is the single most nec- be used in classwork by the business educators essary ingredient in safety and health programs (584). (156). NIOSH and OSHA have begun Project Minerva to develop occupational safety and health train-

COMPUTER NETWORKS AS AN EDUCATIONAL TOOL Improving the quality and quantity of infor- with OSHA’s regulation concerning Access to Em- mation available about preventing work-related ployee Medical Records, which stipulates what injury and illness depends upon finding solutions medical records are to be retained by employers, and communicating the solutions to people who for how long, and who has access to them. In both will benefit from them. The traditional methods cases, NIOSH arranged a conference that involved of information exchange are journal articles, pres- about 10 NIOSH employees in scattered locations; entations at professional meetings, and various there was little overlap in the membership of the kinds of consultation. Computer conferences— two groups. formed by people who share common interests NIOSH participants consider the exchanges to and who “participate” in the conference by sen- have been a success. The computer-time cost for ding and receiving messages in a central comput- the hazard communication exchange was $2,500, er—are an experimental method of information about two-thirds as much as a 2-day meeting at exchange. They have been tried on a limited scale NIOSH in Cincinnati, which would be an alter- for the exchange of information about occupa- native method of gathering needed responses. tional safety and health and on a larger scale within companies to exchange technical informa- A NIOSH-MIT exchange was organized differ- tion. It is unclear what their eventual value will ently. The Reader’s Digest computer data base, be, but they have been praised by some occupa- the Source, has software called PARTICIPATE tional safety and health professionals who have that permits its users to join in conferences. Three used them. professionals with expertise in computers and video display terminals posted a message on the NIOSH Use of Computer Networks Source inviting comments and discussion. Over a dozen participants joined in; the discussions fol- One of the duties of the Technical Information lowed two paths—one concerning possible health Division of NIOSH is to draft responses to effects from radiation and ergonomic considera- OSHA-proposed rules. NIOSH used computer tions, the other stress and quality of work. conferences to respond to two 1982 Advance No- tices of Proposed Rulemaking from OSHA. One Priest (377) sees the NIOSH-MIT exchange as notice concerned “Hazard Communication” (the a limited success. It attracted participants, but be- “labeling standard”), which details what informa- cause it did not have a goal, such as producing tion is to be provided to workers about the chem- a document or answering a question, the exchange icals to which they are exposed. The other dealt strayed, just as undirected conversations do. Ch. 10—Training and Education for Preventing Work-Related lnjury and Illness 201

Privacy and Trade Secret Concerns formation about their methods and, on the other hand, their proprietary business interests and de- In the late 1970s, a large United States-based sire for privacy. There is also the possibility that pharmaceutical firm with over 60,000 employees the firm was concerned about disclosing informa- worldwide developed its own system of computer tion about product safety discussions that might exchanges and over 4,000 employees use it. The be “discoverable” in a legal proceeding (377). firm’s Vice President for Research and Develop- ment considers it a “crucial and integral part” of the firm’s development and management opera- The Future of Computer Exchanges tions (377). Computer exchanges are attractive because they The system is used extensively by employees provide fast and interactive communication. The concerned about product safety. Although the information is “written down” and can be re- firm’s representatives stated that they used the sys- trieved. However, a company inhibited by desires tem considerably for health and safety purposes, to protect its privacy will be reluctant to allow they would not disclose any details about the con- employees to participate in a medium as rapid, tent or process (377). Concern about disclosing fluid, and uncontrollable as computer exchange. business secrets was partially responsible for the The future of government-sponsored and busi- company’s not discussing computer exchange of ness computer exchanges among occupational safety and health information. In addition, a com- safety and health professionals is likely to increase pany representative stated that sharing its health as cost savings become more apparent, as hard- and safety information with other companies ware becomes more widely available, and as more might make the company subject to antitrust ac- people experience and grow accustomed to this tions. Only trade associations and public meetings communication medium. Information exchange were seen as acceptable for sharing information in this manner can enhance face-to-face meetings. (377). There appear to be few advantages of computer The firm in this situation has made an implicit conferences that are peculiar to occupational tradeoff—between, on the one hand, deaths and safety and health. injuries that might be reduced by disclosing in-

SUMMARY

Traditionally, education and training programs safety topics into the curricula of engineering have focused on workers and safety and health schools. professionals. However, with recognition of the OTA sponsored a survey of worker training importance of designers, engineers, and managers programs. Not unexpectedly, the survey found in health and safety, some attention is now given that company programs, which are the majority to providing them with the principles of preven- of programs, emphasize fundamentals of safe tion and control. working habits and recognition of hazards. Both OSHA and NIOSH are involved in edu- Union, university-based, and Committee for Oc- cation and training programs. OSHA has sup- cupational Safety and Health programs prepare ported worker training and education; NIOSH workers for a more active role in recognition and grants are used primarily for education of occupa- control. The company programs are the norm; tional health and safety professionals. Both par- the other programs, which would involve trans- ticipate in Project Minerva, with the objective of fer of some power from management to labor, are introducing health and safety into business admin- more controversial and they do not represent the istration curricula. NIOSH has also sponsored only way to go beyond the traditional programs. workshops directed towards adding health and For instance, General Motors and the United Auto 208 . Preventing Illness and Injury In the Workplace

Workers cooperate in the training of union safety One of the surest ways to disseminate new in- stewards in General Motors’ training programs. formation is to graduate new professionals and NIOSH’s 15 Educational Resource Centers, lo- send them into the work force. As enrollment in graduate programs has declined, ERCs have in- cated across the Nation, represent a departure creased attention to continuing education for prac- from disciplinary training for health and safety ticing health and safety specialists. The flexibility professionals. They emphasize interdisciplinary of ERCs in providing both degree and continu- training of industrial physicians and nurses, in- ing education programs is an argument for their dustrial hygienists, and safety engineers. Unfor- continuation. tunately no solid evaluation of the programs has been carried out, but the ERCs are well regarded Evaluation of Federal education and training by most professionals. programs has consisted largely of counting numbers of graduates and hours of instruction. While The stability, if not the very existence, of ERCs those measures provide an indication of how is threatened by uncertain Federal funding. For much education is going on, they provide no de- each of the last four years, the President’s bud- tails of the impact of the education. Careful get has recommended no Fe&al funding. In each evaluation would aid in making funding decisions. year, Congress restored funding. Nevertheless, the threat to the ERCs has seen parallel decreases in the number of graduate degrees awarded. ●

11. A Short History of Private and Public Activities Page Voluntary Efforts ...... 205 National Safety Council ...... 205 American Occupational Medical Association ...... 205 American National Standards Institute...... 206 American Conference of Governmental Industrial Hygienists ...... 206 Governmental Efforts...... 207 Workers’ Compensation...... 207 Progressive Era Aims...... 207 State Health and Safety Programs...... 209 Early Federal Government Programs ...... 210 Federal Research and Assistance in Occupational Safety and Health...... 211 Federal Legislation and Regulatory Programs ...... 212 11 m A Short History of Private and Public Activities

Accompanying new methods of factory pro- originated in the private sector; others, in the duction in the 18th and 19th centuries were ex- Government. Many of these efforts concerning posures to new working conditions and new haz- occupational conditions were intertwined with at- ards. As some associations between hazards and tempts to improve public health more generally. injuries or illnesses were recognized, efforts were A number of programs were successful, resulting made to improve working conditions and to rein improved working conditions, although prog- duce or eliminate job hazards. Some of these ress has often been strikingly slow (see, e.g., 218).

VOLUNTARY EFFORTS The importance of workplace safety was rec- The council has worked extensively with the De- ognized by some large American firms around the partment of Labor to provide safety training turn of the century in the face of a rising number courses and materials to industry. Industrial firms of injuries associated with the installation of new that belong to the NSC have been reported to industrial machinery. One positive response was have “injury rates well below those of non-mem- the creation of employer policies and practices bers” (249). The cause-and-effect relationship be- directed at reducing the frequency of those injuries tween low injury rates and membership in the (box L). In addition to company-specific efforts, NSC is not clear. It is at least as likely that com- several voluntary organizations have been created panies with low rates join the NSC as it is that to promote occupational health and safety. membership in the NSC results in lower injury rates. National Safety Council

The first national organization devoted entirely American Occupational to occupational safety, the National Safety Coun- Medical Association cil (NSC), was established in 1912 as a response Industrial medicine originated in large compa- to the high industrial accident rate. It popularized nies that employed surgeons to treat traumatic in- the “Safety First” slogan that had been used by juries. By 1915, enough physicians and surgeons U.S. Steel. The council was governed by a busi- were engaged in industrial health to lead to the ness-dominated Board of Directors. It sought to organization of the American Association of In- achieve industry consensus for its recommenda- dustrial Physicians and Surgeons. Seventy-two tions, and favored nonmandatory standards, physicians attended the association’s first meeting training and education, and voluntary safety pro- in 1916. (In 1951, the association’s name changed (361) grams as the best method of improving safe- to Industrial Medical Association and later to ty records. The NSC thus epitomizes the volun- American Occupational Medical Association. ) tary safety movement. Also in 1915, the American Medical Association Although the council has now shifted its em- held its first symposium devoted to industrial hy- phasis to automobile and home accidents, its safe- giene and medicine. The American Occupational ty publications (and particularly Accident Facts Medical Association publishes a monthly journal, (324)) are well known and widely disseminated. the Journal of Occupational Medicine.

205 206 . Preventing Illness and Injury in the Workplace

rected at technical specifications for manufacturing, such as having common thread pitches for screws, nuts, and bolts, some of the standards issued by ANSI are directed at improving occupational safety (249). The membership of ANSI consists of companies (company members) and trade associations, government agencies, and private groups (member bodies). Several unions are also member bodies of the organization. The “consensus” method used by the institute is defined in its constitution as a position “achieved according to the judgment of a duly appointed authority. Consensus implies much more than the concept of a simple majority but not necessarily unanimity.” ANSI standards are developed by standards committees and are reviewed, since 1969, by a Board of Standards Review. The 15-person board has 9 industry members, 2 representatives from the Federal Government, 1 from municipal government, 2 academic representatives, and 1 member from a consumer organization. The board is authorized to decide when a standards committee has reached consensus; this decision includes counting and weighting the votes of the committee’s members (249). The votes and weighting factors are not made public, although ANSI does cir- culate drafts to interested parties for comment throughout the final consensus process. (The ANSI standards concerning occupational health and safety that existed in early 1971 were adopted as regulations by the Occupational Safety and Health Administration (OSHA) using the authority granted in sec. 6(a) of the Occupational Safety and Health Act of 1970. See below and ch. 12. )

American Conference of Governmental Industrial Hygienists The American Council of Governmental Indus- trial Hygienists (ACGIH) was founded in 1938 by a group of industrial hygienists from various levels of government. The ACGIH is a professional organization that issues “recommendations” that are developed by a straight membership vote. By 1968, ACGIH had adopted nearly 400 Threshold Limit Values (TLVs) for hazardous substances. The TLVs are 8-hour time-weighted aver- . . .

Ch. 11—A Short History of Private and Public Activities ● 207

age values that are suggested limits for workday 1969 by the Bureau of Labor Standards of the U.S. exposure; they are “guides for the control of health Department of Labor using the authority of the hazards” and were historically directed to the toxic Walsh-Healey Act (described below). Following rather than the carcinogenic, cytogenic, or muta- its establishment in 1971, OSHA adopted the genic properties of chemical substances. In a Walsh-Healey standards as its own, resulting in special appendix to the 1968 publication, ACGIH the TLVs published in 1968 by ACGIH becom- recommended no exposure to a list of carcinogens. ing occupational health standards for all U.S. in- ACGIH standards have been adopted by several dustry (249) (see also ch. 12). foreign governments and were incorporated in

GOVERNMENTAL EFFORTS

Workers’ Compensation Other employees might be crucial witnesses, but in the 19th century, when there were few govern- The first workers’ compensation program was mental or union job protections, anyone who tes- established in Germany by Otto von Bismarck in tified against an employer would risk being fired. 1884, and other European countries soon adopted More importantly, the law also established three their own programs (279), In the United States, powerful defenses that employers could use the most important early-20th-century activity of against lawsuits brought by employees. These State governments concerning occupational health were: and safety was the creation of workers’ compensation programs. A limited program was estab- ● negligence of other servants or co-workers, lished for Federal workers in 1908, and a num- ● knowledgeable assumption of risk by the em- ber of States established commissions to study ployee, and possible programs at about that time. Interest in ● contributory negligence by the injured workers’ compensation derived from several worker. sources. Progressive Era Aims The focus of these efforts was on the perceived deficiencies of the U.S. legal system concerning In the early 1900s a number of Progressive Era compensation for industrial injuries, Under humanitarian efforts underlined the plight of the Anglo-American common law, individuals can injured worker and paved the way for workers’ sue others for damages if a wrong has been com- compensation programs. Crystal Eastman con- mitted that causes harm. The basic duty of em- ducted the now-famous “Pittsburgh Survey” of ployers was to act with due care for employee 1907-08. She examined the economic conditions safety, as a reasonably prudent person would, and of the families of workers who had been killed to furnish a sufficient number of safe tools and or injured. In over half the cases, she found that equipment, as well as a sufficient number of qual- “the employers assumed absolutely no share of ified employees to do the work. Employers were the inevitable income loss. ” The costs of work ac- responsible for issuing and enforcing rules for cidents fell “directly, almost wholly, and in like- workplace safety, rules that with ordinary care lihood finally, upon the injured workmen and would prevent reasonably foreseeable accidents. their dependents. ” She concluded that a system Finally, there was a duty to warn workers of un- of compensation was necessary to achieve equity, usual hazards. social expediency, and prevention (274). In theory, if the employer failed to live up to At about the same time, a State commission in this standard of conduct, an injured employee Illinois reported that most court awards for in- could sue for damages under the common law. dustrial accidents were small, and that the fami- This was not always easy, however. The first dif- lies of the injured were often forced to live on ficulty was simply proving the employee’s case. charity. Moreover, for employers who had lia- 208 Preventing illness and Injury in the Workplace bility insurance, only 42 percent of payments went tional Civic Federation, the American Associa- to medical care. The remaining 58 percent went tion for Labor Legislation, and a number of the for administration, claims investigation, and legal leading industrialists of the day. expenses (100).

Employer< Attitudes Labor Union Reactions The apparently small awards made to most Unions, on the other hand, initially opposed workers was not the only reason for dissatisfac- workers’ compensation. They generally wanted tion with legal remedies. Employers, who as a workers to retain the right to sue employers and group supported workers’ compensation legisla- advocated abolition of the three common law de- tion before labor unions did, also found advan- fenses. They held this position in part because they tages in compensation programs. There is some thought injured workers would receive larger pay- evidence that just prior to the creation of work- ments under such a plan and because, at the time, ers’ compensation laws, injured workers, at least they generally mistrusted the government and in some circumstances, won a substantial portion feared that governmental intervention would of lawsuits against their employers (100,130,234, weaken unions (667). 316). Union opposition was also based on their per- Moreover, workers’ compensation substituted ception that workers’ compensation was “pallia- a regular, fixed, and predictable compensation tive and not preventive” (279). The belief that payment for uncertain, potentially ruinous liabil- workers’ compensation could provide an econom- ity judgments (274). Employers also feared that ic incentive for prevention was, according to Mac- without a workers’ compensation system, the Laury (279), important in changing labor’s posi- courts would start making more awards to injured tion; it “seemed to tip the scales. ” employees, especially if a worker could show that his/her employer had violated one of the increas- initial Legislation ing number of State safety regulations (249). The very first compensation acts, in Maryland, Finally, employers advocated workers’ compen- Montana, and New York, were ruled unconstitu- sation in order to remove one source of hostility tional. The Federal Government enacted a law in from labor-management relations and possibly to 1908, which covered only certain Federal employ- prevent more fundamental changes in the worker- ees. The first State law to become effective and employer relationship. They specifically opposed remain so was passed in Wisconsin in 1911. Fol- the passage of liability law reforms that would lowing this breakthrough and aided by the com- have eliminated the common law defenses of em- bined support of reformers, business, and labor, ployers. Some large companies had already estab- laws were passed rapidly. Four other States passed lished company benefit plans that provided pay- laws in 1911. By 1925,24 jurisdictions had enacted ments for work injuries. Smaller manufacturers compensation, although it wasn’t until 1948 that favored creation of such plans, but lacked the re- all the States had such laws (249,316,667). sources to do so privately (667). Larger manufacturers feared that if such plans were not created, In some cases, workers’ compensation was set legislators might act to change employer and em- up to supplement rather than to replace the legal ployee rights. In the absence of changes, it was liability system. Lubove states that “[w]here la- feared that the nascent unions would be given a bor had a major voice in shaping compensation boost (100,667). legislation, as in Arizona, injured workers were allowed a choice of remedy after injury.” It ap- For these reasons, some of the initial advocates pears that until that choice was removed under of workers’ compensation included groups like the business pressure a decade later, most injured National Association of Manufacturers, the Na- workers chose the liability route (274). Ch. II—A Short History of Private and Public Activities ● 209

Extent of Coverage The early lead of Massachusetts in establish- The initial laws covered only accidental injuries. ing regulations about workplace safety was fol- 1890, 22 Some state legislatures had no intention of com- lowed elsewhere. By States had passed regulations permitting safety inspectors in mines, pensating occupational diseases and specifically and 14 had factory and workshop inspectors. excluded them from coverage. Three reasons for this have been suggested. First, workers’ compen- However, these early laws were rarely enforced, sation laws were created to supplant common law partly because inspectors, who were often politi- liability. Under the common law, workers were cal appointees, were too untrained to recognize consistently denied recovery for occupational dis- even the most obvious safety hazards. eases. Second, it was thought that compensation In the first decades of this century, Alice Ham- for disease would be so expensive that it would ilton (box M) and other researchers actively pur- best be handled under a general health or disa- sued the work-relatedness of certain diseases. bility insurance program. For example, it was Their work was important in stimulating the in- believed that complete coverage of certain occupa- terest of State governments in occupational tional diseases, such as silicosis in foundries, health. By 1913, programs were organized in Con- mines, or quarries, would be extremely expensive necticut, New York, and Ohio. The first impor- for the compensation system and those particu- tant occupational disease laws, the “lead laws, ” lar industries (261). were passed in New Jersey, Ohio, and Pennsyl- vania (373). It has also been suggested that some of the writ- ers of the early workers’ compensation laws used Despite the establishment of these State occupa- language that would not alarm legislators, but tional health and safety programs in the early part would be flexible enough to allow the courts to of this century, these programs were often defi- extend coverage to occupational disease (46). cient. A 1964 study reported that, on average, Massachusetts was the first State to compensate there was only one occupational health staff mem- disease when the courts acted in just this way. But ber for every 108,000 workers, and there were by 1928, only 10 States covered diseases. From fewer than 1,600 safety inspectors in the various 1931-39, 14 States added coverage, while 18 States State programs combined (249). did so in the 1940s. The 7 remaining States added coverage between 1951 and 1967 (261). (See ch. A survey taken in 1968 found that occupational 20 15 for a discussion of the current workers’ com- health programs were in place in only States pensation system. ) and jurisdictions and that most States had more game wardens than safety inspectors. State occupational safety programs, which were much State Health and Safety Programs more highly developed than occupational health programs, covered only the mining portion of the Most early occupational health and safety ef- work force in 4 States, and 1 State had no safety forts in the United States occurred at the State legislation at all until 1967. State expenditures on level. Occupational safety laws were enacted by various States during the 19th century. As seems occupational safety ranged from 1 cent per employee in Wyoming to $2.70 per worker in Oregon. to be often the case, there appears to have been In addition, as late as 1969, only 21 States gave a tendency to direct the laws at what were, at the safety inspectors the right to shut down a work time, new technologies. For instance, in 1852, Massachusetts passed a law regulating steam en- area that presented an imminent hazard (249). gines and permitting State inspectors the “power Moreover, even in the States with occupational of closure” in situations of grave hazard (249). health programs, the powers to develop and en- Twenty-five years later, in 1877, the same State force occupational health laws and to inspect passed the first factory-inspection law that re- worksites were often diffused through several quired the installation of certain safety devices agencies. The resulting fragmentation of powers (guarding of belts, shafts, and gears), fire exits, contributed to the difficulties of enforcing State and protection on elevators (279). occupational health laws. 210 Preventing Illness and Injury in the Workplace

A final complication of separate State laws is illustrated by what happened when Pennsylvania banned the use, manufacture, storage, and handling of beta-naphthylamine, a carcinogen. Soon after the ban, another facility in another State began producing the chemical (249). This State- by-State approach was also criticized by business representatives. A keynote speaker at a trade association meeting in Washington in 1973 noted: When . . . [there is] a proliferation of different state plans and state enforcement . . . American business [has] great difficulty because most . . . companies . . . are multiproduct, multiplant companies. Having to live with . . . 40 or 50 different approaches . . . as distinguished from a single set of rules . . . concerns me greatly (277a). Thus, by the time Congress considered Federal occupational health and safety legislation in the late 1960s, there was widespread agreement that, as one historian has summarized it, “safety and health laws, historically left to the States, were piecemeal, varied in quality, and often unenforced” (124).

Early Federal Government Programs As early as 1790, the First Congress appeared to take an interest in the safety of merchant seamen by giving the crew of a ship at sea the right to order the vessel into the nearest port if a majority of the seamen plus the first mate believed it unseaworthy (279). In 1798, the Marine Hos- pital Service, which evolved into the Public Health Service, was established to provide care to seamen disabled on the job. The Hospital was paid for through the first system of health insurance in this country: a tax of 20 cents annually deducted from all seamen’s wages (249). In the early 19th century, the Marine Hospital’s physicians were primarily concerned with the control of epidemic diseases, such as cholera and yellow fever, which was more in the realm of public health than worker health. Federal employees benefited from several measures passed in the 1800s: an 1833 law granted compensation to disabled seamen; an 1868 law limited the workday of Federal employees to 8 hours; in 1908 and 1916, workers’ compensation Ch. 11—A Short History of Private and Public Activities . 211 was enacted for Federal railroad and other em- The passage of the Social Security Act in 1935 ployees. Several early attempts (1796, 1852) at led to an expansion of PHS studies of occupational Federal intervention in matters of public health hazards and to the provision of grants to States (such as enforcement of maritime quarantine and for public health work, including industrial hy- State grants to establish asylums for the insane) giene activities. During the 1930s, the Public were rebuffed by the States, which considered Health Service studied the health effects of lead public health their responsibility (249). in gasoline and of fumes of chromic acid and mercury in the workplace. Federal Research and Assistance A 1937 reorganization of the PHS resulted in in Occupational Safety and Health the Scientific Research Division being consoli- dated with the National Institutes of Health (249). The Public Health Service (PHS) activities were The Office of Industrial Hygiene and Sanitation extended to the workplace in 1914 when the Of- along with the Office of Dermatoses Investigations fice of Industrial Hygiene and Sanitation was es- became the Division of Industrial Hygiene of the tablished in the Division of Scientific Research National Institutes of Health. Seven years later, (249). During the next 20 years, the office engaged because of the marked increase in its work with in research to identify occupational health hazards States, the Division was transferred to the Bureau and their effects. It studied lead poisoning, looked of State Services. at hazards in brass foundries and in the glass and chemical industries, and made sanitary surveys In World War II, as in World War I, the PHS in war plants during World War I. It also con- was concerned with the protection of employee ducted studies about the physiological effects of health in government-owned, privately operated lighting, high temperature, fatigue, and other en- munitions plants. A dramatic illustration of the vironmental conditions in the workplace. success of these efforts is provided by mortality associated with TNT manufacture in both wars. A study in 1924 followed up the deaths of 20 During 7 months of World War I, 475 workers people who had been employed in the painting died and 17,000 were disabled by fumes; in World of radium watch dials. Recommended control War II, there were 22 deaths in 35 months. In ad- measures subsequently ended radium poisoning dition, studies were carried out in aviation medi- in the watch industry. Less spectacular, but bear- cine and on the health effects of new chemicals ing on the health of many more workers, was the and metals, such as vanadium and beryllium, Office of Industrial Hygiene and Sanitation’s newly introduced into airplane manufacture. study of the dusty trades. Begun in 1923, this research showed the extent of health impairment as- Other reorganizations followed World War II. sociated with the granite, pottery, cement, cot- In 1953, the Federal Security Agency, which had ton textile, and mining industries. included the Public Health Service, was abolished, and the Department of Health, Education, and During this same early-20th-century period, the Welfare (HEW) was established. The Division of U.S. Bureau of Labor, and its successor, the De- Industrial Hygiene became the Occupational partment of Labor, also sponsored research on oc- Health Program and remained in the Bureau of cupational health and safety. Its reports included State Services until it was designated the Occupa- studies of lead poisoning, phosphorus-caused dis- tional Health Program in the Bureau of Disease ease, the dusty trades, and industrial accidents Prevention and Control in 1966. Other organiza- (124). A 1910 report on “phossy jaw,” published tional moves within the Department followed, by the Bureau, was important in revealing the na- and it was renamed the Bureau of Occupational ture of that occupational health problem. In the Safety and Health (249). second decade of this century, the Bureau employed Alice Hamilton to investigate occupational The Occupational Safety and Health Act in hazards especially in the “dusty trades” and pub- 1970 established the National Institute of Occupa- lished the results of her studies (279). tional Safety and Health (NIOSH), and in 1973 212 Preventing Illness and Injury in the Workplace

NIOSH was transferred to the Centers for Dis- spectors were permitted to close certain establish- ease Control of the PHS. During all the bureau- ments operated by employers guilty of repeated cratic reorganizations since it began in the Pub- serious violations. lic Health Service, the occupational health A disastrous explosion in 1968 killed 78 miners program has produced important studies about in Farmington, WV. This crystallized public at- occupational health and, more recently, safety. tention on mine safety. Citing the Federal Gov- ernment’s “fatalistic attitude” and failure . . . “to Federal Legislation and act vigorously to change [the prevailing bad prac- Regulatory Programs tices], ” the 91st Congress passed the Coal Mine Health and Safety Act of 1969 (the Coal Act) Federal regulatory attention has historically (249). focused on several high-hazard industries, such as mining, longshoring, railroading, and construc- Despite the increase in authority given to the tion. The coal mining industry provides an ex- Federal Government by the 1969 Coal Act, safety ample of Federal efforts to control occupational and health conditions in the mines continued to hazards in these industries. be unacceptable to the Congress; in 1973, approximately one of every 1,500 miners, compared with Coal Mine Safety Legislation one of every 12,400 workers in general industry, were reported to have been killed. In 1977, Con- In the wake of the Monongah, WV, coal mine gress passed the Federal Mine Safety and Health disaster of 1907 that killed 362 miners and other Amendments Act, the first Federal law to consoli- accidents that caused the loss of many lives, the date jurisdiction over both coal and metal mines Bureau of Mines was established in 1910 within and all safety and health matters (except training the Department of Interior to promote mine safe- and research) in one executive department. The ty. Bureau personnel were specifically denied “any Department of Labor was empowered by this Act right or authority in connection with the inspec- to inspect mines, and to develop, promulgate, and tion or supervision of mines.” Although the pow- enforce safety and health standards applicable to erlessness of Bureau personnel was widely de- mines (249). plored, it was not until 1941 that the Federal Coal Mine Health and Safety Act, which granted inspection authority to the Bureau, was passed. An Legislation for Other Industries excerpt from the House of Representatives report The New Deal. —Frances Perkins, selected by accompanying the bill captures the sense of Con- President Franklin D. Roosevelt in 1983 to be the gress at the time: Secretary of Labor, brought experience in occu- Investigation reveals no common standard of pational safety and health to that position. In safety among the States, no common regulations, 1934, she created within the Department the Bu- and, in addition to this, a lack of uniform enforce- reau of Labor Standards, the first permanent Fed- ment of such [State] regulations as are in effect eral agency with a mandate to promote safety and . . . . In order to supplement the work of the health for the entire work force. To a major ex- State agencies, the bill under consideration extent, the Bureau acted by aiding the States in the tends and enlarges the authority of the Federal Bu- administration of their workplace health and safe- reau of Mines. It is not regulatory in any sense. ty laws and by promoting protective legislation. It merely authorizes the Bureau, through its representatives, to make inspections of the under- Three New Deal-era laws contributed to a ground workings and publicize its findings and growing Federal involvement in occupational safe- recommendations. These inspections , . . are to ty and health. As noted above, the Social Secu- be made in conjunction with the local State agen- rity Act of 1935 provided for the Public Health cies so that there is no assumption of the State Service’s funding of State industrial health pro- authority (quoted in 249). grams. In addition, the Fair Labor Standards Act Later laws gradually increased Federal author- of 1938 (the “minimum wage law”) allowed the ity over coal mine hazards; in 1966, Bureau in- Labor Department to bar employment of persons Ch. 17—A Short History of Private and Public Activities 213 . under 18 in dangerous jobs, while the Walsh- The Federal Bureau of Labor Standards almost Healey Public Contracts Act of 1936 directed the never used its inspection and enforcement powers; Department to ensure standards for safe work by in 1969, only 5 percent of the 75,000 firms cov- Federal Government contractors and to “blacklist” ered by Walsh-Healey were inspected. At the contractors who did not comply with the stand- 3,750 worksites inspected by the Bureau in 1969, ards (279). This last act is of particular interest a total of 33,000 violations of safety regulations because it created a Federal regulatory role con- were recorded, while only 34 formal complaints cerning job safety and health and located this were issued. Two companies were blacklisted function in the Department of Labor. (prohibited from bidding on Federal contracts) in 1969, and three had been similarly treated in 1968 Walsh-Healey Act.–This legislation covered all (361). employees working for employers who had contracts with the Federal Government that exceeded The history of events under the Walsh-Healey $10,000 in total value. The McNamara-O’Hara and other acts exemplifies the sporadic efforts of Act of 1966 and the Construction Safety Act of the Federal Government to control occupational 1969 extended Federal regulation to service con- hazards in the years before OSHA. However, a tract employers and Federally funded construc- pattern of increasing Federal involvement, such tion employers, respectively. Employers were re- as the progression that occurred in mining, can quired by the terms of their contracts to comply be seen, particularly in the extra-hazardous with Walsh-Healey safety and health standards, trades—maritime, railroading, and construction. which were recommended by the Bureau of La- In each case, the first laws permitted Federal per- bor Standards of the Department of Labor. The sonnel to inspect specific aspects of hazardous Bureau of Labor Standards was also given the au- operations, such as man-cages (personnel hoists) thority to inspect workplaces covered by the in mines, air brakes on trains, and shackles and Walsh-Healey Act and had the power to prohibit other rigging components in longshoring. This employers who violated the act from bidding on first stage was gradually followed by Federal Federal contracts for a period of three years. assumption of the responsibility for developing Federal involvement in setting safety and health or recommending standards, helping employers standards intensified in the late 1950s. In 1958, to comply with them (and only rarely, in cases an amendment to the Longshoremen’s and Har- of grave danger, using the power of closure), and bor Workers’ Compensation Act extended the finally enforcing them. In all cases, the creation Federal role in protecting safety and health in the of Federal agencies with inspection and enforce- hazardous maritime trades. The amendment au- ment authorities required more than a half-cen- thorized the Labor Department to set standards tury from the time of initial congressional action in those trades and to seek penalties against em- (249). ployers who willfully violated safety and health The Occupational Safety and Health Act of standards, Compliance with the standards was 1970.—One result of the strong criticisms voiced good after enforcement began in 1960, and acci- before and during the 1964 Department of Labor dent rates in the maritime trades declined (279). hearings was a decision of the Department to ex- Acting on its own in December 1960, the La- amine its safety and health policy. A study by an bor Department issued a set of mandatory safety independent consultant characterized the Labor and health standards under the Walsh-Healey Act. Department’s safety laws and programs as frag- However, objections were raised to the rigidity mented (279). During this period of self-examina- of the rules that the Federal Government required tion, the environmental movement was attracting State occupational safety agencies to enforce when public and congressional support in its bids for they inspected Federal-contract workplaces. The Federal laws to protect human health and the envi- criticisms were heeded by the Department, and ronment from the effects of pollution. The envi- hearings about the Federal standards were held ronmental movement spilled over into questions in March 1964. of occupational health because of the attention 214 . Preventing Illness and Injury in the Workplace

paid to chemicals as risks to health and the rea- gave the Secretary of Labor the responsibility of sonable extension of “environment” to include the setting and enforcing standards to protect 50 mil- workplace. lion workers. The bill also had a general duty clause requiring employers to “furnish employ- The Public Health Service published Protecting ment and a place of employment which are safe the Health of Eighty Million Americans (the “Frye and healthful. ” It gave inspectors legal authority Report”) in 1965, which drew attention to threats to enter workplaces without management’s per- to health from new technologies. Although it mission or prior notice. Violators could be pun- highlighted the evidence that some chemicals were ished with civil or criminal penalties. Interested associated with cancer causation, it also empha- states could develop their own occupational health sized that many “old” occupational health prob- and safety programs to replace the Federal one. lems had not been remedied. The report suggested The Department of HEW would provide the La- an approach to improve occupational health that bor Department with scientific information (279). would require a major new effort from the PHS. The AFL-CIO urged President Lyndon Johnson Although hearings were held, that bill did not to respond to the PHS report’s recommendations reach the floor of either the House or the Senate. (279). Part of the reason was the opposition of business, With the President expressing interest in occu- particularly the Chamber of Commerce, to be- pational safety and health, both the Labor De- stowing so much power on the Secretary of La- partment and the Department of Health, Educa- bor and to undermining the role of the States in tion, and Welfare began development of legis- occupational safety and health. Also important lation for a Federal program in occupational safety were other events of 1968: Riots in the inner cit- and health. The departments deadlocked on the ies, protests against the war in Vietnam, and Presi- issue of which one would control the national pro- dent Johnson’s decision not to run for re-election gram in late 1966, and the effort stalled (279). competed with occupational safety and health for public and congressional attention. A dramatic bureaucratic action led the Bureau of the Budget to accept the Department of Labor’s In 1969, Congress passed the Coal Act (see recommendations for legislation rather than those above) and President Richard Nixon introduced of HEW. In 1967, it was learned that abnormally a new version of an occupational safety and health high numbers of uranium miners were dying of law for all U.S. workers. His version skirted the lung cancer. Later that year, the Federal Radia- issue of whether the Department of Labor or HEW tion Council, composed of representatives from was to have the lead in the Federal program. The a number of Federal agencies, met to consider pro- duty of Labor was to inspect workplaces for com- tective measures for uranium miners. They came pliance with standards. The role of HEW was to to an impasse concerning the standard proposed carry out research. The important function of is- by the Atomic Energy Commission versus the suing safety and health standards would be vested more stringent standard proposed by the Depart- in an independent, five-person standards-setting ment of Labor. Unhappy with their indecision, board. The Nixon proposal also stressed the use Secretary of Labor Willard Wirtz adopted the proof existing State government programs and pri- posed Department of Labor standards under the vate industry efforts (279). provisions of the Walsh-Healey Act the very next Objections to the Nixon bill were raised by day. This bold move was instrumental, accord- many Democratic and some Republican congress- ing to MacLaury (279), in the Bureau of the Bud- men. Their concerns involved the independent get accepting the Department of Labor’s recom- standards-setting board, because of the adminis- mendations about legislation. trative confusion it would cause and its lack of In January 1968, President Lyndon Johnson political accountability. Labor unions, in particu- called on Congress to pass job safety and health lar, opposed this board, preferring instead that legislation closely modeled on the recommenda- the authority to set standards be given to the Sec- tions of the Department of Labor. The proposal retary of Labor. In addition, objections were Ch. 11—A Short History of Private and Public Activities ● 215 raised to the bill’s enforcement scheme because ous operations was narrowly rejected. (In addi- it would penalize only willful, flagrant violators. tion, there was some debate on the criteria for Finally, the bill’s reliance on industry-written standards. This is discussed in ch. 14). “consensus” standards, exemptions for small em- In the House, Congressman Steiger (R-WI) pro- ployers, and a three-year delay in its effective date posed as a substitute an amendment that repre- were also points of criticism (279). sented a modification of the original Nixon pro- In response, Democrats in the House and Sen- posal and this substitute was adopted. The ate had already introduced their own bills and conference committee had to resolve a large num- both the House and Senate committees reported ber of differences. They used the framework of to the floor bills sponsored by Democratic mem- the mom liberal Senate bill. The single most im- bers. In the Senate, Peter Dominick (R-CO) pre- portant change from the Senate version was the sented a substitute bill that would have established deletion of a provision that allowed the Secretary two independent boards—one to issue standards of Labor to close down a plant under conditions and one to decide enforcement appeals. This was of “imminent danger. ” Under the provisions of rejected by a two-vote margin. Then Jacob Javits the bill that emerged from conference, the Secre- (R-NY) introduced an amendment that gave the tary is required to obtain a court order before clos- Secretary of Labor the authority to set safety and ing a plant that poses an imminent danger. Presi- health standards and established a separate com- dent Nixon, through the Secretary of Labor, let mission to oversee Department of Labor enforce- it be known that he approved the bill, and both ment of the standards. This amendment was the House and Senate passed the Occupational adopted, although another amendment to restrict Safety and Health Act of 1970 (279). the authority of inspectors to close down hazard- .> . .-

. . .

1 2 m Governmental Activities Concerning Worker Health and Safety ——. — — . . ‘. , ~. ., ● .:. -.~.%. + ,; ‘., + y’ , ,:, , - . . - Cmtt?m!t $: , ‘,..* _ . # <.. J,”.,.Z -* . .“ . - *, ; -’ ..:, .*F . . .--,+?Z’ ‘ .~’”. ‘“” .- . ~ . :- ,’! . . . ‘ ‘. Page . “. Current Federal/State Framework ...... 219 Separation of Research and Regulation ...... 220 Federal Spending for Occupational Health and Safety ...... 221 Assistant Secretaries and Directors ...... 221 +., --r The Occupational Health and Safety Administration ...... 224 ., . OSHAStandard*setting..•• ● ● . . . ● . . . . . ● ● . . ● ● ...... ● ...... ● ...... 224 Enforcement ...... 232 Public Education and Service ...... 237 , State programs and Other Federal Agency Activities ...... 239 .. The National Institute for Occupational Safety and Health ...... 242 Identification ...... 243 Development of Controls ...... $ ...... 247 Dissemination ...... 249 NIOSH Priorities...... 251 Summary ...... 253

LIST OF TABLES Table No. Page 12-1. Federal Spending for Worker Health and Safety ...... 222 12-2. OSHA Assistant Secretaries and NIOSH Directors ...... 223 12-3. OSHA Health Standards ...... 228 12-4. OSHA Safety Standards ...... 229 12-5. Federal OSHA Safety and Health Inspections...... 232 12-6. OSHA Onsite Consultation Program ...... 238 12-7. State Programs ...... 240 128. NIOSH Budget for Control Technology...... 248 12-9. Control Technology Assessments Performed by NIOSH in Fiscal Year 1982 ...... 248

LIST OF FIGURES Figure No. Page 12-1. OSHA Budget 1971-85 ...... 223 12-2. NIOSH Budget 1971-85 ...... 223 12-3. Percent Change in Appropriations for Selected Health-Related Agencies 1980-84 ...... 223 12-4. Occupational Safety and Health Administration ...... 225 12-5. National Institute for Occupational Safety and Health ...... 243 — — 12 Governmental Activities Concerning Worker Health and Safety

This chapter has three major sections. The first vialing public education and services, and moni- describes the framework of U.S. Government ac- toring the performance of State programs. The tivities created by the Occupational Safety and third section describes the activities of the Na- Health (OSH) Act, while the second discusses the tional Institute for Occupational Safety and main activities of the Occupational Safety and Health (NIOSH) in hazard identification, research Health Administration (OSHA): setting stand- on controls, and information dissemination. ards, inspecting and enforcing regulations, pro-

CURRENT FEDERAL/STATE FRAMEWORK

In the Occupational Safety and Health Act of and proposes penalties and prescribes abatement 1970, Congress authorized the creation of three plans for employers found violating the standards. agencies to set and enforce mandatory health and In addition, OSHA provides for public, worker, safety standards; to conduct research on occupa- and employer education and consultation, mostly tional hazards and their control; and to review through grant activities. Finally, OSHA partially contested enforcement actions. The three agen- finances the operations of State agencies operat- cies are OSHA, NIOSH, and the Occupational ing “State plans” and monitors their performance. Safety and Health Review Commission (OSHRC). NIOSH is a research agency that is part of the One other Federal agency, the Mine Safety and Centers for Disease Control (CDC) of the U.S. Health Administration (MSHA) of the U.S. De- Public Health Service, which is part of the De- partment of Labor, specializes in worker health partment of Health and Human Services (HHS). and safety. It is responsible for the health and NIOSH is headed by a Director appointed by the safety of workers in coal mines, as well as in other Secretary of HHS for a term of six years. It was metal and nonmetal mines. It was created as a re- created from what had been the Bureau of Occu- sult of the Metal and Non-Metallic Mine Safety pational Safety and Health and started operations Act of 1966, the Coal Mine Health and Safety Act as NIOSH on June 30, 1971. Congress mandated of 1969, and the Mine Safety and Health Act of that it conduct research and related activities on 1977 (333). Its activities are not described in detail developing criteria or recommendations to be used in this assessment. by OSHA in setting standards, on identifying and evaluating workplace hazards, and on measure- OSHA is a regulatory agency that sets and en- ment techniques and control technologies, as well forces regulations concerning the control of health as provide professional education and disseminate and safety hazards. It began its operations on health and safety information. April 28, 1971. Part of the Department of Labor, it is headed by a Presidentially appointed Assist- The OSHRC has three members appointed by ant Secretary of Labor for Occupational Safety the President, with the advice and consent of the and Health, to whom the Secretary of Labor has Senate, for staggered terms of six years. Its duties delegated authority to administer the OSH Act. are limited to reviewing and resolving disputes OSHA sets mandatory health and safety stand- concerning OSHA citations and penalties. In do- ards, inspects workplaces to ensure compliance, ing so, the Review Commission interprets the

219 220 . Preventing Illness and Injury in the Workplace meaning of OSHA standards and thus determines ulatory agency. The congressional debate con- the nature and scope of many employer obliga- cerning research focused only on the need to tions concerning employee health and safety. enhance occupational safety and health research and to “elevate [its] status” (551). In addition, the act created a temporary commission to examine the workers’ compensation Separating research from regulatory activity system—the National Commission on State may also help ensure the quality and objectivity Workmen’s Compensation Laws. It also created of the research. On the other hand, separation a permanent advisory body for OSHA, known may lead to inefficiencies, especially when the as the National Advisory Committee on Occupa- activities of the two agencies are poorly coordi- tional Safety and Health. nated. One observer argued in 1976 that “the enforcement function and priority setting in OSHA Both OSHA and NIOSH have been criticized are barely connected to the research and man- since their creation in 1971. OSHA has been called power development mandated for NIOSH” (30). to task by employers and their representatives for In 1977, the General Accounting Office concluded issuing standards that are excessively expensive, that OSHA and NIOSH needed to improve the overly stringent, not based on scientific evidence, coordination of their activities concerning the de- or unrelated to employee health and safety. La- velopment of workplace health standards (501). bor unions, on the other hand, have criticized In 1978, an Interagency Task Force also made rec- OSHA for failing to devote adequate resources ommendations to better coordinate NIOSH re- to enforcement, for delaying or failing to set new search with OSHA’s needs (228). The two agen- regulations, and for considering employers’ costs cies have created mechanisms to coordinate as a basis for health and safety decisions. activities, although OTA has not attempted to de- NIOSH, too, has been accused of having failed termine how well these agencies work together to fulfill its mission. OSHA has complained about today. the inadequacy of NIOSH’s criteria documents for Besides being separated from OSHA, NIOSH OSHA standard-setting. NIOSH has been criti- is lower than OSHA in the Federal bureaucracy. cized by the General Accounting Office for the Since July 1, 1973, NIOSH has organizationally quality of its criteria document and Health Haz- been part of the Centers for Disease Control. Thus ard Evaluation programs. Labor groups have crit- the Director of NIOSH reports to the Director of icized NIOSH for being unresponsive to worker CDC, who in turn reports to the Assistant Secre- requests. Management representatives have tary for Health, who reports to the Secretary of claimed that Health Hazard Evaluations were too HHS. OSHA, on the other hand, is headed by aggressively pursued and that NIOSH research an Assistant Secretary of Labor who reports di- was of poor scientific quality. rectly to the Secretary of Labor. Some Members of Congress have criticized the Separation of Research and Regulation placement of NIOSH within CDC. For example, in 1973, three Senators argued that this was “con- The OSH Act separated occupational health trary to the expressly stated intent of Congress and safety research activities from standard- in creating NIOSH, which was to elevate the setting and enforcement by placing these respon- status of occupational safety and health research sibilities into two different departments of the Fed- in HEW from its relatively low level in 1970. . . “ eral Government. This may, in part, simply be In addition, they criticized the average Federal the result of the history of Federal activities, which personnel grade levels that had been established prior to the OSH Act had been found in both the for NIOSH because they were substantially lower Department of Labor and the Public Health Serv- than those for OSHA personnel and personnel at ice of HHS and their predecessor agencies. Dur- the Environmental Protection Agency (232). ing congressional consideration of the OSH Act (see ch. 11), labor unions strongly supported des- The geographical location of NIOSH has also ignation of the Labor Department as the lead reg- generated considerable interest. In 1981, it was Ch. 12—Governmental Activities Concerning Worker Health and Safety ● 221 announced that the NIOSH headquarters would The reduction since 1980 is illustrated in fig- be moved from Rockville, MD, to Atlanta, GA, ure 12-3, which shows the percentage changes in where the headquarters of CDC is located. (Most appropriations for several health-related agencies. NIOSH staff, however, continue to work in the Funding of agencies with responsibilities for reg- NIOSH laboratories in Cincinnati, OH, and Mor- ulating health and safety in the workplace has in- gantown, WV. ) In support of the move of NIOSH creased in current dollars over the last 4 years headquarters, it was suggested that the scientific (OSHA and MSHA), while overall funding for and technical base of NIOSH would be strength- NIOSH has consistently decreased. Overall, ened through greater interaction with other CDC NIOSH’s share of Federal spending for occupa- programs (303). Further, it was thought that tional health and safety has been declining. In NIOSH would benefit from CDC’s expertise in fiscal year 1974, the NIOSH budget of $35.4 mil- disease and health hazard surveillance and that lion was about half as much as the OSHA budg- there would be greater NIOSH involvement in et of $70.1 million. From fiscal year 1975 through environmental health. Cost savings of approx- fiscal year 1980, the NIOSH budget varied from imately $1.5 million per year were predicted to being 34 to 42 percent as large as the OSHA bud- result from this action. get. The President’s budget request for 1985 proposed funding NIOSH at about 26 percent of the But many people from labor, management, aca- level of OSHA funding. demia, and the occupational safety and health professions believed that moving would be detri- The authorized numbers of personnel for both mental to NIOSH’s ability to perform its man- OSHA and NIOSH generally rose from 1971 to dated responsibilities. In 1981 and 1982, Congress peaks in 1980. From 1980 to 1984, the number attached a restriction on the appropriations for of authorized OSHA positions decreased about NIOSH that prohibited this move. At the end of 25 percent, while NIOSH positions dropped about 1982, this restriction was lifted and the move to 16 percent. Atlanta was completed shortly thereafter. Assistant Secretaries and Directors Federal Spending for Occupational Health and Safety Table 12-2 lists the men and women who have served as Assistant Secretaries for Occupational Table 12-1 presents the total budgets (in cur Safety and Health and Directors of NIOSH, The rent dollars and in real, inflation-adjusted dollars) first Assistant Secretary for OSHA was George for OSHA and NIOSH and the authorized per- Guenther, who had been head of the Bureau of sonnel ceilings. Figures 12-1 and 12-2 present the Labor Standards (the Labor Department prede- budget totals graphically. Although OSHA’s bud- cessor agency to OSHA). He was responsible for get in current dollars has grown over time, in real the issuance of OSHA’s first standards and the dollars it peaked in fiscal year 1979, decreasing beginnings of OSHA’s inspection activity. Within nearly 13 percent by fiscal years 1982 and 1983. 2 years he was replaced by John Stender, who had been an official of the Boilermaker’s Union and In current dollars the NIOSH budget grew from a State legislator, but who had no previous pro- 1971 to 1980, but since then it has been substan- fessional background in occupational health and tially reduced, both in current, nonadjusted safety. He presided over the growth of the agency, dollars, as well as in real terms. After adjusting dramatically increased the number of inspections for inflation, the 1983 NIOSH budget is the lowest conducted, and encouraged the development of since 1973 and represents a 42 percent decrease State plans. since the peak in 1980. The 1984 budget includes an increase in current dollars over 1983, largely In late 1975 Stender was replaced by Morton because Congress restored NIOSH funding for Corn, a professor of industrial hygiene at the professional training programs. The 1985 budget University of Pittsburgh, who served as Assistant proposed by the President would reduce the Secretary for just over 1 year. Corn took steps NIOSH budget by completely eliminating these to increase OSHA activity concerning health funds. standards and to conduct more health inspections.

NA NA NA NA NA 1,139 1,175 1,081 1,036 1,105 1,110 1,132 1,116 positions Inspector Federal Government b

NA NANA NA NA NA 13.0 17.0 22.8 22.1 21.0 21.5 State programs Federal grants b For 1971, NA NA NA NA NA NA NA 28.7 21.2 24.6 42.344.1 47.6 49.6 20.7 (millions) (millions) Federal grants (current dollars) (1972 the operating costs of State program

percent for 784 783 785 785 501 745 735848 841 16.7 898 913 37.0 611 611 932 40.0

positions Personnel a

NA NA 19.1 33.2 29.6 34.3 38.5 26.5 44.5 Budget

this table, these grants were

56.4 57.5 25.9 26.3 26.3 25.9 24.5 39.3 28.9 65.9 56.3 36.6 913 31.4 20.4 34.0 62.6 80.4 62.1 67.8 Budget (millions) (millions)

National Institute for Occupational Safety and Health (current dollars) (1972 For the years and

1,696 1,699 2,354 2,258 2,259 2,817 2,944 3,009 2,717 49.7 3,015 positions Personnel

Table 12.1.--Faderal Spending for Worker Health and Safety to the States by Federal Standards Administration a 8.2 970 17.7 NA NA NA 93.1 33.9 61.0 1,83093.1 35.4 30.8 84.790.0 2,494 90.2 on data supplied by 102.4 106.4 103.1 Budget grants

for the

C C 7.6 70.1 33.9 69.395.8 65.6 74.7 2,435 only the 114.9 130.2 172.8 209.4 212.6 217.8 Budget Occupational Safety and Health Administration (millions) (millions) (current dollars) (1972 financed from the . d 1981 . 1984 19850 . 1983...... 1983. 206.6 1982...... 1982. 195.5 1980, . 186.4 1976 . 1977 . . . . . 1978. 138.7 1979. . 1971 ..., . 1972 . 1973 . 1974 . 1975 ...,. NA–Not SOURCE. Office of Technology Fiscal year

Ch. 12—Governmental Activities Concerning Worker Health and Safety 223

Figure 12-3.—Percent Change in Appropriations for Selected Health.Reiated Agencies 1980-84

‘OSHA MSHA’ NIH ‘NIOSH’ CDC ‘ CDC (excluding 1 I I 1 1 Agencies NIOSH) 1971 1973 1975 1977 1979 1981 1983 1985 SOURCE Budget of the United States Government Year Table 12-2.—OSHA Assistant Secretaries and NIOSH Directors

Figure 12-2.— NIOSH Budget 1971-85 OSHA 1. George Guenther ...... April 1971-January 1973 2. John Stender ...... April 1973-July 1975 3. Morton Corn ...... December 1975-January 1977 4. Eula Bingham ...... April 1977-January 1981 5. Thorne Auchter ...... March 1981-March 1984 6. Robert Rowland ...... July 1984-Present NIOSH 1. Marcus Key ...... June 1971-September 1974 2. John Finklea...... September 1974-January 1978 3. Anthony Robbins ...... October 1978-May 1981 4. J. Donald Miller ...... June 1981-Present SOURCE Office of Technology Assessment

She acted to eliminate a number of the “nit-picking” standards, for which OSHA had been criticized, and emphasized the development of health standards and “generic standards” (those that would cover exposures to a group of substances, such as carcinogens, or would provide worker access to information, such as employer records con- Year cerning exposures, medical care, and chemical substance identity). She also established the New Directions grants program and increased the num- He also worked to improve the professional ex- ber of OSHA-funded onsite consultative visits, pertise of OSHA staff, especially its inspectors, especially for small businesses. who had been criticized for their inexperience. Theme Auchter, a construction firm manager, In 1977, Eula Bingham was named to head took office in 1981. He emphasized a ‘balanced” OSHA. She had been a professor of toxicology approach to OSHA activities and improved man- and had served on an OSHA advisory commit- agement of agency operations, established a new tee concerning the coke oven emissions standard. approach for “inspection targeting, ” and en- 224 . Preventing Illness and Injury in the Workplace couraged cooperation with employers, especially Health Service, was appointed as the first NIOSH concerning negotiations about citations, fines, and Director. He focused on making NIOSH a func- abatement. Auchter reconsidered a number of the tioning organization. Key stepped down in Sep- standards issued in previous administrations, re- tember 1974, after 3 years of his 6-year appoint- duced the funding for the New Directions Pro- ment. John Finklea became the second Director gram, increased the funding for onsite consulta- of NIOSH in September 1974 and also served 3 tion, and encouraged the development of State years. He emphasized the production of NIOSH programs. He resigned in March 1984. criteria documents to be delivered to OSHA. In July 1984, Robert Rowland was named to Anthony Robbins, a former Commissioner for head OSHA. An attorney, he had practiced law Public Health in Vermont and Colorado, became privately before being appointed as the Chairman the third Director in October 1978. He deem- of the Occupational Safety and Health Review phasized criteria documents and focused on health Commission. He served in that position from hazard evaluations and epidemiological field August 1981 until his appointment as Assistant studies. He, too, did not complete a 6-year term, Secretary for OSHA. but was fired by HHS Secretary Schweiker in 1981. J. Donald Millar took office as the fourth In June 1971, Secretary Richardson of the De- NIOSH Director in June 1981. Millar has taken partment of Health, Education, and Welfare an- several steps to improve the quality of NIOSH nounced the establishment of NIOSH. Marcus research. Key, M. D., previously chief of the Bureau of Oc- cupational Safety and Health in the U.S. Public

THE OCCUPATIONAL HEALTH AND SAFETY ADMINISTRATION

Figure 12-4 presents the organization of OSHA. The purpose of the general duty clause is to pro- The main activities of its national office in Wash- vide employee protection from some of the ington, DC, are performed by seven directorates, hazards that are not currently addressed by which specialize in developing health and safety OSHA’s more detailed regulations. The interpre- standards, coordinating operations of OSHA’s in- tation of the “general duty clause” is complex and spectors, providing educational and service pro- controversial (see brief discussion later in this grams, monitoring State plans, and furnishing chapter). administrative and technical support. However, Section 5(a)(2), the “specific duty clause, ” re- the majority of OSHA staff, including its inspec- quires employers to comply with the more de- tors, are assigned to area offices that are grouped tailed standards issued by OSHA. These stand- into 10 different regions. ards are Federal regulations that, for example, require employers to observe certain precautions, OSHA Standard-Setting conduct their operations in specified ways, or install and use certain kinds of equipment. The OSH Section 5(a)(1) of the OSH Act, the “general Act provided OSHA, in sections 6(a), 6(b), and duty clause, “ imposes a general requirement that 6(c), with three different methods to issue health workplaces must be kept safe and healthful. It and safety standards. The standards issued under provides that each employer: these three methods have been termed interim, . . . shall furnish to each of his employees em- new or permanent, and emergency temporary ployment and a place of employment which are standards (333,408). But many of the “interim” free from recognized hazards that are causing or standards have not been changed since OSHA are likely to cause death or serious physical harm began. At the same time, OSHA has taken ac- to his employees; . . . tions to change some of the “permanent” stand-

Ch. 12–Governmental Activities Concerning Worker Health and Safety ● 225 226 . Preventing Illness and Injury In the Workplace ards. Mintz (307) suggests that OSHA standards a regulation that mandated a specific shape for be termed: startup standards, standards issued toilet seats (406). In 1978, under Eula Bingham, after rulemaking, and emergency temporary OSHA rescinded about 600 of these “nit-picking” standards. provisions that applied to general industry and about 300 that applied to several special indus- Startup Standards tries. However, the process of revising the startup standards continues. Congress mandated that OSHA adopt, without additional rulemakingl, those health and Of more serious consequence, many types of safety standards that had already been established hazards are not adequately addressed by the star- by Federal agencies or had been adopted as na- tup standards. A Presidential task force in 1976 tional consensus standards (OSH Act, section estimated that the OSHA machine-guarding 6(a)). This authority was limited to the first 2 standards (which had been adopted in 1971) cov- years after the act went into effect (April 1971 to ered only 15 percent of the types of machines in April 1973). use (276). As of 1984, that standard had not yet been revised or expanded. In many cases the orga- The established Federal standards had been nizations that wrote the standards adopted by adopted by the Department of Labor using pro- OSHA in 1971 have revised their standards. cedures that included publication of the proposed OSHA, however, cannot incorporate these changes requirements and provided the public with an op- without going through rulemaking. portunity to comment. The national consensus standards, issued by groups such as the Ameri- Moreover, many of the startup standards speci- can National Standards Institute (ANSI) and the fied in great detail the kinds of equipment neces- National Fire Protection Association, were defined sary for compliance. As pointed out by the 1976 by the OSH Act as standards created in ways that task force, this can hinder the development of im- allowed the consideration of diverse points of proved control techniques and lead to unneces- view and the formation of a substantial consen- sarily costly expenditures for compliance. This sus or agreement by interested persons (Section task force recommended, and OSHA has ac-

3(9)). During hearings before the passage of the cepted, the goal of developing performance stand- OSH Act, industry representatives had argued ards that require employers to control workplace that the consensus standards were already widely hazards without specifying the details of equip- accepted by industry (300). ment design. (The task force also recommended that OSHA publish nonmandatory appendixes to Congress concluded that because these estab- provide information on designs that OSHA con- lished Federal standards and national consensus siders acceptable to meet its performance standards. ) standards were already in effect or represented a voluntary consensus, there was no need for fur- The startup standards were almost exclusively ther rulemaking. Rather, the existing standards safety standards. Most of the relatively few start- could be used to provide a minimum level of pro- up health standards had originated in Federal tection until OSHA could issue its own standards. standards adopted under the Walsh-Healey Act prior to the creation of OSHA. These included Many believe that the new agency, however, the Threshold Limit Values for nearly 400 sub- acted more quickly than required, by completing stances that had been set by the American Con- work on these standards in only 2 months. Un- ference of Governmental Industrial Hygienists fortunately, a number of the national consensus (ACGIH) in 1968. In addition, OSHA adopted standards adopted by OSHA were outdated, un- about 20 ANSI consensus standards that set ex- necessarily specific, or unrelated to occupational posure levels for toxic substances. After eliminat- health and safety. These included a regulation that ing the overlap between these two lists, the start- prohibited the use of ice in drinking water and up permissible exposure limits covered nearly 400 toxic substances. Since 1971, the agency has 1“Rulemaking” refers to the procedures that Federal regulatory adopted new or revised standards for 23 sub- agencies use for adopting regulations. stances and one physical agent (discussed in the ● Ch. 12–Governmental Activities Concerning Worker Health and Safety 227 next section). Some of these were revisions of struction industry, OSHA has adopted a regula- OSHA’s startup standards. Thus the total num- tion requiring consultation with a standing Con- ber of OSHA-regulated substances stands today struction Safety Advisory Committee. Except for at about 410. construction standards, OSHA has not used advisory committees to assist in developing standards The latest Registry of the Toxic Effects of since the mid-1970s. Chemical Substances (586) compiled by NIOSH (as discussed later in this chapter) lists nearly In some cases, OSHA publishes an “Advance 60,000 separate chemical substances. Not all the Notice of Proposed Rulemaking” in the Federal 60,000 are found in the workplace and not all are Register to solicit information from the public. toxic, but there is uncertainty concerning how This step, however, is not required. many hazardous substances are missing from the The first mandatory step is to publish a “Notice OSHA list of regulated chemicals. Most of the ex- of Proposed Rulemaking” that describes the pro- posure limits adopted by OSHA have not been posed new rule, modification to existing rules, or revised, and many are now outdated due to in- revocation of existing rule, and that gives inter- creased knowledge about the hazards posed by ested persons and organizations time in which to particular substances. Furthermore, all the start- comment. An informal, administrative hearing up standards set only a permissible exposure limit. will often take place, which any interested per- None includes additional requirements for expo- son or organization can attend to present testi- sure monitoring, medical surveillance, employee mony and to cross-examine other witnesses. The training and education, record keeping, or warn- agency may later receive written, posthearing ning labels and signs. (In the mid-1970s, OSHA and comments. After the final decisions have been NIOSH started actions to add these requirements made, the agency publishes the text of the stand- in what was called the “standards completion ard and a statement of reasons in the Federal project.” No regulatory actions were ever com- Register. pleted under that project and it is now “dormant.” The only result of that effort is a series of NIOSH A standard typically has staggered startup dates publications with recommendations concerning and deadlines. Some provisions go into effect use, monitoring, surveillance, and protective shortly after publication, while others are delayed equipment for workers exposed to these sub- to allow employers time to plan for compliance. stances. ) Standards that involve the installation of engineering controls generally allow employers a year or more to complete installation. For example, the Standards Issued After Rulemaking OSHA lead standard allows some industries up OSHA also has the authority to issue new to 10 years to comply with the requirements for standards and to modify or revoke existing ones engineering controls. through informal rulemaking. This is authorized The agency is also required to develop an eco- by section 6(b) of the OSH Act, which provides nomic analysis of the expected effects of the stand- for a multistep process. This may start with the ard. This analysis and the content of the proposed receipt of a criteria document from NIOSH, with and final regulations are now subject to review reports from employers, labor unions, or aca- by the Office of Management and Budget prior demics concerning a hazard, or with a petition to publication by OSHA. In addition, after final for a standard from an interested group. publication by OSHA, a major standard is almost ONHA may convene an ad hoc advisory com- invariably the subject of review by one of the U.S. mittee for recommendations. If appointed, such Courts of Appeals after an interested party chal- a committee must have an equal number of rep- lenges its validity. One major issue in the legal resentatives from labor and management, as well challenges to 6(b) rulemaking actions has con- as at least one representative from State health cerned the interpretation of “feasibility” under the and safety agencies. In addition, other persons OSH Act, including the extent to which the agen- with professional expertise may be appointed to cy could or must consider employer costs when the committee. For standards affecting the con- setting standards (see ch. 14). 228 ● Preventing Illness and Injury in the Workplace

Emergency Temporary Standards average time required from first announcement of proposed rulemaking until final publication OSHA is authorized by section 6(c) of the OSH amounts to a little more than 2 years, although Act to issue emergency temporary standards the time for particular standards varied from 6 (ETS) that require employers to take immediate months to 6 years. (This does not include addi- steps to reduce a workplace hazard. As outlined tional time for resolution of legal challenges after by the OSH Act, an ETS can be issued after publication of a final rule, nor does it consider OSHA determines that employees are exposed to the standards that OSHA has begun to develop a “grave danger” and that an emergency standard but has not issued in final form.) is “necessary to protect employees from such danger.” An ETS, issued without providing an op- Ten of OSHA’s final actions on health stand- portunity for comments or for a public hearing, ards established new Permissible Exposure Limits goes into effect immediately upon publication in (PELs) and other requirements for monitoring and the Federal Register. The ETS also initiates the medical surveillance (asbestos, vinyl chloride, process of setting a standard under section 6(b), coke oven emissions, benzene, DBCP, arsenic, with the published ETS generally serving as the cotton dust, acrylonitrile, lead, and ethylene ox- proposed standard. The act mandates that a final ide). Two others did not institute or change a PEL: standard be issued within 6 months of publica- the“14 carcinogens” standard created new re- tion of the emergency standard. quirements for work practices and medical surveillance for a group of carcinogens, while the Major OSHA Rulemaking Actions hearing conservation amendment modified an existing standard with requirements concerning Table 12-3 lists the major health standards noise monitoring, audiometric testing, hearing issued by OSHA since 1971. Through 1984, protection, employee training, and record keeping. OSHA issued 18 separate health standards after rulemaking, or about 3 rules every 2 years. The Thus 12 separate proceedings resulted in new or revised requirements concerning 24 specific substances and 1 physical agent. However, the Table 12-3.—OSHA Health Standards hearing conservation amendment, the 1978 standard for benzene, and the requirements for one of Final the 14 carcinogens have been ruled invalid by the OSHA regulation standard a b courts. (As described in table A-1 in app. A, some 1. Asbestos ...... , ... , , ... , ., 6/07/72 of these standards are still under judicial review 2. Fourteen carcinogens ...... 1/29/74 3. Vinyl chloridea ...... 10/04/74 and several are being reconsidered by OSHA. The 4. Coke oven emissions ...... 10/22/76 application of some requirements in several in- 5. Benzenea...... , ...... 2/10/78 6, DBCPac ...... 3/1 7178 dustries has also been delayed by OSHA or the 7. Inorganic arsenic. , ...... 5/05178 courts. ) 8. Cotton dust/cotton gins ...... 6/23/78 9. Acrylonitrilea ...... 10/03/78 Three regulatory proceedings established new 10. Lead...... , ...... , . . . 11/14/78 “generic” requirements. The access to records 11. Cancer policy. , ...... 1/22/80 12. Access to employee exposure and medical standard created requirements concerning the records ...... 5/23/80 keeping of exposure and medical records and for 13. Occupational noise exposure/hearing providing employee access to those records, while conservation...... 1/16/81 the hazard communication standard requires that 14. Lead—reconsideration of respirator fit- testing requirements ...... 11/12/82 hazardous substances be labeled and information 15. Coal tar pitch volatiles—modification of provided to employees about the substances and interpretation , ...... 1/21/83 16. Hearing conservation-reconsideration . . . 3108/83 the precautions to be taken, The cancer policy set 17. Hazard communication (labeling) ...... 1/25/83 a general OSHA policy concerning future stand- 18. Ethylene oxide ...... 6122184 ards regulating carcinogens. Finally, three pro- %bject of an Emergency TemporaW Standard, bEmeroencv atandarda were issued for aabeatos h 1971 and 1~ ceedings reconsidered and then modified existing cl ,adi~romo.~hloropropane, requirements (respirator fit-testing for lead ex- NOTE: Additional details on these standards can be found in table A-1 of appendix A. posure, coal tar pitch volatiles, and hearing con- SOURCE: Off Ice of Technology Asaesament. servation). . .

Ch. 12–Governmental Activities Concerning Worker Health and Safety 229

Table 12-4 lists the safety standards issued by group of 21 pesticides with an emergency stand- OSHA after rulemaking. In OSHA’s first 13 years ard. (It should be noted that asbestos has been there were 26 such regulations. Many of these the subject of an ETS twice-in 1971 and in 1983. ) safety standards have involved rewriting regula- Of the 11 completed proceedings on specific sub- tions adopted as startup standards, while two stances, six began with the issuance of an ETS. proceedings have revoked a number of specific But because an ETS requires employers to take provisions and advisory language that OSHA in- action before giving them an opportunity to file herited from the consensus standards adopted in comments or objections, these emergency actions 1971. In general, these changes in safety stand- have often been controversial and several of them ards did not impose large increases in costs to have been ruled invalid by reviewing courts. employers. Most of them can be grouped by sub- OSHA’s successes with emergency standards have ject: electrical, mechanical, fire protection, con- been those cases in which labor, management, and struction, or maritime safety. professionals all agreed that a problem existed and that swift action was appropriate. But in those Tables 12-3 and 12-4 also indicate the hazards cases challenged by employers, the courts have and substances for which OSHA has published been reluctant to allow OSHA to impose an ETS. emergency temporary standards. In all, there have been seven separate emergency actions concern- Box N outlines OSHA’s regulation of vinyl ing 19 specific substances, one ETS concerning an chloride monomer. See Ashford (30), Kelman activity (diving), and an attempt to regulate a (245), and McCaffrey (290) for short histories of

Table 12-4.--OSHA Safety Standards

Final OSHA regulation Standard 1. Miscellaneous amendments for construction ...... 2/17/72 2, Cranes/derricks (load indicators) ...... 7/14/72 3. Roll-over protective structures (construction) ...... 4/05/72 4. Miscellaneous amendments for construction ...... 11/16/72 5. Power transmission and distribution ...... 11/23/72 6. Scaffolding, pump jack scaffolding, and roof catch platforms ...... 12/02/72 7. Lavatories for industrial employment ...... 5/03/73 8. Trucks, cranes, derricks, and indoor general storage ...... 6/01/73 9. Temporary flooring—skeleton steel construction ...... 7/02/74 10. Mechanical power presses–(”no hands in dies”) ...... 12/03/74 11. Telecommunications...... 3/26/75 12. Roll-over protective structures for agricultural tractors ...... 4/25/75 13. Industrial slings...... 6/27/75 14. Guarding of farm field equipment, farmstead equipment, and cotton gins . . . . 3/09/76 15. Ground-fault protection ...... 12/21/76 16. Commercial diving operations...... 7/21/77 17. Standards revocation ...... 10/24/78 18. Servicing multi-piece rim wheels ...... 1/29/80 19. Fire protection ...... 9/12/80 20. Guarding of low-pitched roof perimeters during the performance of built-up roofing work...... 11/14/80 21. Design safety standards for electrical standards ...... 1/16/81 22. Latch-open devices (on gasoline pumps) ...... 9/07/82 23, Diving exemptions ...... 11/26/82 24. Marine terminals ...... 7/05/83 25. Servicing of single-piece and multi-piece rim wheels ...... 2/03/84 26. Revocation of advisory “should” and repetitive standards ...... 2/10/84 asubject of an Emergency Temporary standard NOTE: Addltlonal details on these standards can be found in table A-2 In appendix A. SOURCE” Off Ice of Technology Assessment. 230 ● Preventing Illness and Injury in the Workplace .-

Ch. 12–Governmental Activities Concerning Worker Health and Safety ● 231

. .

1, 232 ● Preventing Illness and Injury in the Workplace

some of OSHA’s standards. Mintz (307) provides spections for safety hazards and those for health a number of excerpts of the formal documents hazards (table 12-5). The percentage of health in- concerning many of these standards. spections increased from about 6 percent in the early 1970s to a peak of about 19 percent in fiscal Enforcement year 1979. This has fallen slightly to a range of 17 to 18 percent in the last three fiscal years. (The Inspections and enforcement are the heart of decline would be larger, to about 15 percent, if the regulatory scheme in the OSH Act. Congress “records review” safety inspections were included created an agency that is predominantly an en- in the totals on table 12-5. ) forcement agency conducting unannounced inspections and levying penalties for the “first in- OSHA also classifies its inspections by priority. stances” of violations, as well as for repeated It attempts to investigate first those hazards pos- violations. ing the greatest threat to employee health and safety. The order of priority is: The goals of enforcement are both to correct identified hazardous conditions in inspected plants ● Imminent danger and to provide an incentive for uninspected plants ● Catastrophe and fatality investigations to eliminate or reduce hazards. This second goal ● Employee complaints has often been called “voluntary compliance,” al- ● Special inspection programs though it is misleading to label as “voluntary” ac- ● Programed inspections. tions taken by employers in the face of mandatory Imminent danger inspections are conducted standards with the potential threat of inspection when OSHA learns of a hazard that can be ex- and civil penalties. “Reinspection compliance” is pected to cause death or serious physical harm perhaps a better term (408). In practice, the in- before it could be eliminated through normal en- centive for preinspection compliance is actually forcement activity. Catastrophe and fatality in- quite small, because of both the low probability vestigations, second on the list, are spurred by of inspection and the low level of penalties (see reports of fatal occupational injuries or of in- discussion later in this section). cidents that result in the hospitalization of five or more employees. Inspection Types and Priorities The third priority is employee complaints. OSHA conducts a number of different kinds of Under section 8(f) of the Act, employees and their inspections. The first basic division is between in- representatives who believe that an employer is

Table 12-5.--Federal OSHA Safety and Health Inspections

Establishment Safety Safety Health Health Federal OSHA Inspections Inspections Inspections Inspections Inspections fiscal year (number) (number) (percent) (number) (percent) 1973 ...... 48,409 45,225 93.4 3,184 6.6 1974 ...... 77,142 73,189 94.9 3,953 5.1 1975 ...... 80,978 75,459 93,2 5,519 6.8 1976 ...... 90,482 82,885 91.6 7,597 8.4 1977 ...... 60,004 50,892 84,8 9,112 15.2 1978 ...... 57,278 46,621 81.4 10,657 18.6 1979 ...... 57,734 46,657 80.8 11,077 19.2 1980 ...... 63,404 51,565 81.3 11,839 18.7 1981 ...... 56,994 46,236 81.1 10,758 18.9 1982 ...... 52,8188 43,609 82.6 9,209 17.4 1983 ...... 58,516b 48,269 82.5 10,247 17.5 1984 (first six months) . . . . 30,606’ 25,086 82.0 5,520 18.0 Total...... 734,365 635,693 86.6 98,672 13.4 aDogg not Include 8,444 “Records Review” lnsfMCtiOns. b~g not Include 10,402 “Records Review” inspections. c~~ not Include 4,9S3 “Records Review” iIWeCtiOnS. SOURCE Office of Technology Assessment, based on data supplied by OSHA. Ch. 12–Governmental Activities Concerning Worker Healfh and Safety . 233 violating a health and safety standard may request up inspections has declined, while the percentage an inspection. OSHA schedules inspections to re- devoted to programed inspections has increased spond to what it determines are valid complaints. (see table A-5 in app. A). The fourth priority-special inspection programs— includes programs to give special attention to cer- Enforcement Procedures tain designated hazards and industries. Over the The conduct of all these inspections follows the last decade, the agency has announced several same general outline. The inspector (formally, the such programs: the Target Industries Program, the Target Health Hazard Program, and the Na- Compliance Safety and Health Officer) arrives at tional Emphasis Program. Currently, OSHA is the workplace, almost always without advance warning, presents his or her credentials, and giving special attention to the construction indus- speaks with the employer, manager, or other per- try, oil-well drilling, and grain elevators. son in charge. An opening conference is held with The lowest priority inspections, but by far the the employer (and a union representative, if any) most frequent, are programed ones. (In the past, to describe the purpose of the visit. The inspec- these have been termed “general schedule” inspector asks to see any employer-maintained records tions. ) Although sometimes called “random in- that might be relevant (logs of injuries and ill- spections, ” they now focus on industries with high nesses, exposure and medical surveillance records, injury rates or those with known health hazards. etc. ). Over the past 12 years, OSHA has used a series of different scheduling systems. For safety inspec- Then the inspector conducts a “walk-around,” visiting all or part of the workplace. Both the em- tions, industries are ranked using injury rate in- ployer and an employee representative have the formation from the Bureau of Labor Statistics right to accompany the inspector on this tour. The (BLS) Annual Survey. Only industries with injury inspector observes workplace conditions and takes rates above the national average for the private notes, photographs, and exposure samples as may sector are now selected for safety inspections. For be appropriate. After the walk-around, a closing health inspections, OSHA now selects industries conference is held (either jointly with both em- based on the last 5 years of OSHA health inspec- ployer and employee representatives, or sepa- tions. Industries are ranked based on their respec- rately), at which time apparent violations of tive violation rates. Individual establishments are selected by using commercially available lists of OSHA standards are discussed. Normally, citations are not issued at this closing conference, but employers to identify the establishments with 10 are mailed later. This is to allow consultation be- or more employees in these selected industries tween the inspector and the area director and to (635). obtain the results of laboratory analysis of ex- One other inspection category should be noted. posure samples. Follow-up inspections can be conducted at any The Fourth Amendment to the U.S. Constitu- time to determine if workplace conditions have tion generally protects persons from “unreason- changed following an inspection. In particular, able searches and seizures” by government offi- OSHA is interested in verifying that abatement cials. There are certain circumstances in which of a hazard has taken place by a scheduled date. (For information on the types of inspections, see police are allowed to search without consent, but in general a warrant must be obtained for such 307,333,408,635. ) a search. The question of whether OSHA is also OSHA has, through policy changes, varied the subject to these requirements in conducting its distribution of inspection activity among these inspections of workplaces was raised most prom- categories. For example, the proportion of inspec- inently in a 1976 case concerning an Idaho em- tions triggered by employee complaints increased ployer, Barlow’s Incorporated. The case was ul- from about 10 percent in fiscal year 1976 to over timately appealed to the Supreme Court, which 30 percent in fiscal year 1977, In recent years the ruled that OSHA had the constitutional author- percentage of complaint inspections and follow- ity to conduct inspections, but that if the employer 2.34 ● Preventing Illness and Injury in the Workplace

did not voluntarily consent to the inspection, the age injury rates. There is concern, however, that agency would need to obtain a court-issued war- this policy may also serve as an incentive for sys- rant (see 307,333,408). tematic underrecording of the number of injuries, although inspectors are instructed to verify the Although Congress has never amended the accuracy of employer injury records. OSH Act, it has limited OSHA inspection activity by attaching restrictions to OSHA’s appropria- Compliance with Standards and the tions. Currently OSHA is prohibited from con- General Duty Clause ducting programed safety inspections in establishments with 10 or fewer employees in an Unless they have obtained a variance, employ- industry with a led-workday rate lower than the ers are required by section 5(a)(2) of the OSHA national average; from assessing penalties for the Act (the specific duty clause) to comply with the “first instances” of nonserious violations (unless terms of OSHA standards and regulations. Dur- 10 or more total violations are cited); from assessing inspections, inspectors look for any violations, ing penalties against an employer with 10 or fewer following the procedures and interpretations employees who had requested an onsite consulta- issued by OSHA in its Field Operations Manual, tion and is acting to eliminate identified hazards; Industrial Hygiene Manual, and program from issuing or enforcing standards for small directives. farming operations (10 or fewer employees); and Employers must also comply with section S(a) from activities that would affect recreational hunt- (1) of the act–the general duty clause. This has ing, shooting, or fishing. There are also restric- been used to cover hazards not treated by OSHA’s tions on visits designed to monitor State programs more specific standards. There has been contro- and on OSHA activities on the Outer Continen- versy concerning the interpretation of this clause tal Shelf (489a). and its application to employers by OSHA. Current Inspection Targeting To prove a violation of the general duty clause, OSHA must demonstrate that the employer failed A new procedure for safety inspections was in- to render the workplace free of a “recognized” haz- stituted on October 1, 1981. Since then, OSHA ard that was causing or was likely to cause death inspectors conducting programed safety inspec- or serious physical harm. OSHA and the courts tions have also examined the injury and employ- have held that a hazard is recognized if the em- ment records of the employer and calculated an ployer had knowledge of the hazard or if it is of average lost-workday case rate for the previous common knowledge in the industry in question 2 years (or 3 years, for establishments with 10 to and detectable by the senses or by techniques gen- 20 employees). If the lost-workday case rate is less erally known and accepted by the industry (307, than the national average rate for manufacturing, 333,408). OSHA uses a number of different sources the inspection is terminated. The national aver- of information to demonstrate that a hazard is age injury rate is derived from the BLS Annual “recognized,” including voluntary standards, Survey and currently is 4.2 cases per 100 full-time statements of industry experts, implementation of employees (644). (In table 12-5, these visits are abatement programs by other companies in the termed “records review. ” Termination of inspec- same industry, manufacturers’ warnings, or tion because of a below-average injury record studies conducted by the industry, its employees, does not currently apply to safety inspections in the government, or insurance companies (203). the construction industry, nor does it apply to health inspections in any industry.) The Role of OSHRC In theory, this policy of ending inspections for “low-hazard” establishments can create an incen- An employer who disagrees with OSHA con- tive for employers to take steps to reduce injury cerning a citation, a proposed penalty, or the date rates. It also may help OSHA to use its scarce infor abatement of the hazard can file a notice of spection resources efficiently by concentrating at- contest. Employees also have an independent right tention on establishments with higher-than-aver- to contest the reasonableness of the length of the --—

Ch. 12–Governmental Activities Concerning Worker Health and Safety ● 235 proposed period of time for abatement of a haz- Recent court decisions limit the rights of em- ard. Unless contested within 15 days, an OSHA ployees and unions to object to settlements that citation becomes final. When contested, a hear- they view as insufficiently protective. If the em- ing is held before an Administrative Law Judge ployer is no longer contesting the citation, the (ALJ), who is an employee of the OSHRC, an in- rights of employees and unions to object is limited dependent review body. to the “reasonableness of the abatement period. ” At the hearing before the ALJ, both OSHA and (In addition, OSHA enforces the provisions of the employer present their sides of the case. Af- the OSH Act that prohibit employers from dis- fected employees and their representatives have criminating against employees who exercise any a right to participate in these proceedings. Dur- of the rights provided by the OSH Act. See ch. ing the period when the citation is under contest, 15 for a discussion of this protection and other no abatement is required. employee rights. ) The ALJ examines the evidence and decides Incentives for OSHA Compliance whether to affirm, vacate, or modify OSHA’s citation and proposed penalties. After this decision, In theory, OSHA’s enforcement activities will any party can petition the Occupational Safety lead employers to comply with OSHA regulations and Health Review Commission to review the and to improve employee health and safety. But decision of the ALJ. The Commission can grant in practice OSHA has never had the resources to such review either upon request or by its own inspect more than a relative handful of the Na- choice. Unless it is ordered to be reviewed within tion’s workplaces. In addition, the average 30 days by a member of OSHRC, the ALJ deci- penalties levied for violations uncovered in the sion becomes a final order of the Commission. “initial” visits to workplaces are very small, espe- Any adversely affected person can then petition cially when compared with the costs of many a U.S. Court of Appeals for judicial review types of controls (see box T in ch. 15). Thus, there (307,333,408). is only a weak incentive for employers to comply prior to an OSHA inspection. Informal Conferences In fiscal year 1983, OSHA conducted about 58,500 establishment inspections and about 10,400 Current OSHA policy encourages the use of in- “records review” visits, The various State pro- formal conferences between employers and OSHA’s grams conducted an additional 104,000 establish- area directors to reach settlements concerning cita- ment inspections and about 2,500 “records re- tions. Using such conferences to settle citations view” visits. This makes for a Federal-State total might facilitate prompt abatement of hazards and of about 162,000 establishment inspections and improve employee health and safety, as well as 13,000 “records review” visits. As there are about reduce the time spent by government personnel, 4.6 million private sector establishments in the employers, employees, and their representatives United States (553), OSHA and the State pro- resolving contested citations. The first directive grams can inspect less than 4 percent of U.S. on this was issued by Eula Bingham in 1980. establishments. Assistant Secretary Theme Auchter strongly The probability of a health inspection is sub- encouraged the use of these settlements by stantially less than the probability of a safety in- OSHA’s Area Directors. During his tenure as head spection. Of all fiscal 1983 inspections, about of OSHA, the number of informal settlements in- 21,000 were health inspections. Thus, of each creased dramatically. But there have been sugges- 1,000 establishments, about 31 receive a safety in- tions that some of these settlements may not suf- spection, 3 receive a “records review” visit (which ficiently protect worker health and safety (440). covers only safety), and fewer than 5 receive a Moreover, if the settlement agreement provides health inspection. for eliminating penalties, or reducing them to very low levels, the incentive for preinspection compli- OSHA and the States do, however, concentrate ance is virtually eliminated. their activities on certain industries. For example, 236 ● Preventing Illness and Injury in the Workplace

in fiscal year 1983 the agency conducted about As discussed above, for establishments in man- 58 percent of its establishment inspections in con- ufacturing, the probability of inspection is only struction and about 33 percent in manufacturing. about one in six, while the average proposed pen- This would imply about 54,000 establishment in- alty for “serious” violations was about $172. A spections in the approximately 321,000 manufac- decision based purely on a desire to maximize prof- turing establishments nationwide (assuming that its would be to spend on controls only when the the State plans conduct the same proportion of annual cost of controls is less than the annual ex- inspections in manufacturing as does OSHA). pected costs of noncompliance. If it is assumed Therefore, OSHA and the State programs are cur- that the inspection will detect the violation, the rently able to inspect a maximum of about 17 per- expected costs for each serious violation equals cent of manufacturing establishments (assuming one-sixth of $172 or about $29. The average pro- one inspection per establishment in any given posed penalty for all violations was $39. With a year). one in six probability of inspection, this implies an expected penalty for noncompliance of about At this rate of inspection, OSHA and the State $6.50. (See 64,127,287, and 685 for similar esti- programs can inspect each establishment in man- mates based on older data.) ufacturing only once every six years. The inspection rate for most other industries with fixed Of course, employers may take actions out of establishments—transportation, wholesale and altruistic concern about the health and safety of retail trade, finance, services, and agriculture— their workers or because of the other incentives is substantially less than this. (Because of the limits described in chapter 15. In addition, for employers of available statistics, it is difficult to estimate the who have been inspected recently, there is the probability of inspections at construction sites.) threat of a repeat inspection. In these cases, the fines for “failure to abate” an identified hazard Even if an establishment is inspected, the penal- can be very substantial. Thus the incentives for ties for violations are quite low. The average pro- compliance are much larger after an OSHA in- posed penalty for a “serious” violation (one that spection. What is clear, however, is that the eco- creates a “substantial probability [of] death or seri- nomic incentive for preinspection compliance is ous physical harm”) for OSHA and the State pro- small. grams in fiscal year 1983 was about $172. Of course, this is only for violations that are actually In fiscal year 1983, compared with fiscal year discovered and cited, and does not include the 1980, OSHA issued 40 percent fewer serious viola- penalties levied for “other than serious,” “willful,” tions, 55 percent fewer repeat violations, and 85 “failure to abate,” and “repeat” violations. In fiscal percent fewer willful violations. Consequently, year 1983, OSHA and the State programs pro- the total penalties it has proposed declined by posed penalties totaling about $13.4 million for nearly 60 percent (see tables A-8 and A-10 in app. all violations in the 162,000 establishment inspec- A). This change has also been accompanied by tions that were conducted, an average penalty of a decline in the number of inspections with con- about $83 per inspection. tested citations (from nearly 12 percent of inspections to less than 2’ percent) and may lead to As discussed in chapter 14, employers making prompt abatement of hazards in firms that have decisions to maximize profits will tend to choose been inspected. However, the reduction in penal- the least costly alternative. Thus the employer ties also significantly weakens the small economic faces a choice of spending a certain sum to con- incentive for employers to comply with OSHA trol a hazard and comply with OSHA standards, standards. or not controlling a hazard and possibly being inspected and penalized for noncompliance. The At least one commentator has suggested that decision will be based on a number of factors, in- “OSHA will not be a meaningful deterrent until cluding the probability of inspection and the likely the cost of noncompliance becomes greater than penalties if an inspection takes place and the haz- the cost of compliance” (406). Violations of some ard is detected. requirements for pollution control have been —. —

Ch. 12–Governmental Activities Concerning Worker Health and Safety 237 penalized with fines that are equivalent to the fi- representing employers and employees as to ef- nancial benefits that a firm receives from noncom- fective means of preventing occupational injuries pliance. This began in 1973 in Connecticut, and and illnesses. ” at the Federal level with the 1977 Clean Air Act Amendments. Recently, the Environmental Pro- Consultation tection Agency announced plans to extend this policy to other pollution control requirements, Consultation is the provision of information, along with other changes in its policies concern- measurements, and advice about controls with- ing civil penalties (163). out threat of citation or penalty. Section 9(a) requires that when OSHA personnel discover viola- An Interagency Task Force on Workplace Safe- tions of standards, the agency “shall with ty and Health recommended that OSHA develop reasonable promptness issue a citation to the em- procedures to set its penalties at levels equal to ployer” (emphasis added). This absolute language the benefits of noncompliance (228). Such a has been interpreted to prohibit the agency from change, although clearly providing employers a engaging in employer consultation at the worksite. larger incentive to comply with OSHA standards, Rather, OSHA personnel must issue citations if would probably increase the level of controversy violations are found. This language originated in in OSHA enforcement proceedings. response to the history of frequently ineffective State health and safety agencies, which prior to Public Education and Service 1970 often gave inspectors the discretion of issuing or not issuing citations and penalties. The third major category of OSHA activities can be called public education and service. This OSHA, therefore, has limited its direct consul- includes OSHA-funded consultation, the projects tation to phone calls, letters, office visits, and assisted by grants under the New Directions pro- speeches at employer gatherings. Partly from con- gram, and OSHA’s “voluntary protection pro- gressional interest, expressed through proposed grams.” These activities, although different in legislation and the appropriations process, and many ways, are all designed to improve work- partly from OSHA itself, actions have been taken place health and safety through methods other to provide for onsite consultation for employers than direct standard-setting and enforcement. desiring this service. Because of the section 9 language, Federal OSHA personnel do not them- The OSH Act devotes only a few lines to edu- selves conduct these visits. Rather, through cation and service activities, in contrast to the various contractual and grant mechanisms under many paragraphs establishing the new agencies sections 18(b) and 7(c)(1), OSHA pays for visits and specifying the authority of these agencies for by State personnel (usually in departments of la- standard-setting, enforcement, and research. Sec- bor, industry, or health) or private consultants. tion 2(b)(2) declares that one purpose of the Act is to The purpose of the onsite consultation program . . . encourage] employers and employees in is to provide employers with a confidential eval- their efforts to reduce the number of occupational uation of the health and safety hazards in their safety and health hazards at their places of em- workplaces, and to recommend means of hazard ployment, and to stimulate employers and em- abatement. An onsite consultation will let an employees to institute new and to perfect existing ployer know how the business measures up to the programs for providing safe and healthful work- relevant OSHA standards. This service is pro- ing conditions; . . . vided at no cost to the employer, and priority is Section 21(c) provides specific authorization for given to requests from small businesses. Visits oc- OSHA to conduct “education and training of cur only at the request of the employer and the employers and employees in the recognition, results of the visit are provided only to the em- avoidance and prevention of unsafe or unhealthful ployer. Consultation is not available to employ- working conditions” and to “consult with and ad- ees or their unions, nor do they have a right to vise employers and employees, and organizations the information provided to the employer by the 238 . Preventing Illness and injury in the Workplace consultant, although employers may voluntarily OSHA has also modified its policy concerning turn that information over to employees. This the provision of onsite information by agency in- program has increased substantially from 1975 to spectors. OSHA still issues citations for appar- today. In fiscal year 1983, OSHA funded about ent violations, but inspectors are now required 28,000 onsite visits, at a cost of $23.4 million (see to provide “general assistance” to employers in table 12-6). identifying abatement methods for alleged violations (629). In addition, OSHA has recently an- In 1978, the General Accounting Office (GAO) nounced plans to create “full-service area offices. ” issued a report criticizing OSHA’s management The goal is the creation of “resource centers” that of the consultation program, suggesting that employers can use to obtain information on occu- OSHA’s policies were not sufficient to ensure pro- pational safety and health and on compliance with tection of worker health and safety (505). Changes OSHA standards. in the program have been made since then, but a continuing controversy concerns the relative em- New Directions Program phasis of enforcement versus consultation, which parallels the debate about the relative merits of The New Directions program was established mandatory standards versus voluntary efforts. by OSHA in 1978 to provide grants to employee, Many think that Government should engage in employer, educational, and nonprofit organiza- consultation in order to assist well-meaning but tions for the purpose of providing workplace ignorant employers to improve health and safety. health and safety training, resources, and serv- Others suggest that such an approach is largely ices for employers and employees. Grantees have ineffective and tends to divert resources from en- used these funds to develop educational materials, forcement activities that require, rather than conduct training sessions, provide technical assist- merely encourage, health and safety improvements. ance concerning health and safety hazards, and hire technical staff. OSHA has provided the bulk In 1982, OSHA began an experimental program of the funds for this program, and additional in two regions to grant employers a one-year ex- money has come from the National Cancer Insti- emption from programed inspections if they apply tute, NIOSH, the Federal Emergency Management for and receive a health and safety consultation. Agency, and the National Institute of Mental In 1984 that program was made permanent and Health (636). extended nationally. Now employers who agree to a comprehensive onsite consultation, correct The program has recently been cut back sub- all hazards detected during the consultation, and stantially, from $13.9 million in fiscal year 1981, implement the “core elements of an effective safety to $6.8 million for fiscal years 1982 and 1983. and health program” can be given a one-year ex- (This does not, however, include funds from the emption from programed inspections (355, 648). National Cancer Institute that are also included However, given the low probability of an inspec- in this program. ) Similarly, the number of orga- tion, this one-year exemption may not represent nizations receiving grants has declined from 156 a very large incentive to participate in this in fiscal year 1980 and 142 in fiscal year 1981 to program. 100 in fiscal year 1983 (34). Ch. 12—Governmenta/lActivities Concerning Worker Health and Safety 239

Voluntary Protection Programs other Federal agencies who need health and safety training. OSHA also makes available a certain In 1982, OSHA created three programs that number of classroom spaces for private sector per- both recognize the achievements of companies sonnel. In fiscal year 1983, approximately 6,700 that are “leaders” in providing health and safety trainees completed a course at the OSHA insti- to their employees and provide additional oppor- tute—nearly 1,850 from Federal OSHA, about tunities for OSHA-employer consultation and co- 1,650 from State agencies, 2,800 from other Fed- operation. Collectively referred to as “voluntary eral agencies, and about 500 from the private sec- protection programs, ” they are individually tor (34,630,636). named: “Star, “ “Try,” and “Praise.” They are a mixture of consultation and recognition for superior performance. In the course of develop- State Programs and Other ing a program, OSHA personnel may make an Federal Agency Activities onsite visit and offer suggestions for program im- State Programs provement. So far, only a handful of companies have applied to participate in these programs. Par- Many who supported passage of the Occupa- ticipating companies are exempted from OSHA tional Safety and Health Act in 1970 believed that programed inspections and are promised ex- a Federal program was necessary because existing pedited action on variance applications. State programs were inadequate. But during congressional consideration, a coalition of business These programs represent a more cooperative leaders, State officials, a number of State-oriented approach toward employers than OSHA has members of Congress, and the Nixon administra- taken in the past and are intended to enhance tion pushed for the inclusion of State-inn OSHA worker health and safety. Critics of voluntary programs in the act. It has been suggested that protection, on the other hand, are concerned the act could not have passed without a provi- about the general policy of exempting participat- sion allowing State plans (30,460). ing companies from scheduled inspections and fear that these programs divert OSHA resources In passing the OSH Act, Congress established away from inspections and standard-setting activity, a mechanism that enables States to regulate worker health and safety subject to Federal mon- Other Programs itoring and approval. A State program must, in general, “provide for the development and en- OSHA has also, from time to time, set up co- forcement of safety and health standards which operative programs with trade associations, . . . are or will be at least as effective” as Fed- unions, etc. to disseminate health and safety in- eral standards (OSH Act, Section 18). formation. In 1982, for example, OSHA agreed to assist the American Electronics Association The development of a State program is a step- with its efforts to conduct training courses and wise process. After application and initial ap- prepare booklets on health and safety. Another proval from OSHA, the State can begin to en- agreement was signed with the National Agricul- force health and safety standards. States can adopt tural Chemicals Association to develop a slide/ the existing Federal standards, as have 18 of the tape program on good workplace practices for 24 plans approved before 1984. Half of these 18 pesticide manufacture and formulation (632). also enforce some State standards. The other 6 States enforce mostly their own standards, which OSHA also has an in-house training institute OSHA has deemed to be “at least as effective” as in Des Plaines, IL. It offers 52 health and safety the corresponding Federal ones. courses of 1 to 2 weeks each at no cost to the students. Since its start in 1972, approximately The first 3 years of a State plan are called the 40,000 trainees have attended the institute. The developmental stage. At the time of initial plan training is generally designed to meet the ad- approval, both OSHA and the State agency have vanced training needs of OSHA Compliance Of- concurrent jurisdiction —both have the authority ficers, State agency personnel, and staff from to conduct inspections and cite employers, and 240 ● Preventing Illness and Injury in the Workplace employers must therefore comply with both Fed- came the 25th State program, although it applies eral and State standards. (Any time after initial only to State and local government employees approval, however, as soon as the State is “oper- (table 12-7). “Final approval” had not been ational, ” OSHA may suspend its concurrent en- granted to any jurisdiction until 1984, when the forcement jurisdiction through an “operational Virgin Islands, Hawaii, and Alaska received this status agreement.”) After all developmental steps designation. are completed and approved, OSHA can issue a An important issue for OSHA policy concern- certification of the plan. If the State meets all of ing State programs has been a continuing dispute OSHA’s requirements, it becomes eligible for concerning the level of funding and the number “final approval” 1 year after certification. Even of compliance personnel a State must have to meet after “final approval, ” Federal OSHA still moni- the requirements of the OSH Act. In the mid- tors the State program. 1970s OSHA had interpreted this to mean that The requirement that States maintain a pro- State staff levels need only be equivalent to those gram “at least as effective” as OSHA’s means that of OSHA, even though OSHA’s staffing levels if OSHA issues new or revised regulations, State were not considered optimal. The AFL-CIO sued agencies must follow suit by adopting the change, OSHA on this issue. issuing an equivalent change, or making the case In a major decision in 1978 (AFL-CIO v. Mar- that, for local reasons, there is no need to alter shall), the reviewing court agreed with the AFL- the regulation. (For example, a State without tex- CIO. According to its opinion, the “at least as ef- tile mills need not adopt the OSHA cotton dust fective” requirement applies only to the standards standard. ) issued by the State program. For staffing and There are also procedures for OSHA to with- funding, the court ruled that Congress intended draw approval and/or reintroduce Federal en- that OSHA could use “current Federal levels of forcement for a State that does not comply with personnel and funds as benchmarks for State pro- OSHA’s requirements (307,333,408). In all stages grams, provided they are part of a coherent proof its operation, OSHA monitors the quality of gram to realize a fully effective enforcement ef- the State program. Monitoring may involve “spot fort at some point in the foreseeable future.” Thus checks” (inspections by Federal personnel after a while current Federal levels, described by OSHA State-conducted inspection) or “accompanied” itself as only “a percentage of what is really monitoring visits in which Federal personnel needed, ” were deemed to be adequate as interim observe a State inspector during an inspection. benchmarks, they were not adequate as goals for (More recently, a computerized data system has “fully effective” State programs (307). replaced much of this onsite monitoring. )

History of State Programs Under the OSH Act Table 12-7.–State Programs (1984) The policies of OSHA toward State programs have varied-sometimes permitting State pro- Jurisdlctions with approved plans: grams to operate relatively free from oversight, Alaska North Carolina Arizona Oregon while at other times setting high standards for California Puerto Rico State programs. OSHA’s efforts to allow and en- Hawaii South Carolina courage or disallow and discourage State pro- Indiana Tennessee Iowa Utah grams have also been subjects of litigation Kentucky Vermont (249,307). Maryland Virgin Islands Michigan Virginia From 1978 to 1983, 24 State programs were in Minnesota Washington Nevada Wyoming operation—21 States plus Puerto Rico and the New Mexico Virgin Islands covered private and public employ- Plans cover/rig only pub//c employees: ment and one program (Connecticut’s) covered Connecticut New York only public employment. In 1984, New York be- SOURCE: (408). Ch. 12–Governmental Activities Concerning Worker Health and Safety . 241

In 1980, with the concurrence of the AFL-CIO, programs devote insufficient resources to health OSHA submitted to the court staffing-level “bench- and safety and are ineffective. Although many marks” designed to achieve the goal of State people in the business communities of affected programs becoming “fully effective” rather than States support such programs, large businesses only “at least as effective” as Federal operations. with establishments in more than one State often This plan would have required the jurisdictions express a desire for regulatory consistency-some- with approved State programs to increase the total thing that is hard to achieve with a variety of dif- number of safety inspectors from 849 to 1,154 ferent State standards. Recent employer support within 5 years and of health inspectors from 332 for an OSHA “labeling” standard, for example, to 1,683 (subject to an annual reassessment of the derives in part from a desire to preempt State and availability of qualified health personnel). The local laws on this subject. agency also released a “benchmark model” that Local control over job safety and health could could be used to generate or revise staffing levels be desirable in its own right, and the use of State as necessary. The model considered the hazard programs can create a combined Federal and State rank of various industry sectors and the number effort that is larger than the Federal effort alone of establishments of each rank in each State when could be. But most States do not have the research calculating the number of programed inspections, capability needed to set standards, nor would it the factor that ultimately determines State health be efficient to have each conducting the same re- and safety staffing requirements. search for standards development (30). However, Since then, OSHA has taken steps to change because they can simply adopt Federal standards these required benchmark levels. Beginning in verbatim, it is usually not necessary for them to 1981, it has sought congressional approval to pro- have such a capability (34). hibit the spending of any funds to achieve State But the possibility exists that States may com- staffing levels that are greater than current Fed- pete with each other in order to attract new eral levels. Although enacted in two continuing businesses by relaxing the enforcement of health resolutions concerning appropriations in fiscal and safety standards. In theory, this should be years 1982 and 1983, this language was not in- restricted by the requirement that State programs cluded in the final appropriations actions for those be “as effective as” Federal OSHA. Examination years (34). In 1984, it was reported that OSHA of inspection data shows that, on average, State is no longer seeking this appropriations language programs issue fewer serious violations and, for (350). In 1982, OSHA also formally proposed to this reason, propose lower penalties overall than recalculate the benchmark formula. One result of OSHA does. Moreover, the States with programs, this would be the lowering of the required State as a group, devote a smaller fraction of their in- staffing levels (249). spection resources to health inspections than does OSHA (see tables A-l to A-11 in app. A). On the Pros and Cons of State Programs other hand, State programs conduct proportion- A number of arguments have been made for ately more inspections than OSHA does and some and against State takeover of OSHA respon- States have proportionately larger enforcement sibilities. Proponents of State programs argue that staffs, particularly in safety (34). In 1976, the they can adapt to local needs better, are more ef- GAO criticized OSHA’s policies toward the deficient and more fairly enforced, and continue the velopment of State programs and expressed con- traditional State and local roles in occupational cern that these policies were not sufficient to en- health and safety regulation. Opponents argue sure employee health and safety (500). that, compared with OSHA, these programs are The friction over benchmarks, final approval, less effective, understaffed, underfunded, inefficient, less evenly enforced, and more susceptible and Federal monitoring of State programs reflects to local political influence (408). the underlying tension between the national and the State governments, which historically had Organized labor, especially at the national been responsible for job safety and health. Fed- level, is principally concerned that many State eral policies concerning State programs have re- 242 ● Preventing Illness and Injury in the Workplace

sponded to those tensions, as well as to more gen- “working conditions” for which other Federal eral views concerning the rightful role of the agencies and certain State agencies “prescribe or Federal Government. enforce standards or regulations affecting occupational safety or health” (OSH Act, Section 4). For For example, the 1973 President’s Report on Oc- example, the health and safety of coal miners is cupational Safety and Health referred to the ap- regulated by the Mine Safety and Health Admin- proval of 20 State programs in 1973 as a step that istration, while protection of certain railroad em- moves the United States closer to the “Fed- ployees is provided by the Federal Railroad Ad- eral/State occupational safety and health partner- ministration. ship intended by Congress. ” Assistant Secretary Bingham took a very different view, stating that Health and safety conditions for most public “[t]he Federal agency is given a leadership role sector workers are not directly regulated by which does not lend itself to a traditional part- OSHA. State and local employees, however, are nership of equality as I believe a number of States covered by State programs in the 25 States with desire.” Thorne Auchter returned to an earlier pol- approved plans. The heads of Federal agencies are icy of actively encouraging State programs. In his required by the OSH Act to provide their employ- words, “It is my belief—and the belief of this ees with an occupational safety and health pro- administration —that in the last analysis, local gram that is “consistent with” OSHA standards. problems are best addressed by those closest to Three different Executive Orders have been issued them” (249). concerning health and safety protections for Fed- eral Government employees. Lastly, the regulations issued by several other Other Federal Agencies Federal agencies also affect job safety and health, Although OSHA and the State programs are even though workplace conditions are not the pri- directly responsible for ensuring the health and mary focus of these agencies. The constellation safety of most private sector workers in the United of governmental bodies with workplace safety and States, some workers are the responsibility of health responsibilities is summarized in table A- other agencies. The OSH Act does not apply to 12 in appendix A.

THE NATIONAL INSTITUTE FOR OCCUPATIONAL SAFETY AND HEALTH

The need to provide for research to identify, Welfare) to carry out research and related activi- evaluate, and control work-related illness and in- ties. Congress listed several major activities for jury, to disseminate information to workers, em- NIOSH: ployers, and health professionals, and to train occupational safety and health professionals did not ● develop criteria for recommended occupa- provoke much controversy during the debate tional safety and health standards; about passage of the OSH Act. All parties agreed ● conduct educational programs to provide an there was overwhelming evidence that scientific adequate supply of qualified personnel; knowledge about work-related injury and illness ● conduct informational programs on the im- was lacking, that the supply of trained personnel portance of the use of adequate safety and was inadequate, and that meaningful statistical health equipment; data were unavailable. o conduct Health Hazard Evaluations; and ● conduct industry-wide studies of the effects The OSH Act established the National Institute of chronic or low-level exposures. for Occupational Safety and Health within the De- partment of Health and Human Services (then In contrast to OSHA, most NIOSH personnel called the Department of Health, Education, and work in only two locations, at the NIOSH facil-

Ch. 12—Governmental Activities Concerning Worker Health and Safety 243 ities in Cincinatti, OH, and Morgantown, WV. The funding for the three activities is not,equal. As mentioned earlier in this chapter, NIOSH Identification now receives the largest share of headquarters are now in Atlanta, GA. There are NIOSH’s budget, with information dissemination also very small staffs, usually two to three, in the and control technology research receiving sub- 10 HHS regional offices (located in the same cit- stantially less funding. In addition, although its ies as OSHA regional offices). mandate extends to both health and safety, NIOSH has concentrated almost exclusively on NIOSH’s organization reflects the various top- questions of occupational health. ics of its research and dissemination program. These activities are conducted by NIOSH’s seven divisions, each of which conducts research con- Identification cerning a particular aspect of occupational health Identifying work-related illness and injury and and safety or provides for the dissemination of understanding the mechanisms of disease and in- information or training and education (fig. 12-5). jury causation involve using the skills of medi- For this description, OTA divides NIOSH activ- cine, industrial hygiene, safety engineering, epi- ities into three major program areas: demiology, toxicology, and statistics. It is the first step toward prevention (chs. 3 and 4). The spe- ● Identification of occupational health and cific NIOSH activities directed toward identifica- safety problems; tion include illness and injury surveillance sys- ● Development of controls to prevent work- related illnesses and injuries; and hygiene, which refers to recognition, evaluation, and control. OTA ● Dissemination of findings and recommenda- has grouped what NIOSH calls identification and evaluation into tions and provision of Professional Train- one category, largely because there really are no clear boundaries 2 between these two activities, and both depend on a variety of pro- ing and Education. fessional skills. Moreover, the term “evaluation” is commonly used in the Federal Government to refer to the process of assessing the itself divides its activities into five areas: identification, administration and impacts of particular programs. Administration, evaluation, control, dissemination, and administration (s79). The the fifth NIOSH program area, is not a research area and is not dis- first three appear to have derived from the language of industrial cussed in detail in this chapter.

Figure 12-5.—National Institute for Occupational Safety and Health 244 Preventing Illness and Injury in the Workplace terns, Health Hazard Evaluations, and NIOSH’s These surveys represent a large effort. The re- toxicologic and epidemiologic research programs. cent survey, patterned after the first one, covered a random sample of nearly 5,000 worksites in 67 Surveillance major metropolitan areas over a 2-year period starting in 1981. Potential exposure estimates for NIOSH surveillance systems identify workplace the Nation are based on the probability sample hazards and work-related injuries, diseases, in which over 9,000 chemicals and physical agents disabilities, or deaths, and generate hypotheses were observed in industrial use. These observa- concerning the possible relationships between tions included exposures of workers in 450 sepa- workplace exposures and adverse effects. NIOSH rate occupational classifications. conducts work-related illness and injury surveillance in several ways. For illnesses, NIOSH’s The results of these two surveys can provide activities have included two large surveys of national estimates of potential exposure to work- working conditions and the Health Hazard Evalu- place hazards by industry and occupational ation program (discussed in detail below). groups that can be useful for research planning, for estimating the impact of proposed standards, Until relatively recently, NIOSH did very lit- and for planning OSHA’s enforcement activities. tle on the identification of the causes of injuries. For instance, occupational safety and health pro- Now it collaborates with the Bureau of Labor Sta- fessionals have used the 1974 NOHS to identify tistics and OSHA in conducting work injury sur- groups of workers exposed to various hazards or veys, which are special-topic mail surveys of in- combinations of hazards. One finding from NOHS jured workers. Epidemiologic methods for injury was that over 22 million of the 38 million work- investigation are being developed. National workers in the survey universe were exposed to at least related injury data are being collected in conjunc- one potential hazard. tion with the Consumer Product Safety Commis- sion’s National Electronic Injury Surveillance Trade-name chemical products (products with System. a commercial name but often lacking information about the identity or properties of the chemical National occupational hazard and exposure sur- constituents on their containers) are prevalent in veys.—The first NIOSH survey of working con- the workplace. Analysis of the NOHS data showed ditions, called the National Occupational Hazard that potentially toxic substances are frequently Survey (NOHS), was completed in 1974. Pres- present in them. About 70 percent of 86,000 prod- ently, a major task of NIOSH’s surveillance pro- ucts identified in NOHS (made by 10,500 manu- gram is analysis of data collected in the second facturers) had trade names. NIOSH requested national inventory of potentially hazardous ex- information about the composition of these posures to workplace agents, the National Occu- approximately 60,000 separate trade-name prod- pational Exposure Survey (NOES).3 The data for ucts from the 10,500 manufacturers. The manu- both of these surveys were collected through visits facturers supplied information on 45,000 trade- to a random sample of workplaces. Potential name products. Of these, 40 percent contained at exposures to workplace hazards were observed least one OSHA-regulated substance and over 400 and recorded during comprehensive tours of these of these trade-name products contained at least worksites. The surveyors collected information one of the OSHA-regulated carcinogens. on the nature of the potential hazards, the num- However, while NOHS estimates are the only ber of workers at risk, and the controls in place national base of information on exposure to po- to protect them. tential workplace hazards, they have been questioned for their accuracy, for being nonquan- 3The word “hazard” in the title of the NOHS survey was changed to “exposure” in the second survey in part because employers’ rep- titative, and for their validity. NOHS estimates resentatives argued that the survey was really an inventory of all for agents for which there is a high degree of potential exposures, of which hazardous ones were only a part. awareness (such as asbestos) sometimes appear Ch. 12—Governmental Activities Concerning Worker Health and Safety ● 245 at odds with figures put forward by others. Esti- In fiscal year 1981, NIOSH received 513 re- mates of the numbers of people exposed to low- quests for mining and general industry HHEs, a toxicity substances, such as sodium chloride, 10 percent increase over 1980. Twenty-one per- dominate some parts of the data base, leaving a cent of these requests came from employers, 55 casual observer with the impression that the data percent from employees or unions, 23 percent are weak for determining exposure to more haz- from Government agencies, and 4 percent from ardous substances. Many users have wished for other sources. Twenty percent of these requests quantitative estimates of actual exposure levels, came from establishments with fewer than 100 since these would strengthen the data and increase employees. NIOSH investigators made over 600 its usefulness. Collecting actual exposure data, site visits and produced 234 final reports docu- while theoretically possible, would be expensive menting HHEs conducted. since accuracy and validity would depend on col- In 1978, the General Accounting Office criti- lection and chemical analysis of a large number cized the administration of the HHE program of samples taken over sufficient time to make sure (503). GAO suggested that NOSH needed to pub- the results were representative of workplace con- licize the program more, issue HHE reports more ditions. quickly, establish an evaluation program, and The usefulness of NOHS is reflected by requests disseminate HHE reports more widely, When An- for it and reports developed from the collected thony Robbins was its Director, NIOSH quadru- data. In fiscal year 1982 alone, there were over pled the number of evaluations (465). This rapid 5,000 requests for data from NOHS. Further anal- growth of the HHE program created some con- ysis of potential exposures to hazardous sub- cern in the business community. It has also been stances, including analysis of trends, will be pos- suggested that the the HHE program has reached sible when analysis of NOES, the second survey, too few workers, that the investigations take too is completed. A similar NIOSH environmental long, that NIOSH has attempted to reach conclu- survey focusing on the mining industry is planned sions about toxicity based on too few data, and to comply with the mining surveillance mandate that the program has not found many new work- of the Mine Safety and Health Act of 1977. related illnesses and injuries, especially when compared with the expense of conducting the studies. Health hazard evaluation and technical assist- In some cases, the data gathered by an HHE ance programs. —The identification program area are insufficient to identify a particular cause for also includes the Health Hazard Evaluation (HHE) worker health complaints. For instance, individ- and Technical Assistance (TA) programs. The ual HHEs concerning health problems from un- HHE program responds to employee and employ- known sources of indoor air pollution have sel- er requests from the private sector for evaluation dom identified a causal agent. The accumulated of specific hazards, while the TA program re- findings from these studies, however, have sponds to requests from government agencies. pointed to lack of ventilation as a common fac- However, the technical aspects of HHEs and tor. Frequently, reported illnesses disappear after TAs are the same. The purpose of both of these improvements in ventilation. programs is to “determine toxic effects of chemical, biological, or physical agents . . . in the State surveillance programs.—NIOSH is now workplace” through medical, epidemiologic, and attempting to involve State health departments industrial hygiene investigations of the worksite in surveillance programs for identifying work- of concern. Upon completion of an evaluation, related illness and injury. After surveying State NIOSH reports the results back to the worker and Health Officers to learn about their current ca- to the employer. If the substance proves to be pabilities for carrying out surveillance and pre- toxic as it is used and exposure is not covered by vention programs, NIOSH established agreements any standard, NIOSH is required to report its to cooperate with five States. NIOSH is devising findings to OSHA. methods to assign Epidemic Intelligence Service 246 ● Preventing Illness and injury in the Workplace

Officers to State Health Departments for general bestos dust exposure and a method for grading epidemiological duties related to preventing work- disease severity. An expert committee of pul- related illness and injury. monary pathologists assembled in cooperation with the College of American Pathologists re- Epidemiologic and Toxicologic Studies viewed surigical and autopsy evidence and reached The goal of epidemiologic and toxicologic a consensus as to what the criteria should be. studies is to understand the causes of occupational There have also been investigations of the im- health and safety problems and thus reveal areas munologic aspects of work-related lung disease for preventive actions. Since 1971, NIOSH has to examine the natural defense systems of the lung sponsored a number of studies through its re- and potential ways they are stressed by work- search grants programs and NIOSH staff have related exposures. For instance, the effect of cer- themselves conducted important studies. tain particulate matter often found in the work- NIOSH describes its current research efforts by place was tested on the interferon system in a referring to its list of 10 leading work-related dis- laboratory experiment. One finding was an associ- eases and injuries. These are occupational lung ation of inhibition of interferon production with diseases; musculoskeletal injuries; occupational increasing coal rank (coal, that is, with higher car- cancers; amputations, fractures, eye loss, lacera- bon content such as anthracite). The growth of tions, and traumatic deaths; cardiovascular dis- influenza virus, introduced into a treatment sys- eases; reproductive disorders; neurotoxic dis- tem, increased more among those treated with orders; noise-induced hearing loss; dermatoses; coal particulate than among controls. and psychologic disorders. As examples of cur- Work-related cancer. —Both laboratory and rent NIOSH activities, a few studies of lung dis- epidemiologic studies are conducted to determine ease, cancer, and reproductive health are dis- the carcinogenic potential of workplace exposures. cussed. In cooperation with the National Toxicology Pro- Work-related lung disease. -Lung disease is gram, NIOSH sponsors carcinogenicity assays for the highest priority work-related health problem workplace chemicals. and includes a number of debilitating diseases. As an example of epidemiologic studies, a The health and safety HHS prevention objectives NIOSH retrospective cohort mortality study of (see discussion below) set a goal of virtually no paper and pulp workers found a suggestive in- new cases of asbestosis, byssinosis, silicosis, and crease in lymphatic disease, although associations coal workers’ pneumoconiosis. NIOSH efforts with exposure to workplace agents including concerning lung disease focus on these four dis- wood dust, formaldehyde, sulfur compounds, and eases plus lung cancers and occupational asthma. other chemicals were unclear. A proportional NIOSH studies in this area include animal mortality study of automotive workers suggested testing to study the carcinogenicity of short that makers of wood dies and models suffer more asbestos fibers, those less than 5 um in length. The often than expected from fatal colorectal cancer, study lasted 2 years, at which time the animals’ leukemia, and other cancers. lungs were examined. Asbestos fibers were found NIOSH has developed a registry of workers ex- in the lung, but fibrosis, lung tumors, and other posed to dioxins that contaminate certain chemi- lesions were absent. The researchers concluded cals. Follow-up studies of the members of the that although short asbestos fibers did not by registry will, it is hoped, provide clarifying infor- themselves appear to cause tumors, there is still mation about the possible carcinogenicity of di- a need for further study of the possible interac- oxin. Although information about the exposure tions between short asbestos fibers and other sub- levels of these “dioxin workers” is limited, at least stances that cause cancer. some of them were exposed to much higher levels NIOSH researchers have sponsored the devel- than any “environmental” exposure through air, opment of criteria for the pathologic diagnosis of water, or soil. Therefore, the registry will be im- pleuro-pulmonary disease associated with as- portant in the dioxin controversy. This activity Ch. 12--Governmental Activities Concerning Worker Health and Safety 247 also illustrates that studies of industrial exposures, Development of Controls often carried out by NIOSH, have important impacts on other aspects of public health. Development of control technologies includes developing, assessing, and improving measures Work-related reproductive health problems.— to reduce workplace hazards, especially through Studies of work-related reproductive system control technology, protective equipment, work health problems include both animal studies and practices, and hazard-detection devices. NIOSH pregnancy outcome studies among wives of male investigations of control technologies commands workers. NIOSH tested several industrial chem- a smaller proportion of NIOSH resources than icals—ethylene oxide, propylene oxide, and n-butyl any other research area. In fiscal year 1983 only acetate—for teratogenicity in exposed female ani- 12,8 percent of the NIOSH budget of $57.5 mil- mals. Rats and monkeys were exposed to high lion and less than 14.0 percent of the 911.7 person- levels of these substances prior to breeding. The years of staff time were allocated to the control- test animals had a significant reduction in the technology budget, number of fetal implantation sites and live fetuses, NIOSH’s efforts in control technologies are but it was not possible to conclude that these were divided into three research program areas: about definitely teratogenic effects. The investigators did 21 percent of the control technology budget is conclude that in these cases teratologic effects did spent for control systems research; 14 percent for not occur at exposure levels below those that are respirator research; and 3 percent for other per- toxic to the mother animals. sonal protective equipment research (table 12-8). Epidemiologic studies are being conducted However, the control technology budget also in- among workers exposed to polychlorinated bi- cludes funds for performing chemical analyses in phenyls (PCBs) to determine the effect of exposure support of NIOSH industrial hygiene studies (37 on pregnancy outcome among women. Animal percent); for developing methods for sampling studies of PCB suggest that it is toxic to the fetus and analyzing airborne contaminants (15 percent); and that it may be teratogenic. NIOSH investi- and for testing respirators (10 percent) (584). gators are conducting a case-control study of neonatal deaths and infants with low birth weight Control Systems Research among infants of women workers at a capacitor plant using PCBs. NIOSH conducts Control Technology Assess- The effects of carbon disulfide, which is used ments (CTAs) to identify, evaluate, and document to manufacture rayon, on the central nervous sys- the most effective engineering controls used for tem have long been known, but recent studies particular hazards within a given industry. Infor- mation collected in a CTA is reported back to the have suggested that its health effects also include industry so that other plants can use it to solve increased risk of death and illness from cardio- problems. Table 12-9 shows some of the areas in vascular disease and increased risk of reproductive system effects among both men and women. which CTAs have been done. Other research has been done concerning improved local exhaust ven- Sexual dysfunction, loss of libido, semen abnor- tilation and in the use of air recirculation for gen- malities, and impotence have all been found among exposed male workers and menstrual ab- eral dilution ventilation. normalities and increased risk of fetal loss and Engineering control research concentrates on in- premature births among exposed female workers. dustrial processes, since it is here that toxic sub- NIOSH is now conducting a case-control study stance emissions can be controlled. Research is of the wives of exposed workers to determine the split by industry: chemicals, mining and minerals, potential effects on pregnancy outcome from materials processing, and general industry. Re- fathers exposed to carbon disulfide. cent research includes evaluation of emission con- 248 ● Preventing Illness and Injury in the Workplace

Table 12-8.—NIOSH Budget for Control Technology

Nonstaff Total Person (dollars in (dollars in Program area Projects years thousands) thousands) Percent Control systems. ... , . . 10 29.0 542.9 1,535.8 20.86 Respirator research ., . . 10 16.7 462.6 1,000.0 13.58 Other PPE ...... 3 4.8 99.3 253.7 3.45 Sampling/analysis . . . . . 12 36.9 1,500.8 2,751.8 37.38 Method development, . . 6 24.2 272.8 1,090.5 14.81 Respirator testing . . . . . 4 15.7 224.2 729.4 9.91 Totals ...... 45 127.3 3,102.6 7,361.2 100.00 SOURCE (584)

Table 12-9.—Control Technology Assessments male workers, and menstrual irregularity among Performed by NIOSH in Fiscal Year 1982 exposed female workers. Because the batches of Completed: these products are relatively small, the process has Seals and fittings in chemical processing technology not been mechanized and there are many oppor- Nonferrous metals production including aluminum tunities for worker exposures. Good engineering reduction and nonferrous smelters Fire building controls used in these situations included isola- Foundries tion, highly efficient local exhaust ventilation and, Spray painting as an additional safeguard, the use of supplied- Pesticide manufacture Dry cleaning air suits. Ongoing: Other assessments have been made in industries Dust control in dusty unit operations Chemical processing unit operations expected to grow in future years. These include Unit processes used in general manufacturing hazardous waste disposal, semiconductor manu- SOURCE: (584) facturing, and fermentation processes. The studies of waste disposal included investigations of waste trol of the suspected carcinogen formaldehyde incinerators and automated drum handling. (both in its manufacture and as it is found in adhesives during hot-press wood veneering) and Toxic materials used or found in the manufac- control of worker exposure to styrene, an agent ture of microelectronic components include lead, that can cause dermatitis and neurotoxic illness, arsine, phosphine, boron trifluoride, carbon in boat building plants. tetrafluoride, phosphorous oxychloride, and hy - drofluoric acid. In many cases, NIOSH found NIOSH is also conducting control technology well-designed and effective engineering controls research in selected petroleum-refining operations. and ventilation systems for these hazards, but also Although most processes in this industry are con- discovered one previously unidentified exposure tained, workers may be exposed to toxic sub- problem—arsenic off-gassing of silicon wafers im- stances through leaks, during equipment failure pregnated with arsenic. or maintenance, while collecting quality control samples, while loading or unloading materials, Fermentation processes are expected to be used and during waste treatment. NIOSH investigated more frequently in biotechnology. NIOSH is in- engineering and work practices at a hydrogen vestigating the enzyme production industry to fluoride acid alkylation unit, at a benzene-loading learn which control technologies are most effec- facility, and during other processes in petroleum tive in containing potentially hazardous biological refining. material (582). CTAs have been conducted in some unit proc- Respirator Certification and Research esses. In the pharmaceutical industry, for example, the substances used to manufacture contra- NIOSH, in conjunction with the Mine Safety ceptives are capable of causing gynamastia or and Health Administration, tests and certifies enlarged breasts among both exposed female and respirators. During fiscal year 1982, 44 new res- Ch. 12—Governmental Activities Concerning Worker Health and Safety ● 249 pirators were evaluated for approval and 60 were ment industrial hygiene analytical labs participate evaluated for extension of approval. Audits were in the NIOSH Proficiency Analytical Testing pro- conducted of 33 off-the-shelf respirators to ensure gram, which periodically sends out reference that certified respirators continue to meet test re- samples for analysis. For example, a sample of quirements. Respirator quality-control programs airborne lead dust would be analyzed by each lab- are also reviewed to make sure that certified oratory, NIOSH then collates results from all of respirators continue to meet requirements. As the participating laboratories and reports them detailed in chapter 8, however, there are many back in summary so each lab can see how closely complaints about respirators and about NIOSH’s its results match those of other laboratories. Par- inability to publish new regulations concerning ticipating in this program and performing analy- respirator-testing requirements. ses within the quality control boundaries for the tested substances is required for maintaining lab- Field testing is conducted to measure certified oratory certification by the American Industrial respirator performance in actual working condi- Hygiene Association. tions. Research is under way on sorbent efficiency for organic vapor respirators, filter efficiency and Sampling and Analytical Methods Development optimum flow rates for aerosol air-purifying respirators, the effects of filter resistance and effi- To facilitate accurate and precise assessment of ciency on protection factors and reduced air flow worker exposure to chemical hazards, NIOSH use in powered air-purifying respirators, and the develops and refines methods for sampling and physiological effects of using respirators simul- analysis. In 1981, it published the seventh volume taneously with protective clothing. of the Manual of Analytical Methods, which added 21 methods for monitoring chemical Other Personal Protective Equipment Research hazards. Chapter 8 outlines NIOSH investigations of NIOSH has also participated in the develop- various types of personal protective equipment, ment of new sampling devices. In 1981, it devel- including head protection (hard hats), eye pro- oped performance specifications, testing protocol, tection (protective glasses), face protection (face and evaluation criteria for passive monitors, shields), - foot protection (steel-toed shoes and which are devices requiring only the natural mo- metatarsal guards), hand protection (protective tion of contaminant molecules in air, thus sav- gloves), motion restraints, and protective clothing the costs of sampling pumps. NIOSH and the ing. All of those studies were done in the late Bureau of Mines collaborated in a comparison of 1970s. Currently, resource constraints have lim- X-ray diffraction and infrared spectroscopy for ited research in this area to chemical protective analyzing quartz or crystalline silica. These tests clothing. The absence of assurance that personal were done to help refine reliable, low-cost ana- protective equipment works as advertised could lytical methods. Other sampling research has led be addressed by NIOSH, with sufficient funds and to real-time, direct-reading sampling methods. effort, or by NIOSH in cooperation with other Such devices, while costing more at the outset, organizations. will reduce the overall cost of monitoring chemical workplace hazards by eliminating the costs, Sampling and Analysis risk of error, and delays in obtaining results that are associated with laboratory analysis. Analysis of chemical samples is necessary for many industrial hygiene and medical studies. In fiscal year 1982, 17,200 exposure samples were Dissemination analyzed for 42,000 determinations. About one- NIOSH disseminates its research findings through quarter of the samples were done in NIOSH lab- criteria documents, reports, and published papers oratories and three-quarters on contract. informing professionals and the public of identi- NIOSH also has a role in laboratory quality fied problems and solutions, as well as through control. Approximately 375 private and govern- Health Hazard Evalution reports. Efforts are now 250 Preventing Illness and Injury in the Workplace

being made to disseminate HHE findings to hard- research division-the Division of Surveillance, to-reach audiences. For example, a finding that Hazard Evaluation, and Field Studies—submitted irritating vapors from duplicating machines were over 300 reports on industrial hygiene and medi- affecting teachers’ aides and that an easy way to cal studies to the National Technical Information control the exposure was available was dissem- Service, published 63 articles in technical journals, inated through teachers’ organizations. published 3 articles on research findings describing hazards and means for reducing them in in- The OSH Act provides that NIOSH shall make dustry and labor trade journals, had 6 articles in recommendations to OSHA concerning health the Morbidity and Mortality Weekly Report, pub- and safety standards. Since 1971, NIOSH has lished 2 summaries of approximately 80 recently transmitted to OSHA over 100 “criteria docu- completed HHEs, and provided over 200 reports merits, ” which contain extensive bibliographies of to people requesting information from NOHS. available scientific literature on the chemical or process in question, followed by an analysis and The OSH Act requires that MOSH publish an- assessment of the substance’s toxicity and a rec- nually a list of all known toxic substances or ommendation to OSHA for a potential standard. groups of toxic substances, their observed effects, The criteria documentation process has covered and the concentrations at which they occurred. about 151 substances since 1970. Priority for the In compliance with this requirement, NIOSH pub- substances for which criteria documents were lishes the Registry of the Toxic Effects of Chemi- written was based in part on estimates of the num- cal Substances. The current edition contains ber of workers exposed, the volume of produc- 218,746 listings of chemical names, of which tion of the substance, and the severity of toxicity. 59,224 are different chemicals (the rest of the The third NIOSH Director, Anthony Robbins, names are synonyms). The listings are extracted deemphasized the production of criteria docu- from the published literature, including journals ments in part because many of them “were mostly published abroad, but are not evaluated for qual- toxicological reviews” that “ . . . did not contain ity, accuracy, or reproducibility, The substance name is given, followed by its synonyms, chemi- a great deal of epidemiology, which is needed. They were inconsistent” (482). In ad- cal data, and toxicity data by the general catego- dition, as discussed in chapter 13, OSHA has not ries of irritation, mutation, reproductive effects, issued standards for most of the hazards addressed tumorigenic effects, and general toxicity. The cita- by NIOSH criteria documents. tion for each toxic effect reported is included so that the orginal article can be found for further Few criteria documents are being developed detail. now, but from time to time NIOSH has prepared other kinds of reports, transmitting them to NIOSH is also becoming involved in education OSHA, as well as disseminating them to the pub- to prevent work-related illness and injury. A lic. These include Occupational Hazard Assess- health motivation working group has been formed ments, which contain recommendations but that to examine ways in which NIOSH’s research could are less thorough and less specific than criteria be applied to combine health protection (control documents. In addition, NIOSH prepares Current of work-related illness and injury) and health Intelligence Bulletins, which contain new findings promotion (improvement of personal health about workplace hazards, are also published and behaviors). transmitted to OSHA, to worker and employer In addition, NIOSH has responsibility for the representatives, and to health and safety profes- education of occupational safety and health pro- sionals. NIOSH also participates in the public fessionals through both direct short-term train- hearings that are part of OSHA’s standard-setting ing and academic programs, It is estimated that process and provides recommendations concern- over 5,000 professionals have received training ing the standard under consideration. through these NIOSH programs. (These activi- NIOSH technical publications are widely dis- ties are described in detail in ch. 10. ) Training seminated. In fiscal year 1982, for example, one grants for Educational Resource Centers have Ch. 12–Governmental Activities Concerning Worker Health and Safety ● 251 been successful in establishing coordinated multi- objectives. The Prevention Objectives, shown in disciplinary programs. They have received some Box O, are divided into several categories: attention recently because of proposals by the cur- ● Improved health status (measured by fewer rent administration to eliminate their Federal deaths, injuries, and illnesses); funding (102). ● Reduced risk factors (by implementing hazard controls); ● Improved public/professional awareness (as NIOSH Priorities reflected by increased worker, employer, and professional knowledge about occupational Setting priorities for preventing work-related hazards); illness and injury has been a challenge to NIOSH ● Improved services/protection (through the management. During John Finklea’s tenure as Di- use of generic standards and increased rector, NIOSH activity was concentrated on the NIOSH activity in studying hazards); and production of recommendations to OSHA con- ● Improved surveillance/evahation (including cerning health and safety standards-NIOSH’s creation of coding systems and enhancing ex- “criteria documents. ” He believed that the criteria isting efforts to include occupational factors) documents should be the primary product of (556). NIOSH and that once published would be considered public health policy documents that would The NIOSH list of 10 leading work-related force OSHA into setting and revising health health problems (see table 5 in ch. 3) was com- standards. piled by NIOSH’s division directors. Frequency of occurrence, severity, and amenability to pre- Anthony Robbins, the next NIOSH Director, vention of the work-related injury and illness were reemphasized criteria documents and focused the criteria used for selection (583). This priority NIOSH activity on health hazard evaluations and list is used by NIOSH to identify subjects for cri- epidemiologic studies (465,482). The fourth teria document development and to set priorities NIOSH Director, Donald Millar, has set special for research for hazard assessment and control. emphasis on efforts to assure high-quality research It was also published to encourage discussion and focus research on the most important work- among occupational safety and health profession- related problems. He is also working to expand als and to assist them in setting their control pri- the participation of State and local health agen- orities. NIOSH intends to collect data on the 10 cies and to increase workplace health promotion and periodically update it as conditions change efforts (584). (581). Present research priorities are influenced by the The National Toxicology Program coordinates HHS Prevention Objectives for the Nation, the the Federal Government’s testing of chemicals for NIOSH list of 10 leading work-related health possible human health hazards. Two of its activ- problems, and the HHS National Toxicology Pro- ites are important for NIOSH. First, it carries out gram. These are used to identify subjects for cri- the testing of substances that NIOSH identifies as teria document development and to set priorities possible concerns, and NIOSH contributes to the for research on hazard assessment and control. costs of those tests. Second, the results of tests The Prevention Objectives include 20 objectives requested by other agencies are also provided, so for preventing work-related injury and illness. NIOSH can evaluate the potential workplace haz- Thirteen of these have been designated priority ards presented by these substances. [page omitted] This page was originally printed on a dark gray background. The scanned version of the page was almost entirely black and not usable. Ch. 72—Governmenta Activities Concerning Worker Health and Safety 253

’16. By 1990, generic standards and-other forms of technology transfer should be established, where possible, for standardized employer attention to such major common problems as: chronic lung hazards, neurological hazards, carcinogenic hazards, mutagenic hazards, teratogenic hazards, and medical monitoring requirements. 17. By 1990, the number of health hazard evaluations being performed annually should increase tenfold, and the number of industry-wide studies being formed annually should increase three- fold. (In 1979, NIOSH performed approximately 450 health hazard evaluations; 50 industry-wide studies were performed.) Improved surveillance/evaluation: -18. By 1985, an ongoing occupational health hazard/illness/injury coding system, survey, and surveillance capability should be developed, including identification of workplace hazards and related health effects, including cancer, coronary heart disease, and reproductive effects. This system should include adequate measurements of the severity of work-related disabling injuries. 19. By 1985, at least one question about lifetime work history and known exposures to hazardous substances should be added to all appropriate existing health data reporting systems, e.g., cancer registries, hospital discharge abstracts, and death certificates. 20. By 1985, a program should be developed to: 1) follow up individual findings from health hazard and health evaluations, reports from unions and management, and other existing surveillance sources of clinical and epidemiological data, and 2) use the findings to determine the etiology, natural history, and mechanisms of suspected occupational disease and injury.

“Priority objectives of NIOSH.

SOURCE: (556).

SUMMARY

OSHA and NIOSH were created by the Occu- startup standards have provided some worker pational Safety and Health Act of 1970 and both protection, there were many problems with these began operations in 1971. The separation of reg- standards and OSHA has been slow in revising ulatory activities from research may help ensure them. Since 1971, OSHA has completed 18 sepa- the objectivity of the research, but it has also been rate proceedings for setting health standards and criticized because oftentimes the two agencies 26 proceedings for safety standards. However, the have not coordinated their activities. After a scope of some of these standards has been very period of growth during the 1970s, the budgets limited, and several involved only a rewriting of for both OSHA and NIOSH have declined in re- requirements that had been issued earlier by cent years. After adjusting for inflation, OSHA’s OSHA. Those that involved major changes have has decreased nearly 13 percent since its peak in frequently been the subjects of judicial review. 1979. NIOSH’s has declined 42 percent since a OSHA has also published nine emergency tempo- peak in 1980 and is now lower, in real terms, than rary standards, but the courts have been reluc- any NIOSH budget since 1973. tant to support these standards. OSHA’s main activities are setting occupational OSHA inspection activity has generated a great health and safety standards, conducting work- deal of controversy because employers can be place inspections to ensure compliance with those compelled to install health and safety control tech- standards, monitoring State programs, and pro- nology. Together, Federal OSHA and the State viding for several different types of educational programs operating under the OSH Act con- and service programs. Although many of OSHA’s ducted about 162,000 establishment inspections 254 ● Preventing Illness and Injury in the Workplace and 13,000 “records review” visits in fiscal year professional education. The largest share of the 1983. However, for most workplaces, the threat NIOSH budget is now devoted to identification of an inspection is quite low. On average, OSHA of hazards. This includes a large survey of penalties are also low, often lower than the costs workplaces to determine the extent of potential of many controls. exposures to toxic substances, the Health Hazard Evaluation program, and NIOSH research in epi- OSHA’s public education and service activities demiology and toxicology. include its consultation program, New Directions grants program, and Voluntary Protection Pro- Control technology receives a relatively small grams. All these programs attempt to approach portion of the NIOSH budget. This research in- health and safety in a way that is more coopera- cludes Control Technology Assessments, respira- tive than OSHA citations and penalties. But there tor certification and research, research on other have been concerns expressed about OSHA’s types of personal protective equipment, as well management of funds in these areas, the quality as developing sampling and analytical techniques of the work performed, and the relationship of and performing laboratory analysis of exposure these activities to OSHA’s inspections. samples. Information dissemination includes the preparation of reports on Health Hazard Evalua- Federal OSHA also monitors the 25 jurisdiction, industry-wide studies, Control Technology tions that have assumed responsibility for the Assessments, and guides to good practice. Sever- health and safety of all or some of the work force al NIOSH publications are standard reference within their borders. These State programs can sources for industrial hygiene, and the Registry potentially meet local needs better than a cen- of the Toxic Effects of Chemical Substances is a tralized Federal agency and can enhance the total comprehensive listing of the literature about the governmental resources devoted to worker health toxicity of over 59,000 different chemicals. NIOSH and safety. But there has been controversy here has also prepared Criteria Documents containing too, especially concerning appropriate staffing recommendations to OSHA concerning new or levels for these operations. revised health and safety standards. Finally, NIOSH activities are chiefly related to hazard NIOSH grant programs have provided for the identification, research on control technology, education of a number of health and safety pro- and providing for information dissemination and fessionals. 13. Assessment of OSHA and NIOSH Activities

\ Contents

Page Comparison of Recommendations and Standards...... 257 Methodology for Comparison ...... 258 Results...... 259 Discussion ...... 260 Impacts of OSHA...... 262 . Cost of OSHA Compliance ...... 262 OSHA’s Impact on Injury Rates ...... 264 Trends in Exposure Levels. . 26a Conclusions About OSHA’s Impacts of NIOSH ...... 269 Summary ...... 270 LIST OF TABLES Table Page 13-1. Comparison of Protective Levels for selected substances...... 259 13-2. McGraw Hill Survey of Capital Spending for Worker Health and Safety ...... 263 .—.

Assessment of OSHA and NIOSH Activities

This chapter presents a discussion of several as- NIOSH, and the American Conference of Gov- pects of the activities of the Occupational Safety ernmental Industrial Hygienists (ACGIH). Second and Health Administration (OSHA) and the Na- is a discussion of estimates of employer invest- tional Institute for Occupational Safety and ments in health and safety, followed by a sum- Health (NIOSH). It is not a complete assessment mary of the research on the effectiveness of OSHA of Federal and State activities concerning occupa- in reducing injury rates and toxic exposures. tional health and safety, but only a relatively brief Finally, there is a short discussion of the dif- treatment of several important areas. First is a ficulties in assessing the results of NIOSH presentation of the results of an OTA comparison activities. of the standards and recommendations of OSHA,

COMPARISON OF RECOMMENDATIONS AND STANDARDS

This section discusses an OTA comparison of the current time, however, this is not required. current OSHA standards for certain toxic and haz- (As discussed in ch. 16, one option might be to ardous substances with NIOSH Recommended encourage companies to meet more protective Standards and with the Threshold Limit Values limits when undergoing “reindustrialization. ”) (TLVs) recommended by the American Confer- OSHA currently has standards for about 410 ence of Governmental Industrial Hygienists. This chemical substances. In most cases, OSHA speci- comparison provides an analysis of the stringency fies the maximum levels for employee exposures of OSHA standards, and NIOSH and ACGIH rec- (the Permissible Exposure Limits, or PELs). As de- ommendations. In most cases, OSHA standards scribed in chapter 12, nearly 400 of these stand- for chemical exposure appear to be less stringent. ards were adopted in 1971 by OSHA under sec- When there is more than one required or rection 6(a) and consisted of consensus standards and ommended level, engineers and designers can find established Federal standards. In addition, OSHA themselves in a quandary as they seek to design has issued 18 separate health standards. Twelve a productive process and a system for the proof these covered 24 specific substances and one tection of worker health. Obviously, the designers physical agent. Among these 12 standards, there must comply with the OSHA standard and can- were 10 that set new or revised Permissible Ex- not choose to comply only with a less protective posure Limits. recommendation from another organization. Each year the ACGIH publishes a list of TLVs. But when the recommendations of NIOSH or These are recommended to the ACGIH member- ACGIH are more protective than OSHA stand- ship by the TLV Airborne Contaminant Commit- ards, it would be desirable, from a health stand- tee—17 members and 9 nonvoting consultants point, for companies to adhere to these more pro- who represent companies and other countries. tective recommendations. This is especially true ACGIH has a second committee that considers when building new facilities or rebuilding old recommendations for physical agents. Although plants, when it is possible to increase worker pro- qualified technical consultants from unions have tection and to take advantage of the reduced costs been and are sought by ACGIH (244), unions of controls introduced as part of the design. At have not participated on either of these commit-

257 258 . Preventing Illness and Injury in the Workplace tees in recent years because they believe that by OSHA. For coke oven emissions, a standard standard setting should be a Government activ- was issued in 1976. For noise, OSHA issued a ity. The committee members are unpaid and may Hearing Conservation Amendment in 1981 and meet two to four times each year to deliberate. revised it in 1983. For both versions of the Hear- New TLVs that are accepted by the membership ing Conservation Amendment, however, the pub- at the annual business meeting are placed on a lished provisions covered only monitoring, audio- list of “intended changes” for at least 2 years, dur- metric testing, hearing protection, training, ing which comments and additional data are re- warning signs, and record keeping. The NIOSH quested. At the end of this period the member- recommendation to lower the Permissible Expo- ship votes on whether to add the TLVs to the list sure Limit from an 8-hour time-weighted average of Threshold Limit Values. of 90 decibels to 85 decibels has not been acted on. (The ACGIH TLV for continuous noise is also The 1983-84 TLV list contains recommended 85 decibels for an 8-hour exposure. ) exposure limits for 615 chemical substances and mineral dusts. In May 1984, ACGIH adopted Finally, NIOSH has made recommendations TLVs for 15 substances, and proposed to add concerning 148 specific chemical substances. But TLVs for 7 substances not currently on their list only 123 of those recommendations include a as well as to revise the TLVs for 21 other sub- specified numerical exposure limit. stances (352). Changes in ACGIH TLVs are not automatically incorporated by OSHA, even for Methodology for Comparison the PELs originally based on the 1968 ACGIH list. OTA’s analysis compares the protective levels As outlined in the Occupational” Safety and either recommended or required by NIOSH, Health (OSH) Act, NIOSH makes recommenda- ACGIH, and OSHA, for the 123 specific sub- tions, in the form of Criteria Documents or other stances included on the NIOSH list. The use of documents, to OSHA concerning health and safe- the NIOSH data set as a basis for comparison does ty standards. It was apparently intended that not mean that the NIOSH exposure levels are the OSHA would issue mandatory standards after re- most important. The NIOSH data are simply a ceiving recommendations from NIOSH. convenient source for analysis. (See app. A for A NIOSH list, Recommendations for Occupa- a discussion of the selection of substances and the tional Health Standards, details the proposals major points of inconsistency among these orga- made to OSHA since 1971 for exposures to 163 nizations. ) hazardous substances and working conditions. Unfortunately, OSHA, NIOSH, and ACGIH Five of these concern hazardous working condi- use different terminology for describing their tions: logging, hot environments, coal gasifica- standards and recommendations. OSHA refers to tion, confined spaces, and emergency egress from the basic requirement of its health standards as elevated work stations. None of these five has 1 a “Permissible Exposure Limit. ” This is the re- resulted in a completed OSHA rulemaking. quirement that employee exposures be kept below Ten recommendations concern exposures to a specified numerical limit. NIOSH prepares “Cri- categories of hazardous substances or to harm- teria for Recommended Standards” and transmits ful physical agents (benzidene dyes, chrysene, those recommendations to OSHA. ACGIH uses coke oven emissions, fluorocarbon decomposition the term Threshold Limit Values for its recom- products, ethylene thiourea, kepone, noise, pes- mended exposure limits. ticides, ultraviolet radiation, waste anesthetic For these comparisons, OTA has used the term gases). Only two of these, coke oven emissions “Protective Level” to refer to the standards, rec- and noise, have resulted in any regulatory action ommendations, and TLVs. Protective Levels can be specified in two basic ways: K)ne other NIOSH recommendation, concerning underwater diving operations, is not included in the NIOSH list. This recommen- ● time-weighted averages (TWAs), and dation was issued in August 1976 after OSHA had already issued an Emergency Temporary Standard on diving operations. OSHA’S ● ceiling limits, short-term limits, or peak final standard was published in July 1977. levels. Ch. 13–Assessment of OSHA and NIOSH Activites 259

Protective Levels can have an 8-hour Time- health effects and the dose-response relationship Weighted Average, a Ceiling Limit, or both. for each of these 123 substances. Thus, the results ACGIH recommendations often have TLVs for only describe the relative stringency of the stand- both a TWA and a short-term limit. OSHA standards and recommendations of these three orga- ards often have only an 8-hour TWA. NIOSH rec- nizations. ommendations also generally have just one Pro- tective Level. A Protective Level is defined to be only the Results number listed as the PEL, recommendation, or The rows of table 13-1 present three different TLV. No attempt has been made to quantify the comparisons –OSHA and NIOSH, OSHA and number of workers exposed either above or below ACGIH, and NIOSH and ACGIH. To simplify that Protective Level or to assess the additional presentation, in most cases the table provides only protection provided by monitoring, medical sur- the number of cases for which the first organiza- veillance, and other health and safety activities tion listed has a numerical Protective Level that often included in standards and recommendations. is less than the corresponding Protective Level of This comparison does not quantify all possi- the second organization. bIe aspects of worker protection. Rather, it rep- For example, the OSHA standard (TWA) for resents only a numerical comparison of the stand- 3 carbon dioxide equals 9,000 mg/m , while the ards and recommendations for a group of toxic NIOSH recommendation is 18,000 mg/m3 (see substances. The results of this comparison are ex- table A-13 in app. A). In this case, the OSHA pressed in terms of strictness or stringency. A standard is more stringent and would be included standard or recommendation is more stringent in the total number of cases presented in the first than another if the specified numerical exposure column of table 13-1. For phosgene, both OSHA limit is lower. Depending on the nature of the haz- and NIOSH have the same protective level for an ard, the expected health effects, the relationship 3 8-hour TWA (0.4 mg/m ). Because these TWAs between exposure and these health effects, are equal, phosgene is not included in the first col- “stringency” and the degree of worker protection umn of the table. But because NIOSH has also afforded by a standard or recommendation are set a Ceiling Limit, while OSHA has not, there usually closely related. would be an entry in the second column of table There can be cases for which a more stringent 13-I. There will be an entry in column 3 or col- protective level does not provide improved pro- umn 4 only if both the TWA and ceiling limits tection. However, for this comparison, OTA has are more stringent or equal. (Although not in- not conducted a detailed analysis of expected cluded in the table, the number of times that two

Table 13-l .—Comparison of Protective Levels for Selected Substances

Number (percent) of substances Both TWA and Both TWA and TWAb is more Ceiling Limitb is Ceiling Limits are Ceiling Limits are Case stringent more stringent more stringent equally stringent OSHA standards compared with NIOSH recommendations...... 28 (23%) 5 ( 4%) 2 ( 2%) 22 (18°/0) OSHA standards compared with ACGIH TLVsb ...... 11 ( 9%) 12 (10%) 10 ( 8%) 15 (12%) NIOSH recommendations compared with ACGIH TLVsb ...... 35 (28%) 50 (41%) 28 (23%) 15 (12%) aFOr 123 ~ubstarrce~ for which NIOSH has made recommendations to OSHA. %WA Is an abbreviation for Time-Weighted Average exposure level. The TWAS compared are typicstly calculated for an &hour workday and a 40-hour workweek. Ceiling Lirnlts are deflnad for this comparison, in most caaes, as a 15-minute Time-Weighted Average exposura level ACGIH uses the term Short Term Exposure Limit (STEL) for this. TLV or Thre8hold Llmlt Value is ACGIH’S term for Its recommendations concerning airborne concentrations of substances. See text and app. A for a more complete discussion. cThe num~r presented Is the num~r of cases for which the first Organlzatlon ]lsted has a prot~tlve level more Stringent tharl the Swond Orgaflkdioll. SOURCE: Office of Technology Assessment. 260 ● Preventing Illness and Injury in the Workplace organizations have identical TWAs or ceiling this is because ACGIH has not set any TLVs for limits is included in the following discussion. ) these substances. NIOSH has stricter TWAs than ACGIH in 35 cases (28 percent) and stricter Ceil- Overall, NIOSH recommendations are more ings in so cases (41 percent). NIOSH recommen- strict than the OSHA standards. Even when the dations and ACGIH TLVs are equal in 15 cases Protective Levels are separated into TWA and (12 percent). Sometimes one set of Protective Ceiling Limits, OSHA generally permits higher Levels, such as the TWA, is equivalent and the exposures. Only 28 TWA PELs set by OSHA are other is not (33 cases), For example, the TWAs more stringent than NIOSH’s. In all but 2 of those for xylene are the same for both NIOSH and cases (lead and carbon dioxide) the TWAs used ACGIH, but ACGIH’s Ceiling is stricter than by OSHA appear to be more stringent because NIOSH’s. There is only one chemical with an NIOSH did not set a TWA but only set a Ceiling OSHA PEL and an ACGIH TLV for which NIOSH Limit. In many cases (22 of the 28), this NIOSH has not set an exposure limit (boron trifluoride). Ceiling Limit is lower than or equal to the TWA used by OSHA. There are, however, 5 instances in which the OSHA Ceiling Limit is more strict than the NIOSH Discussion one, although in each case this is because NIOSH These three organizations use different formal has not recommended a Ceiling Limit. There are decision rules when setting their standards and 33 instances (27 percent) where OSHA PELs are recommendations. However, in practice, many equal to NIOSH recommendations, either in the more informal factors enter into the decisions, in- TWA or Ceiling or both. Of these, 22 represent cluding the experience of the individuals involved, situations in which both the TWA and Ceiling their professional judgments on what is protec- Protective Levels are identical. tive of worker health, and the values and goals As mentioned, ACGIH updates its TLVs every for public policy that are held by these individuals year but OSHA’s PELs remain virtually frozen at and their organizations. the levels adopted by ACGIH in 1968. Obviously, For OSHA and NIOSH, the formal guidelines then, OSHA PELs and ACGIH TLVs will be equal for recommendations and standards derive from in those cases in which ACGIH has not changed the OSH Act, which appears to distinguish be- its 1968 levels (15 instances). It is noteworthy that tween the “criteria” developed by NIOSH and the ACGIH now has TLVs for over 600 toxic and haz- “standards” set by OSHA. NIOSH is required to ardous substances—some 200 substances more develop “criteria” solely on health and safety than OSHA regulates. grounds, without consideration of the feasibility OSHA standards are stricter overall than ACGIH of the “criteria. ” OSHA, on the other hand, is TLVs in 10 cases (8 percent). ACGIH has no rec- directed to set standards that are “reasonably nec- ommendation for 4 of those substances: acetylene, essary or appropriate” and, in the case of health DBCP, ethylene dibromide, and petroleum sol- standards, to protect employee health “to the ex- vents. OSHA’s TWAs are stricter than ACGIH tent feasible. ” (For a discussion of OSHA deci- TLVs in 11 cases (9 percent) and OSHA’s Ceiling sionmaking, see ch. 14. ) Limits are stricter than ACGIH’s in 12 cases (10 ACGIH sets its TLVs at levels “under which it percent). In all but 6 of these, if the TWA used is believed that nearly all workers may be repeat- by ACGIH is weaker than OSHA’s, then the edly exposed day after day without adverse ef- ACGIH Ceiling Limit is stricter than OSHA’s (or fect.” ACGIH cautions, however, that because of vice versa). On the whole, instances where OSHA variation in individual susceptibility, some work- is more strict occur because ACGIH has not made ers may still develop an occupational illness from any recommendation at all. exposures at a TLV. The TLVs “should be used NIOSH recommendations are stricter than as guides in the control of health hazards and ACGIH TLVs for 28 substances (23 percent) out should not be used as fine lines between safe and of the 123. In 4 instances, as mentioned above, dangerous concentrations” (8). Ch. 13—Assessment of OSHA and NIOSH Activities ● 261

This comparison shows, in a limited way, that In 1983, it was reported that a memorandum OSHA has tended to lag behind NIOSH’s and had been prepared by OSHA staff recommend- ACGIH’s Protective Levels. There are several in- ing that the agency stop working on the devel- terrelated reasons for this lag. The first, and most opment of revised exposure standards for 116 sub- important, is that OSHA is a governmental reg- stances that are regulated under OSHA’s startup ulatory agency. Employers can, if they wish, ig- standards. OSHA’s activities concerning these 116 nore NIOSH recommendations and ACGIH TLVs. chemicals include requests for information on 61 But even though OSHA’s inspection activity is of the chemicals, preparation of advance notices limited (see ch. 12), OSHA standards can poten- of proposed rulemaking on 55 of them, issuance tially be used as the basis for civil penalties and of proposed rules for 23, and hearings on rules required abatement. Thus they receive more at- to regulate 9 of the chemicals. According to the tention. memorandum, a decision to stop working on rules for these chemicals would be practical, as the Opposition has often accompanied that atten- agency no longer has the staff or resources to pur- tion. Most of OSHA’s health standards have been sue this activity, Although OSHA has stated that challenged in the courts. Resolution of these cases a decision not to pursue regulation of these chem- has taken time–in most cases, several years. The icals has not been made, there has been little re- courts now require that OSHA standards be based cent activity regarding these 116 substances. The on adequate evidence that the hazard addressed last hearing to be held on any of them concerned by the standard poses a “significant risk” and that beryllium, in 1977, and there has been no public compliance is “feasible” for the affected industry. activity on any of the others since 1975 (110,366, Moreover, Executive Orders since 1974 have re- 517a). quired that OSHA examine the economic impact of its standards prior to issuing them. (For a dis- As discussed earlier, from 1971 to 1984, OSHA cussion of this aspect of OSHA decisionmaking, issued 18 separate health standards. But only 12 see ch. 14. ) of these established Permissible Exposure Limits or other requirements for 25 specific exposures. In addition, although OSHA is to use NIOSH’s Requirements for 3 of these have been overturned recommendations as one element for its standard- by the courts. Thus, of the approximately 410 setting, it is not required to respond to those rec- OSHA-regulated substances, nearly all are still the ommendations within any specified deadline. Be- same as those OSHA initially adopted in 1971. yond this, OSHA has often concluded that the NIOSH Criteria Documents did not provide an In direct contrast to this is the experience of adequate basis for developing standards. Thus some European nations. For example, in July OSHA has usually developed its own scientific 1984, Sweden issued a revised list of health stand- and technical information concerning hazards and ards that included 18 revisions of existing exposure controls. These factual and legal requirements, as limits and the addition of limits for 22 new sub- well as the requirements of the Executive Orders, stances previously not regulated. This is the re- have consumed time, stressed OSHA’s limited result of only 3 years of effort, as the last revision sources for standard-setting, and slowed the of the Swedish health standards was published in agency. June 1981. Although it is generally believed that the Swedes are leaders in occupational health and NIOSH and ACGIH have their own resource safety, it is difficult to gather quantitative infor- constraints, too. But they do not have the same mation on actual exposure conditions in Sweden potential for public opposition and legal chal- and the United States. Thus, it is not clear to what lenge. They can issue recommendations based on extent the Swedish limits are translated into ac- their judgments about employee protection from tual protection. It is clear, however, that the proc- adverse health effects, without formal analysis of ess of revising the exposure limits in the United the feasibility or costs of these recommendations. States is dramatically slower. 262 . Preventing Illness and Injury in the Workplace

IMPACTS OF OSHA Cost of OSHA Compliance merely represent costs previously borne by injured workers in terms of lost wages, pain, and suffer- There are a number of difficulties and problems ing. In addition, the increased costs to employers in securing accurate and meaningful information often represent increased revenues for suppliers on the costs of health and safety regulation. These of control technologies and an increased number problems generally plague both estimates of the of jobs for health and safety professionals. More- costs of any particular regulation or proposed reg- over, when complying with regulations, employ- ulatory change and estimates of the total costs of ers often are able to offset, at least partially, the all OSHA regulations. costs of compliance with improvements in pro- First, the basic source of data on the costs of ductivity and reductions in the costs of lost wages, compliance with regulations is employers, either medical care, down time, etc. that are associated because they are the most knowledgeable about with workplace injuries and illnesses. day-to-day conditions and equipment in their plants or because they have given permission to Third, there are several difficulties in correctly outsiders to visit their facilities. But businesses separating and attributing health and safety ex- generally do not arrange their accounting systems penditures. The first of these concerns how to al- to keep track of all health and safety expenditures. locate the costs of equipment and personnel that Although information is sometimes available for perform multiple functions, including normal certain health and safety expenses (e.g., the sal- business activity, or that enable compliance with ary of a company physician), there is generally several regulations. For example, an employer inno account entry for the total expenditures for stalls a ventilation system that both reduces the health and safety. Thus, estimates of current costs level of a toxic air contaminant and provides heat- and predictions of future costs are generally based ing and air conditioning. What percentage of the on inexact information and guesses rather than cost of the duct work, fans, switches, installation on data that can be verified and audited. costs, etc. should be attributed to the industrial hygiene function and what percentage is a nor- Furthermore, when asked to provide these esti- mal business expense for heating and air condi- mates, there is a tendency for employers to over- tioning? estimate regulatory costs. Employers have a strategic interest, if they oppose a particular reg- A second, and related, difficulty arises most ulation or Government intervention in general, particularly in attempts to estimate the costs rein overestimating the costs. In addition, lower- lated to existing OSHA regulations or proposals and middle-level managers, when asked how for new standards. This concerns how to sepa- much a regulation will cost in their areas of rate incremental costs (those due solely to the pres- responsibility, will tend to overestimate the costs ence of regulation) from total costs, which include because, in the words of a senior official at spending that would have taken place even in the ALCOA, “it is better to be under budget than over absence of regulation. These could include the budget” (140). Facing the possibility that they will costs incurred because of corporate good citizen- have to budget expenditures within the estimates ship, collective bargaining agreements, and fear they make, managers may overestimate costs in of legal liability. For new, more stringent regula- order to have a sufficient budget. tions, the distinction is often made between the total costs of compliance and the incremental Second, compliance cost estimates have gener- costs for reducing hazards beyond the existing really not been offset by the benefits of health and quirements. safety compliance. In fact, in many cases what appear as new costs to the business firm are Finally, cost estimates generally assume the use merely costs that previously had been borne by of currently available technologies, neglecting the workers and society. For example, the costs to a potential for cost savings with improvements in business of preventing occupational injuries may control technology. Over time, industry may —

Ch. 13—Assessment of OSHA and NIOSH Activities ● 263 learn how to control more cheaply and control Table 13-2.—McGraw Hill Survey of Capital Spending technologies themselves may improve. Thus con- for Worker Health and Safety trol may become less costly. Basing the costs on Health and safety investment current technology will thus tend to overestimate Current 1972 Percent of the costs of compliance (33). dollars dollars a capital Year (millions) (millions) spending Estimates of Total OSHA Compliance Costs 1972 ...... 3,279 3,279 2.7 1973 ...... 3,616 3,552 2.6 Robert Smith (446) has reported a National 1974 ... , . . 4,403 4,028 2.8 1975 ...... 3,842 3,044 2.4 Association of Manufacturers estimate that its 1976 ...... 3,415 2,550 2.0 members needed to spend between $35,000 and 1977 ...... 4,291 3,043 2.2 $350,000 each to comply with OSHA startup 1978 ...... 6,645 4,439 2.9 1979 ...... 4,317 2,719 1.6 standards. No other details of the methodology 1980 ...... 4,128 2,441 1.4 behind this estimate were given. 1981 ...... 5,112 2,848 1.6 1982 ...... 4,503 2,459 1.4 The Economics Department of the McGraw- 1983 ...... 4,945 2,704 1.6 Hill Publications Co. annually surveys a group aAdjuSted to Ig7z dollars using the implicit price deflator for nonresidential wO- ducers’ durable equipment Economic Reporf of the Preslderrt (1984), table B-3 of large companies about actual and planned cap- SOURCE (298) ital expenditures and about the percentage of their capital spending that is for employee health and safety and for air, water, and solid waste pollu- fell below an average for all industries of 2.0 per- tion control, McGraw-Hill then develops national cent from 1972 to 1978, has been in the range of estimates based on these survey responses (298). 1.4 to 1.6 percent for the last 5 years for which There is no independent verification of the survey data are available (298). This decline in recent responses, although a limited check is conducted years in the percentage share of capital spending to ensure internal validity of survey question- devoted to health and safety is similar to the de- naires. Information is not collected on annual cline in the share of capital spending for pollu- operating expenses, but only on capital spending. tion control (410). Nor is information collected on what companies Another source of information on OSHA com- would be spending on health and safety even in pliance costs is a study prepared for the Business the absence of OSHA activity. Roundtable by the accounting firm of Arthur The total capital spending for occupational Andersen & Co. They surveyed 48 very large cor- health and safety was estimated to be $4.5 bil- porations to estimate the incremental costs of lion in 1982 and $4.9 billion in 1983, represent- Government regulations of six Federal agencies ing 1.4 percent and 1.6 percent of total capital in 1 year-1977. The 48 companies were estimated spending, respectively (table 13-2). When meas- to incur about $2.6 billion in compliance costs. ured in current dollars, capital spending for health Of this, only $184 million or about 7 percent was and safety has ranged from $3.3 billion to $6.6 attributed to OSHA. About 37 percent of the billion (from $2.4 billion to $4.4 billion in 1972 OSHA compliance costs were capital costs, 56 dollars). The latest estimate of investment, about percent operating and administrative expenses, $5 billion, is only about one-sixth as large as the and 6 percent research and development costs National Safety Council estimate regarding the (26). costs of work injuries alone—over $30 billion each Wiedenbaum and DeFina have developed what year (324, 530). is probably the most widely quoted estimates of In both current and real dollars the peak for the costs of Federal regulation. Their estimate for reported capital spending appeared in 1978. From the total costs of all Federal regulations in 1976 1979 to 1983, the expenditure ranged between $2.4 amounted to about $65 billion (669), while for and $2.9 billion (in 1972 dollars). In addition, the 1979 their estimate was over $100 billion (670). share of total capital spending that has been de- (See 195,428,470 for criticism of these estimates.) voted to employee health and safety, which never They did not, however, develop any new or origi- 264 ● Preventing Illness and Injury in the Workplace

nal cost estimates for OSHA compliance. Rather, Thus, the estimated adverse impact of OSHA their estimate for OSHA was based on the McGraw- on productivity is quite small. Moreover, in some Hill survey plus an estimate of the costs incurred cases OSHA compliance has been accompanied by universities for OSHA compliance. by improvements in productivity. As discussed in box N (in ch. 12) and chapter 16, these cases OSHA and a number of participants in its reg- include the OSHA standards regulating exposures ulatory hearings have provided estimates of the to vinyl chloride and cotton dust. costs of complying with many of the proposals offered by the agency. The quality of these esti- One other area in which Government regula- mates has been quite variable and has usually been tion may have had an adverse impact is in increas- colored by the controversies surrounding the par- ing the burden of paperwork on businesses. Con- ticular regulatory proposal. Because they have cern about the burden of Federal forms and other been especially subject to the difficulties discussed record keeping was important in congressional above, OTA has not attempted to summarize consideration and enactment of the Paperwork these estimates. Reduction Act of 1980 (Public Law 96-511). But OSHA record keeping is only a very small part Impacts on Productivity and Paperwork Burdens of the total record-keeping “burden. ” For the year ending in September 1983, the time spent in keep- OSHA may also have an impact on economic ing OSHA-required records is estimated to be 36.9 productivity by diverting resources from “produc- million hours. This amounts to slightly less than tive” uses to compliance with health and safety 2 percent of the total for all Federal information regulations. Unfortunately, available information collection activities (171,358). on economic productivity generally does not include the benefits of improved worker health and safety. But even without considering those off- OSHA'S Impact on Injury Rates setting factors, the negative effects of OSHA regulation on traditional measures of productivity During congressional debates concerning the appear to be small. OSH Act, one sponsor of the act expressed the hope that the creation of a Federal regulatory Denisen (137) has studied the determinants of agency would lead to a 50 percent decline in in- U.S. economic growth and has attempted to ex- jury rates (459). Have OSHA standards and in- plain the sources of declines during the 1970s in spections reduced occupational injury rates? As the economic growth rate. He estimated the incre- discussed in chapter 3, analysis of injury rate mental costs of occupational safety and health reg- trends since the creation of OSHA is difficult, but ulation in three areas: costs of safety equipment it is hard to find any large changes in national, on motor vehicles used by businesses, costs of aggregated injury rates that do not appear to be mine safety regulation, and the costs of OSHA associated with the effects of the business cycle. regulation. He then estimated the percent of economic inputs (land, labor, and capital) that had A number of researchers have conducted sta- been “diverted” to provide for health and safety tistical analyses of the possible effectiveness of protection. In 1975, the latest year of his esti- OSHA in reducing occupational injury rates. Be- mates, business vehicle safety equipment had cause of their greater detail, these studies can pro- diverted 0.09 percent of inputs, mine safety reg- vide more information than a simple analysis of ulation had diverted 0.24 percent, and OSHA reg- trends. Two basic approaches have been used. ulation had diverted 0.09 percent, compared with The first is to develop a statistical model to “ex- a 1967 baseline. All three together, then, diverted plain” injury rates and changes in those rates. 0.42 percent of inputs. Thus, if occupational safe- Such an explanatory model can include a variable ty and health regulation and other economic in- that measures the activity of OSHA (usually the puts had been the same in 1975 as they were in number of inspections per year). The analysis can 1967, net output would have been 0.42 percent examine whether changes in injury rates correlate larger than what was actually produced. with OSHA activity. Another use of this approach Ch. 13—Assessment of OSHA and NIOSH Activities ● 265 — — is to extrapolate results from before OSHA ex- by the current scheme than others are, and that isted to predict injury rates. If the actual rates are other types (e. g., musculoskeletal problems such below those predicted, OSHA may have been the as back injuries) are not currently addressed at cause. all. Analyzing data that aggregates all injury types together may mask a favorable OSHA effect on The second major approach is to compare the some types of injuries. injury rate experience of plants that have-been inspected with those that have not. This approach Fourth, OSHA’s shift from conducting almost encounters difficulties if OSHA tends to inspect entirely safety inspections toward conducting plants with the highest injury rates. A variation more health inspections may change the expected is to compare the injury records of plants in- OSHA effect on injury rates. Fifth, the effective- spected “early” in the year with those inspected ness of Federal OSHA and the various State pro- “late” in the year. Because the data are collected grams may differ. Analysis of a data set that com- annually, the records of “early inspectees” will bines safety and health inspections and that probably reflect whatever effect OSHA has had groups OSHA activity with State program activ- during the year, while the records for the “late ity may not detect the positive effects or either inspectees” will reflect plant experience before the OSHA or individual State programs. Sixth, the inspection. Any “OSHA effect” will be seen the effectiveness studies that have been done have following year. Although this approach has the used data from employer-maintained records. If advantage of being able to estimate effects on in- there are biases in those data, or if employers keep spected plants, it will not be able to detect any better records after an inspection than they did deterrent effect of OSHA in plants that have not before inspections, the effects of OSHA may be been inspected. difficult to evaluate. In both approaches, researchers must try to cor- To date, research results on this question are rect for other factors that may influence the num- mixed. DiPietro (141) found that inspected firms ber of injuries and injury rates. Some of these fac- had higher injury rates, a result that DiPietro at- tors include the kind of industry, the effects of tributed to a tendency for OSHA to inspect more the business cycle, the pace of production and hazardous plants. Smith (446) used a model to overtime worked, the demographics of the work predict injury rates in several high-hazard indus- force (including age, experience, and family in- tries targeted by OSHA. He found declines over come), changes in the administration of workers’ time, but these were not statistically significant. compensation, changes in the practice of medi- In a later study (447), Smith compared early and cine, and variations in the mix of industries and late inspections, finding a significant decline in occupations. 1973 and an insignificant decline in 1974. McCaf- frey (291) used the same methodology and found In addition, several other factors make the task no significant OSHA effects in 1976, 1977, or of discovering an “OSHA effect” difficult. The 1978. Cooke and Gautschi (121) compared early first concerns the low probabilities of inspection, and late inspections in Maine for 1970-76 and the low average OSHA penalties, as well as varia- found a statistically significant reduction in the tions in inspection probabilities and penalties number of lost workdays per worker. among industries and geographic areas. Second, many OSHA inspections do not find any viola- Mendeloff (300) was able to disaggregate injury tions or any “serious” violations. Research that data for California by type of injury. Using a groups inspections without violations with inspec- model to predict injury rates in the absence of tions that found violations may mask OSHA’s ef- OSHA, he found significant decreases in several fect on employers who violate OSHA standards. types of injuries he judged more likely to be Third, not all types of injuries are equally preven- preventable by OSHA activity. This parallels re- table by current OSHA safety standards and search by the State of California that shows large OSHA’s inspection activity. Indeed, it is likely declines in amputations, explosions, and crushing that some types of injuries are better addressed injuries (668). Mendeloff found no OSHA effect, 266 . Preventing Illness and Injury in the Workplace

however, for California or the Nation using data injuries per year. They also estimated that the that aggregated all types of injuries. Smith findings of 5 to 16 percent reductions in injury rates in inspected workplaces imply a re- Two studies have used variables concerning the duction of 144,000 to 450,000 injuries in a year level of OSHA inspection activity. Viscusi (658), with 180,000 inspections, using data that did not distinguish between Fed- eral and State OSHA programs, found no signif- Assigning a dollar value to these reductions is icant effect related to the level of Federal activ- very difficult (see ch. 14). Using one estimate of ity. His results did reveal, however, a significant the minimum social losses due to disabling in- decline in injury rates over time, which may have juries, Green and Waitzman (195) estimate that been a statistical artifact or the result of a the benefits of OSHA safety regulation are, using favorable OSHA effect. Bartel and Thomas (45) Mendeloff’s results, at least $408 million to $610 limited their study to States covered by Federal million annually. Smith’s results for lost-workday OSHA, and found that OSHA activity had a sig- cases, they argue, imply monetary benefits of up nificant effect on employer compliance with to $4.59 billion. The reduction of 350 deaths per OSHA standards, but that this compliance led to year might be valued at over $5 billion if one only a modest reduction in injury rates. “willingness to pay” estimate of the value of life is used. Taken together, these results tend to support the conclusion of chapter 3 that most of the in- The research concerning OSHA can be com- jury rate changes since 1972 have been due to the pared with research about Federal regulation of effects of the business cycle and are not related mine safety. The number of fatalities in mining to OSHA activities. (For a more detailed discus- has fallen during the last half-century. In 1926- sion of these studies, see Working Paper #l. ) 30, a total of 11,175 miners were killed on the job, while during 1971-75, the figure was 715 (536). OSHA has analyzed the effect of one OSHA standard on the reported frequency of injuries. Congressional activity concerning coal mining has included legislation in 1941, 1952, 1969, and 1977. Between 1970 and 1978, about 47 injury-producing accidents occurred each year with one type Two studies have evaluated this legislation of wheel rim used on trucks and buses. In 1980, using aggregated data over time. Lewis-Beck and OSHA issued a new regulation concerning the Alford (270) examined fatality rates from 1932 to servicing of these wheels, and since then the 1976 and concluded that Federal legislation passed agency estimates that the frequency of injuries has in 1941 and in 1969 significantly diminished the fallen by 76 percent, to about 11 per year (631). risk of fatal injuries in mining, but that the 1952 A workplace standard issued by the State of legislation had no significant impact on fatality California concerning another type of wheel has rates. Weeks and Fox (665) concluded that there also led to an injury reduction of about 80 per- had been no change in underground fatality rates cent (631). Although these declines are encourag- from 1950 to 1969, but a significant decline from ing, they represent only a very small change in 1970 to 1980. This decline in mining fatality rates the total number of work-related deaths from all may have recently reversed. Weeks and Fox found causes. Moreover, this appears to be the only case that in 1981 the fatality rate was significantly for which OSHA has reported the actual effects higher than would be expected from the rates from of any particular safety standard on injury rates. 1970 to 1980. The Mendeloff (300) and Smith (447) studies Three other studies have used more detailed in- have been used to estimate the possible benefits formation to control for nongovernmental factors in the United States of OSHA safety regulation. that influence injury rates in mining. Boden (63) Green and Waitzman (195) have calculated that found that Government inspections reduced in- the 5 percent reduction in fatalities and a 2 to 3 juries and fatalities for the period 1973-75. Con- percent reduction in lost-workday cases found by nerton (114), using data from 1965 to 1975, found Mendeloff translate into a nationwide reduction that the 1969 Coal Act had reduced fatality rates, of 350 deaths and prevention of 40,000 to 60,000 but that nonfatal injury rates had stayed the same Ch. 13—Assessment of OSHA and NIOSH Activities ● 267 or increased slightly. Neumann and Nelson (327) detected by an OSHA inspection on the day be- concluded that the 1969 act had significantly re- fore the accident. Based on the panel’s evaluation, duced fatality rates (they estimated that in 1976 Mendeloff concluded that only 55 percent of the it had been lowered by 9 percent), but that nonserious violations would have been detected. fatal accident rates had risen. It is not completely clear why fatality rates have fallen while nonfatal The California Department of Industrial Rela- injury rates have increased since the 1969 act. Part tions has analyzed occupational injuries and of the reason may be improved reporting of non- reports that between 30 and 50 percent of the in- fatal injuries (536). juries examined in several different industries could have been prevented by compliance with Several reasons have been suggested to explain standards. Nearly all of the remaining injuries, why mine safety regulation appears to have been they concluded, could be prevented by improved more effective than OSHA’s safety regulation in training and education (84, 85, 86, 87, 88, 89). nonmining industries. First, the funding of the Bacow (37) concludes that most injuries cannot Government agencies has been greater for mine be prevented by OSHA activity. To reach this he safety, about $150 per worker, as opposed to $3 relies on studies of “unsafe acts” and “unsafe con- per worker in general industry. Second, inspec- ditions” (summarized in ch. 4) and two studies tion coverage is much greater. Every coal mine in Wisconsin and New York that found violations is inspected every year—at least twice a year for of standards to be related to, at most, 30 percent surface operations and at least four times a year of nonserious accidents and 57 percent of fatali- for underground operations. Third, the hazards ties. A Federal Interagency Task Force on OSHA of mining are more similar from establishment to concluded that only about 25 percent of work- establishment than for the wide variety of indus- place fatalities could be prevented by current tries covered by OSHA. Thus, standards and in- OSHA standards (228). spections can be more narrowly focused. Finally, mine safety regulations require mandatory safety However, the information contained in inspec- training of all miners, which includes both train- tion reports can be limited, especially for deter- ing for new employees and refresher training for mining injury causes and in designing preventive current employees (530). measures. Moreover, often these studies do not Some critics believe that safety regulation itself consider how changed standards could prevent represents the wrong approach to the problem of injuries. Many of these studies either explicitly or workplace injuries; they suggest injury taxes, implicitly rely on the belief that most accidents workers’ compensation, and tort liability as pre- are due to “unsafe acts” by workers. But, as ferred alternatives (see discussion in ch. 15 and pointed out in chapter 4, oftentimes changes in 16). Moreover, some have criticized OSHA’s safe- equipment and workplace design can prevent t y standards because they mostly concern equip- these “unsafe acts” from occurring or reduce the ment and not worker activities. OSHA’s safety potential for injury from “human errors.” Simi- inspections can only detect relatively permanent larly, new standards concerning design and instal- features of the workplace and not the transient lation of equipment might increase the propor- hazards that lead to many injuries. Thus, these tion of injuries that are deemed “preventable. ” critics claim, it is not surprising that the studies on OSHA’s effectiveness have found only limited In fact, several of these studies automatically exclude motor vehicle fatalities as “not prevent- impact or no significant impact at all. able.” Current OSHA standards do not generally For example, Mendeloff (302) found that only address motor vehicle safety, but there are tech- 13 to 19 percent of the 645 deaths reported in 1976 nologies available to reduce the incidence of in California were related to violations of safety deaths in vehicle crashes (664). These could be standards. He then asked a panel of safety engi- mandated by the Federal Government, probably neers to review the written reports of these cases through regulations issued by the National to determine if the violations would have been Highway Traffic Safety Administration. 268 ● Preventing Illness and Injury in the Workplace

Critics of safety standards frequently point to cubic centimeter to two fibers per cubic centimeter the importance of worker training for preventing and thus may be linked to the OSHA standard. injuries. Industrial leaders in occupational safety However, other factors, particularly the increase also emphasize worker training programs. Al- in the number of lawsuits concerning asbestos- though perhaps more difficult than regulating ma- related disease, may also have contributed (see chinery design, OSHA could encourage or even ch. 15). require safety training programs. Thus, new It has been estimated that the OSHA asbestos standards that improve the design of equipment, standard issued in 1972 would result in a reduc- address hazards not currently regulated, or that tion of between 630 and 2,563 deaths per year, improve training programs may have a beneficial resulting in social cost savings, in 1970 dollars, effect on injuries. of between $110 million and $652 million per year (432). These totals may underestimate the benefits Trends in Exposure Levels of the asbestos standard (433). In the case of vi- It is often asserted that exposures to toxic sub- nyl chloride, exposures declined substantially after stances have been going down over time. How- the issuance of a new, more stringent OSHA ever, data that would permit evaluation of this standard (see box N). It has been estimated that claim are scarce. Some individual firms have con- this standard would prevent about 2,000 deaths ducted industrial hygiene measurements for some over the years 1976-2000 (32). time (see 215 for several examples). But it is not Two case studies commissioned by OTA for clear, even when records from these firms have this assessment also show favorable effects after been maintained and are available, that it would the issuance of new, more stringent OSHA health be appropriate to generalize from their experience. standards (see ch. 5 for a fuller discussion). Rut- Moreover, while some exposures have declined, tenberg (413) reports that cotton dust exposures others have risen. have declined substantially in the past few years, The analysis of exposure levels over time could halving the number of workers who were exposed be used to measure the impact of OSHA. Of above the new permissible limit. Several textile course, this evaluation encounters similar, but not mills appear to be completely in compliance, while identical, problems to those found when evaluat- others fully expect to be in the near future. In ad- ing OSHA’s impact on injury rates. For hazard- dition, these changes appear to have been accom- ous exposures, however, comparatively little re- panied by or created by the installation of new search has been conducted to identify the factors technologies that both decrease employee expo- that influence exposure levels. sures and increase productivity. Mendeloff (301) analyzed information con- Goble, Hattis, et al. (184) report that in the last tained in the OSHA Management Information two years there have been large declines in both System about OSHA health inspections for asbes- employee exposures to lead in the air as well as tos, trichloroethylene, silica, and lead from 1973 in measured employee blood lead levels. For ex- to 1979. Although there are a number of limita- ample, the percentage of workers exposed above tions to those data, the data show substantial de- 200 micrograms/ma in primary smelting has declines in asbestos exposures over this period. For clined by nearly 20 percentage points. Blood lead trichloroethylene, silica, and lead, Mendeloff’s levels have declined even more dramatically. New analysis reveals no major changes over time. OSHA standards for vinyl chloride, cotton dust, and lead have clearly r-educed workplace exposures. Using OSHA inspection data, Carol Jones (237) has also found declines in asbestos exposures. She Conclusions About OSHA'S Impacts estimates that the decline in average exposures from the period 1972-76 to 1977-79 amounted to OSHA’s activities can be grouped as standard- about three fibers per cubic centimeter. This ex- setting, enforcement, and public education and posure decline is equal to the decline in 1976 of service (see ch. 12). OSHA has had the resources the permissible exposure limit from five fibers per to develop only a few new standards each year. Ch. 13—Assessment of OSHA and NIOSH Activities . 269

But many of its revised safety standards (see table larly important that the people who are always 12-4 in ch. 12) have been limited in their scope in the workplace —workers and managers—be and have not addressed major workplace hazards. fully informed about occupational hazards. In ad- There are many areas for which standards could dition, steps can be taken to provide other incen- be issued, to improve equipment design, address tives to ensure that appropriate control technol- hazards not currently covered, and establish ogy is installed (see chs. 15 and 16). workplace training programs. Moreover, many of the current safety standards most frequently Analysts can disagree on the number of cases cited by OSHA inspectors are only rarely in- of occupational disease and injury that the cur- volved in workplace fatalities (302). rent regulatory scheme may be able to prevent. But, in practice, there have been substantial Many critics of OSHA’s safety standards, how- limitations on these regulatory activities. Any ever, believe that there must still be a role for changes in this, however, would require agree- health standards (446, 685)). Indeed, analysis of ment by Congress and the executive branch to in- several of OSHA’s health standards reveals sub- crease OSHA staff and funding, as well as to ex- stantial reductions in workplace exposures, reduc- pand its ability to influence business decisions on tions that should lead to improved worker health. workplace investments and operations. However, analysis of OSHA standards and the recommendations of ACGIH and NIOSH reveals There is a general belief that the presence of that OSHA’s adoption of health standards has OSHA has increased manager and worker aware- lagged behind professional recommendations. ness of occupational health and safety. Kochan, et al. (253) report, from interviews with company Criticisms about OSHA’s health standards are and union officials, that “management has as- less likely to be about whether they are needed signed a higher priority to plant safety, the ability than about the desirable level of protection. In of the union to influence management decision- particular, employers have been concerned that making on safety issues has increased, and the role OSHA has not taken account of the costs of these of the union-management safety committees has standards (see ch. 14), or that OSHA requires the been bolstered” since the passage of the OSH Act. use of engineering controls to reduce or eliminate This increased attention has also created a need these hazards, instead of allowing the use of per- for health and safety professionals and probably sonal protective equipment (see ch. 9). Labor, on increased their role in company decisionmaking. the other hand, has criticized the slow pace of The presence of a Federal regulatory agency may standard-setting, as well as the increasing atten- lead employers to anticipate potential health and tion to the predicted costs of standards. safety problems and solve them before regulatory Finally, OSHA and the State programs have action becomes necessary. The OSH Act also cre- been able to inspect (combining health and safety ated new rights for worker information and par- inspections), at most, 4 percent of all establish- ticipation concerning health and safety. Although ments and less than 20 percent of manufacturing all these changes may be desirable, it appears that establishments each year. In addition, the penal- OSHA activity has, thus far, not had a very large ties for violations are, on average, very low, and effect on injury rates. It has had some effect on in most cases, much smaller than the potential several clearly defined health hazards, but its ef- costs of controls. Because of the low frequency fects on the many hazards it has not addressed of inspection and the low penalties, it is particu- are still in doubt.

IMPACTS OF NIOSH

OTA has divided NIOSH research activities mation (see ch. 12). Assessing the impact of into three categories: hazard identification, devel- NIOSH in these areas is even more difficult than opment of controls, and dissemination of infor- assessing that of OSHA. In theory, quantitative 270 ● Preventing Illness and Injury in the Workplace

measures of inspection activity and injury rates clear what effects this will have on the quality of can be analyzed although, of course, there are NIOSH research. great difficulties in doing so. But the impact of As in the case of OSHA, the ideal would be to an agency that conducts research is not subject know what the situation would have been in the to even these limited measures. In fact, quan- absence of NIOSH activities. Certainly some re- titative measures can be especially misleading” search and information dissemination would have when assessing research. It is not the number of taken place at universities and in the private sec- studies completed or papers published that is im- tor, even without Government activity. Although portant, but the quality of the research. it is difficult to quantify this, it is unlikely that Does this research and dissemination contrib- private sector parties by themselves would have ute to the advance of knowledge in the field of devoted the same level of resources to health and occupational health and safety? Are the epidemio- safety research as the Federal Government has logic and toxicologic studies based on well- through NIOSH. Because information is, to some designed protocols followed by comprehensive extent, a “public good, ” private parties have only and accurate data collection? Can the studies be a limited incentive to develop it on their own. reproduced? Do the NIOSH-developed sampling Once published, many can benefit, but because and analytical methods provide accurate and valid the information is already public, the original results? Do the control technologies developed or researcher would encounter great difficulty in described actually work as indicated? Is the in- charging each beneficiary for that information. formation provided by NIOSH accurate and use- Thus, it can be argued, the Government needs to ful? Are the educational programs sponsored by be involved in order to provide this “public good.” NIOSH worthwhile? Finally, do all these activi- NIOSH is the only Federal agency dedicated to ties lead to improvements in working conditions occupational health and safety research and and in the health and safety of the work force? dissemination. Although NIOSH is not the only In the late 1970s, some concern was expressed organization that conducts or sponsors epidemio- that the quality of NIOSH research was suffer- logic and toxicologic studies of workplace haz- ing (primarily following criticism by affected com- ards, its studies have advanced knowledge in this panies and industrial consultants of a NIOSH field. Controls are developed in the private sec- study of workers exposed to beryllium). When tor, but NIOSH has provided many of the detailed Donald Millar became Director, he took several sampling and analytical methods used by private steps to improve NIOSH research. One of these sector industrial hygienists. NIOSH publications was the establishment of a Board of Scientific are widely distributed and serve as an important Counselors to advise the Director on all aspects source of reference on occupational hazards and of research conducted by NIOSH. The Board con- controls. Today, most newly graduated occupa- sists of 10 scientists who are knowledgeable about tional health and safety professionals in the United occupational safety and health research. The States are educated in programs that receive fund- board has only recently been appointed and held ing from NIOSH, and many come from the pro- its first meeting in early 1984. Although the ap- grams at the institutions that have been designated pointment of a board is a concrete step, it is not as Educational Resource Centers.

SUMMARY

An OTA comparison of the standards and rec- Threshold Limit Values published by ACGIH in ommendations from OSHA, NIOSH, and ACGIH 1968. ACGIH has increased the number of sub- reveals that OSHA has tended to lag behind the stances on its list to over 600 and has revised the recommendations of both NIOSH and ACGIH. TLVs for many substances from the 1968 list. OSHA’s startup standards included nearly 400 Since 1971, NIOSH has formally transmitted rec- Ch. 13—Assessment of OSHA and NIOSH Activities . 271 ommendations concerning over 160 different sub- significant effects. The limited research on ex- stances, classes of substances, or hazardous work- posure levels appears to show positive effects for ing conditions to OSHA, After adopting the initial hazards that were the subjects of new or revised group of startup standards, OSHA has issued new OSHA regulations during the 1970s—vinyl chlo- or revised Permissible Exposure Limits and other ride, asbestos, cotton dust, and lead. requirements for 10 substances, and work prac- Because NIOSH’s major activity is research, its tice, monitoring, and personal protection require- impacts are even more difficult to quantify. The ments for 14 other substances and one physical quality of NIOSH research has been criticized al- agent. A detailed numerical comparison for a though NIOSH has recently taken steps for im- group of 123 substances shows that, overall, provement. A number of other aspects of its oper- OSHA standards are less stringent than NIOSH ations have been criticized as well. However, and ACGIH recommendations. many occupational health and safety professionals The impacts of OSHA and NIOSH are hard to have graduated from NIOSH-sponsored training evaluate. Accurate estimation of the costs of programs. NIOSH has also been an important OSHA regulation is difficult for a number of rea- source for the dissemination of information in this sons. The most comprehensive cost estimates de- field. rive from a survey conducted by McGraw-Hill. One final impact of both OSHA and NIOSH According to this survey, the share of capital is that their presence has served to increase the spending devoted to employee health and safety attention given to occupational health and safety* has changed little in the last 5 years, remaining by workers, employers, and health and safety pro- at a percentage substantially below the levels of fessionals. This increased attention facilitates the the 1970s. identification of hazards and the development of Assessing OSHA’s impacts on injury rates and controls. Thus, indirectly, OSHA and NIOSH exposure levels is also difficult. The research on activities have spurred improvements in worker OSHA effects on injury rates divides into two health and safety, although these effects probably groups—studies that find a statistically significant, cannot be quantified. but small effect, and those that do not find any 14. Decisionmaking for Occupational Safety and Health: The Uses and Limits of Analysis Contents

Page General Issues in Decisionmaking ...... 275 Private Decisionmaking ...... 276 Public Decisionmaking...... 276 Decision Tools and Rules ...... 277 OSHA Decisionmaking on Standards ...... 278 Principles Embodied in the OSH Act ...... 278 Regulatory Relief...... 279 OSHA Standard-Setting...... 2$1 Current OSHA Criteria ...... , ...... Effects of Decision Rules ...... 289 The Uses and Limits of Economic Analysis ...... Value of Economic Analysis ...... 2$9 Difficulties in Implementation ...... 2$9 Ethical Considerations ...... 292 summary ...... ●...... 293 TABLE Table No. Page 14-1. Willingness-to-Pay Estimates of the Value of Life ...... , 291 14 ■ Decisionmaking for Occupational Safety and Health: The Uses and Limits of Analysis

A number of different criteria are used for judg- This chapter discusses some general issues con- ing the desirability of changes that might improve cerning decisionmaking for occupational safety workers’ health and safety. Some have proposed and health, the history of the use of economic that greater reliance be placed on the use of cost- analysis by the Occupational Safety and Health benefit analysis, cost-effectiveness analysis, and Administration (OSHA), and the merits and other techniques of economic analysis for deci- limitations of the techniques of formal analysis. sions concerning workplace health and safety.

GENERAL ISSUES IN DECISIONMAKING

The process of improving occupational safety hazards and the best means to reduce or elimi- and health involves the identification of hazards, nate them. Labor, on the one hand, believes that the development of control techniques, and the employers have often not done enough to reduce decision to control. Choices need to be made con- or eliminate workplace hazards and desires a cerning products and workplace design by many strong governmental presence in setting and en- actors—the engineers in charge of design, the forcing standards. Employers, on the other hand, managers who make decisions concerning produc- are often concerned about investing money in un- tion, and public officials with responsibilities for necessary controls, believing either that their protecting the health and safety of workers and workplaces are not hazardous or that effective, the public. but less costly control methods are available. They also often contend that they, as employers, should The decisions we need to make. . . turn on the be permitted to decide on appropriate control question: Which hazards are acceptable, which not, in what amounts, and why?, . . And who methods, without the involvement of Government should be empowered to make this decision? (281) agencies or labor unions. In many cases, the answers to these questions Some disputes about the Government’s role in are not simple. Oftentimes facts are uncertain and occupational health and safety concern very tech- in dispute. Sometimes, values and ideals conflict nical questions about the application of the prin- with each other. To resolve such disputes, soci- ciples of industrial hygiene and safety engineer- eties rely on various institutional arrangements. ing; some are debates at the very frontiers of In the field of occupational health and safety in scientific knowledge on the mechanisms of tox- the United States, the principal institutions are icity. But beyond these disputes are more general Congress, the courts, the Occupational Safety and debates concerning the criteria on which decisions Health Review Commission, and OSHA itself. will be made and who shall be empowered to make them. The proposed resolutions of these This issue of decisionmaking—the question of issues are based on interpretations of technical in- who is authorized to make decisions and on what formation about risk as well as the moral and ethi- basis—is important because the participants can cal values, the political beliefs, and the immedi- differ greatly on the nature of occupational ate economic interests of the various parties

275 276 ● Preventing Illness and Injury In the Workplace involved. In fact, one prominant feature of be as profitable as possible, it will not take ac- disputes on occupational health and safety is that tions to protect employee health and safety solely debates over technical questions are often com- because health and safety are important goals, but bined with discussions of more general issues of only when there are financial benefits to the firm. policy, ethics, and social values. If the firm does not follow this rule, it may find itself spending money to improve health and safe- To develop and present the information needed ty without receiving any corresponding financial for a fully informed decision on workplace haz- benefit. Competitors that do not also make such ards, various observers have suggested different improvements will be able to produce the same kinds of formal analysis. The techniques are, in products at lower cost and then increase sales many cases, connected with proposals to specify and/or profits at the expense of the firm that in- the criteria on which decisions must be based. vested in health and safety. These criteria range from a vague injunction to consider “all relevant effects” to requirements that Health and safety professionals have often cited controls can be required only when “the benefits the slogan “Safety Pays” as a justification for im- exceed the costs. ” proving occupational health and safety. Although in many cases this is true, in other cases it is not. Private Decisionmaking Moreover, even when it does pay, it generally pays only up to a point. On what basis do employers make decisions concerning worker health and safety? What deci- Therefore some businesses will not always take sion rules will they follow? Some employers, of voluntary actions because they will not reap any course, take actions to protect the health and advantage over their competitors. One benefit of safety of their employees either out of altruism Government regulation is that it puts all firms on or enlightened self-interest. But in general deci- an equal footing. A company can undertake in- sions made by employers follow the dictates of vestments in employee health and safety without the competitive market system in which they fear that it will lose money to competing firms operate. that do not do so. Expressed differently, Govern- ment regulations may require certain measures Economists have developed models to explain beyond those that “pay” from the point of view this market-oriented behavior. The simplest and of the individual firm (245). most commonly used model seeks to explain the actions of firms in terms of profit-maximization This is not to say that all employers will ignore and cost-minimization. If this is a company’s goal, occupational safety and health. Indeed, there are it will take an action only if the expected reve- numerous examples of employers and their pro- nue is at least equal to the costs of that action. fessional staffs who have taken extensive actions Using the technical language of economics, this to improve worker health and safety, even with- means that the firm will produce only up to the out pressure from a Government agency. This point at which the marginal revenue from the last could be because of the commitment of profes- additional product unit is equal to the marginal sionals and companies to goals of ethical behavior costs of that unit. or corporate altruism; a decision by the firm to pursue goals other than maximizing profits; or the Applied to investments and expenditures for oc- belief by company officials that such investments, cupational safety and health, this model predicts although not profitable in the short term, will that to prevent an occupational injury or illness, ultimately be to the long-term benefit of the firm. employers will spend only as much as they can (See additional discussion on voluntary activities expect to gain in terms of reduced workers’ com- in ch. 15. ) pensation costs, improved employee productivity, reduced “down time, ” etc. Thus, any investment Public Decisionmaking for health and safety must be justified in terms of the short-term and long-term financial bene- During the congressional debates concerning fits to the firm. If the company is attempting to the Occupational Safety and Health (OSH) Act,

Ch. 14—Decisionmaking for Occupation/ Safety and Health: The Uses and Limits of Analysis 277

vious assessments concerning medical technology, OTA has considered cost-effectiveness analysis and cost-benefit analysis as parts of a family of related techniques (539). Both are designed to . compare the costs and effects of projects or alternative projects. The principal difference between them is that in a cost-benefit analysis, both costs and benefits are expressed by the same measure, which is nearly always monetary. In cost-effectiveness analysis, on the other hand, costs and benefits are expressed by different measures. Costs are usually expressed in dollars, but benefits or effectiveness are ordinarily expressed in terms such as “years of life saved, ” “days of morbidity Photo credit Department of Labor, Historical Office or disability avoided, ” or other relevant measures ● Rescue workers attempt to extricate a trapped worker (539). Cost-benefit techniques became widely used in many references were made both to the rights of the United States in the 1930s and 1940s for the American workers to safe and healthful working evaluation of public works projects (in particu- conditions and to the high cost of work-related lar, for analyzing investments of public capital in illnesses and injuries. Congress concluded that too projects for irrigation, hydroelectric power, and many illnesses and injuries were occurring and flood control). The Flood Control Act of 1936 ex- that the efforts being made and the institutions plicitly called for a cost-benefit decision rule by of that day were not sufficient to achieve the goal permitting the Government to finance a water of preventing disease and injury in the workplace. project only when “the benefits to whomsoever they may accrue are in excess of the estimated Broadly speaking, two different types of rea- costs” (280). In the 1960s and 1970s, the tech- soning have been suggested to justify Government niques of cost-benefit and cost-efectiveness anal- intervention. The first is based on ethical argu- ysis were applied in formal budget planning sys- ments, MacCarthy (277) describes four different tems in the Federal Government. To some extent, ethical justifications for Government action that these techniques have also been applied to ques- are based on utilitarianism, workers’ rights, dis- tions of health policy (539). tributive justice, and public values. Others argue that Government intervention must be justified Analysis can be a tool to assist in the decision- by economic criteria. In particular, the Govern- making process or it can be used as a formal deci- ment should intervene only in cases of “market sion rule. As a tool, the collection of information failure” and only after balancing costs and bene- and its analysis can assist policymakers to reach fits of its decisions (425,446,463,685). Three sound, well-informed, reasoned decisions about sources of “market failure” are particularly im- the management of workplace hazards. There is portant for worker health and safety: inadequate widespread agreement, at least in principle, that information, lack of labor mobility and unequal decisionmakers need to have some minimal under- bargaining power, and the presence of exter- standing of the important features of the prob- nalized costs. lem to be addressed, the factors involved in the decision, and the implications of various courses of action. There are, of course, disagreements con- Decision Tools and Rules cerning the application of this principle. In particular, there are often disputes about how much The techniques of cost-benefit and cost- information needs to be collected and to what ex- effectiveness analysis have been developed to tent policymakers should be allowed to act on the assist private and public decisionmakers. In pre- basis of uncertain and incomplete information. 278 ● Preventing Illness and Injury in the Workplace

As a decision rule, however, formal analysis could only act on the basis of the quantified com- is considerably more controversial. A decision parison of costs and benefits. rule specifies the criteria on which decisions must be based and usually requires that certain find- As a decision-assisting tool, cost-effectiveness ings be made before action is permitted. For in- analysis can be used to describe many of the con- stance, a cost-benefit decision rule would be to sequences of an action. But because the costs (in “select the alternative that produces the greatest dollars) and the benefits (e.g., in lives saved) are net benefit” (463). left in incomparable units, cost-effectiveness analysis will provide a clear decision rule only in two As a tool, a cost-benefit analysis of a proposed circumstances. If a health goal is specified, then action can be provided to a decisionmaker, who a decision rule based on cost-effectiveness analy- can still decide to undertake the proposed action sis will mandate the selection of the least costly even if the analysis shows that the quantified costs alternative that achieves that goal. Alternatively, exceed the quantified benefits. The decision could if a budget or expenditure is fixed, then a cost- reflect a concern for the distributional conse- effectiveness rule will require selecting the alter- quences of a failure to act or consideration of native with the greatest health benefit. some important benefits of the action that could not be quantified. Thus a cost-benefit analysis In occupational health and safety, two other provides information, but the decisionmaker still types of analysis have become important—risk has the flexibility to act even when the “bottom assessment and feasibility analysis. What these im- line” of the analysis says, “Don’t act.” If a for- ply for OSHA, however, has been the subject of mal cost-benefit decision rule were in effect, the considerable dispute. The meaning of these terms decisionmaker would not be able to include those and requirements concerning their use by OSHA other considerations in the final decision, but have evolved over the last 14 years.

OSHA DECISIONMAKING ON STANDARDS

Principles Embodied in the OSH Act agency to set and enforce mandatory occupational health and safety standards. Section 6 laid forth When the OSH Act was passed in 1970, Con- the procedures for adopting these standards: pub- gress clearly decided to involve the Federal Gov- lic notice of proposed actions, an opportunity for ernment in research and regulatory activity. In public comment, the conduct of a public hearing particular, Congress mandated the adoption of if requested, and a final decision based on the evi- health and safety standards and set forth a mech- dence presented to the agency. (Further details anism for writing new standards. To some extent, about these procedures are given in ch. 12. ) the law removed decisions concerning health and safety from the competitive marketplace by limit- Two subsections of the Act define health and ing employer discretion. Congress, however, was safety standards and specify the criteria on which less clear about the precise decision rules that standards are to be based. Section 3(8) of the Act OSHA would have to follow when setting these provides: standards. The term “occupational safety and health At the beginning of the OSH Act, Congress standard” means a standard which requires con- declared a purpose for Government activity in the ditions, or the adoption or use of one or more occupational health and safety field: “to assure practices, means, methods, operations, or proc- so far as possible every working man and woman esses, reasonably necessary or appropriate to pro- in the Nation safe and healthful working condi- vide safe or healthful employment and places of tions.” The act empowered the newly created employment. .

Ch. 14–Decisionmaking for Occupational Safety and Health: The Uses and Limits of Analysis .279

Section 6(b)(5) specifies the criteria for standards There is no evidence that cost-benefit analysis concerning toxic materials or harmful physical was ever explicitly proposed during those debates agents: as a decision rule for the new agency. Congress had previously issued laws that did contain cost- The Secretary . . . shall set the standard which benefit decision rules, starting with the flood con- most adequately assures, to the extent feasible, on the basis of the best available evidence, that trol legislation of the 1930s. The nonuse of such no employee will suffer material impairment of a rule in the OSH Act, if not its explicit consider- health or functional capacity even if such employ- ation and rejection, was important in the Supreme ee has regular exposure to the hazard . . . for the Court’s decisions concerning the decisionmaking period of his working life. . . . In addition to the authority of OSHA (discussed below). highest degree of health and safety protection for the employee, other considerations shall be the MacLean (280) suggests that one possible rea- latest available scientific data in the field, the fea- son that cost-benefit analysis did not enter the sibility of the standards, and experience gained debates is that it simply did not occur to Congress under this and other health and safety laws (em- that such an analysis could be applied to balance phasis added). health risks against economic costs. At that time— 1970—the theory and techniques of cost-benefit The extent to which the goals of environmental, consumer, and worker protection are to be bal- analysis, which had been used to evaluate public works projects, were still being developed for ap- anced against the costs of protection has defined much of the debate on regulatory policy during plication to health and safety issues. the last decade. Discussions of the standard-setting Because of concern about the costs of the OSH authority of OSHA and the proper role for cost Act, language was added to require that OSHA considerations have focused on these two sub- take into account the “feasibility” of its standards. sections. But the final bill also included the goal that “no employee will suffer material impairment. ” In the The legislative history of the OSH Act that is floor debate, Senator Ray Yarborough (D-TX) ex- relevant to this question is, in the words of the pressed the sentiment behind this language: “We Supreme Court, “concealedly not crystal clear. ” In particular, it does not explain what the Con- are talking about people’s lives, not the indifference of some cost accountants. ” As commen- gress meant by “feasibility” or how OSHA was to balance economic and technological considera- tators have noted, since the passage of the OSH tions in setting standards. Neither the original Sen- Act “the degree to which the complete protection ate bill nor the House version included specific of safety and health should be compromised by provisions about regulating toxic substances; in the technological difficulty and economic cost of achieving that protection has been an issue of con- both, the section was added in committee. stant controversy” (51). The criterion of “feasibility” was first proposed by Senator Jacob Javits (R-NY), who was concerned that the other bills then under considera- Regulatory Relief tion “might be interpreted to require absolute In the early to mid-1970s, policymakers began health and safety in all cases.” The final amendments that resulted in section 6(b)(5) were added to face the simultaneous problems of price inflation and unemployment. Added to these was the on the Senate floor. They reflect the objections growing perception, which was not always based to that section from several Senators. Senator on empirical analysis, that the Government’s Peter Dominick (R-CO) argued, in particular, that health, safety, and environmental regulations “[i]t is unrealistic to attempt. . . to establish a were at least partially to blame for these economic utopia free from any hazards. ” After some dis- ills. (For discussions of more general issues in Fed- cussion, a compromise was reached (280). The language adopted by the Senate was later accepted eral regulation and regulatory reform, see 3,72, 133. ) by the House-Senate Conference Committee and incorporated in the final bill signed by President A series of Executive Orders and other proce- Nixon. dural requirements reflected this new perception 280 ● Preventing Illness and Injury in the Workplace and placed new requirements on the regulatory latory agencies, and to participate in agency agencies. In 1971, the Office of Management and rulemaking proceedings “in order to present its Budget (OMB) issued a memorandum to all heads views as to the inflationary impact that might re- of departments and agencies to improve inter- sult” (659). agency coordination concerning standards and The text of Executive Order 11821 did not spe- guidelines for environmental quality, consumer cifically refer to cost-benefit analysis. However, protection, and occupational and public health in its review of regulatory activities, CWPS and safety. Known as the Quality of Life Review defined as “inflationary” those regulations for Program, it covered, in theory, any agency ac- which the costs exceeded the benefits (304). tion that would significantly affect other agencies, impose costs on “non-Federal sectors, ” or increase Just before leaving office, President Ford ex- the need for Federal funds. Agencies were ordered tended his Executive Order, without any signifi- to submit to OMB schedules of future activities cant change in substance, with E.O. 11949, issued and prepublication copies of proposed and final on December 31, 1976. The name of the required actions. In addition, proposed and final regula- statements, however, was changed to “Economic tions were to be accompanied by a summary that Impact Statements” (177). indicated the principal objectives of the rules, President Carter issued his own Executive Or- alternatives that had been considered, “a com- der on the topic of regulatory procedures (E.O. parison of the expected benefits or accomplish- 12044, March 23, 1978; renewed by E.O. 12221, ments and the costs . . . associated with the alter- June 27, 1980), which bore the title “Improving natives considered, ” and the reasons for picking Government Regulations. ” The name of the re- the selected alternative (424). quired statements was changed to “Regulatory Although this program was theoretically appli- Analyses” and in them the agencies were to pre- cable to all regulatory agencies, in practice only sent: activities of the Environmental Protection Agency . . . were examined by OMB (158,401). The memo a succinct statement of the problem; a description of the major alternative ways of deal- had little effect on the activities of OSHA (47). ing with the problem that were considered by the President Ford issued the first formal Executive agency; an analysis of the economic consequences Order (E.O. 11821) on regulatory relief on Novem- of each of these alternatives and a detailed ex- ber 27, 1974, as part of his “Whip Inflation Now” planation of the reasons for choosing one alter- program. This order required all major rules or native over the others. regulations issued by executive branch agencies Regulatory Analyses were required for any reg- to be accompanied by “a statement which certifies ulation that would have an “annual effect on the that the inflationary impact of the proposals has economy of $100 million or more” or a “major been evaluated” (176). These were called “Infla- increase in costs or prices for individual indus- tionary Impact Statements, ” probably after the tries, levels of government or geographic regions” “Environmental Impact Statements, ” required by (98,99). the National Environmental Policy Act. The primary thrust of E.O. 12044 was to im- A Presidential “oversight process” concerning prove the content of Government regulations and Federal Government regulatory activity was also to encourage agencies to compare alternative ap- established at this time (659). A new agency, the proaches. Cost-benefit analysis was viewed as a Council on Wage and Price Stability (CWPS), was tool “to compare alternative approaches to a given established in the Executive Office of the Presi- goal; . . . not to evaluate the goal, itself” (157, dent to monitor the inflationary implications of 170). The CWPS continued to prepare analyses private sector wages and prices. CWPS was also of agency regulations. The Carter administration authorized to monitor the inflationary impact of also created two other organizations: the Regu- Federal regulatory activity, to review the Infla- latory Council to compile calendars of future reg- tionary Impact Statements prepared by the regu- ulatory actions and to encourage innovative reg- Ch. 14—Decisionmaking for Occupational Safety and Health: The Uses and Limits of Analysis ● 281 ulatory techniques, and the Regulatory Analysis “burden” of federally required paperwork. OSHA Review Group to review particularly important and other Federal agencies are required to obtain regulations (157). the approval of OMB before conducting research President Reagan issued E.O. 12291 on Febru- or issuing regulations that require the “collection ary 19, 1981, as a central component of his cam- of information .“ This includes survey questionnaires and written report forms as well as require- paign for “regulatory relief.” Once again, the anal- ments for record keeping. In addition, Congress ysis requirement was renamed, this time to has considered, but has not enacted, a number “Regulatory Impact Analysis” (382). Unlike the previous orders, however, which had only re- of bills that would require Federal agencies to con- quired the agencies to evaluate the economic im- duct risk assessments and cost-benefit analyses and that would change other regulatory pro- pacts of their decisions, E.O. 12291 set an explicit cedures. cost-benefit decision rule: “Regulatory action shall not be taken unless the potential benefits to society for the regulation outweigh the potential costs. ” OSHA Standard-Setting This Executive Order further specifies that “regulatory objectives shall be chosen to maximize the OSHA standard-setting activity has been a net benefits to society” and requires, to the ex- source of disputes between management and latent possible, that all benefits and costs be quan- bor, between advocates of stringent regulation tified in monetary terms, Thus, E.O. 12291 was and those desiring regulatory relief, and between the first to require explicitly that regulatory deci- those proposing the use of cost-benefit analysis sions be based on a comparison of quantified costs and those who reject any consideration of eco- and benefits. nomic effects. Important to understanding those disputes is the legislative history of the OSH Act, In addition, the Reagan administration cen- the attitudes of business and labor, and the at- tralized regulatory review in one agency -OMB— and required agencies to submit to OMB copies titudes of the health and safety professionals who staff OSHA and the National Institute for Occu- of proposed and final regulations and the accom- pational Safety and Health. The disputes concern- panying Regulatory Impact Analysis in advance ning OSHA standards have often been settled in of publication. Although the legal authorit for y the courts. proposing and issuing regulations still remains with the heads of the regulatory agencies, in prac- During its first decade, OSHA was criticized tice the requirement for submission to OMB has by business representatives for failing to take ac- meant that the agency must receive approval from count of the costs of complying with its stand- OMB prior to publication. A Presidential Task ards and for not ensuring that those costs bore Force on Regulatory Relief, chaired by the Vice a “reasonable relationship” to the benefits of the President, was also established to resolve disputes standards. Labor representatives, on the other between OMB and regulatory agencies. hand, criticized OSHA for including compliance costs as a factor in its decisions. Congress has also added procedural requirements for Federal regulatory agencies. In Septem- The values of OSHA personnel have been de- ber 1980, the Regulatory Flexibility Act (Public scribed as “pro-protection” and, it has been Law 96-354) was enacted. It requires agencies to argued, derive from the professional training and prepare “regulatory flexibility analyses” for reg- background of the health and safety professions ulations that would have a “significant economic and from their view of the agency’s mission. Most impact on a substantial number of small entities” OSHA staff have worked or been educated in the (defined as small businesses, small nonprofit orga- occupational health and safety professions. As one nizations, and small governmental jurisdictions). observer noted, “they believe strongly that work- In December 1980, Congress enacted the Paper- ers ought to be protected from hazards and that work Reduction Act (Public Law 96-511). In gen- larger reductions of risk are preferable to smaller eral, the purpose of this act was to reduce the ones (without much thought of cost). ” These val- 282 Preventing Illness and Injury in the Workplace

ues led the agency to tend to adopt “the more pro- concerning the guarding of mechanical power tective of alternatives presented to them by the presses (“no hands in dies”). In this case, the Third parties (from outside the agency] or by credible Circuit Court of Appeals ruled that while the OSH scientific research” (245). Act was a “technology forcing piece of legislation,” OSHA’s determination that the standard The agency did, however, recognize the need was technologically infeasible was adequately sup- to collect information concerning the technical ported. In addition, this court, following the rea- feasibility and estimated costs of its regulatory soning of the earlier asbestos decision, ruled that proposals. From as early as OSHA’s first year OSHA could consider “economic consequences” (1971), the agency has endeavored to develop this when setting standards. In particular, OSHA information using a combination of in-house staff could not “disregard the possibility of massive and outside consultants. The goal of this early economic dislocation caused by an unreasonable activity was not to perform cost-benefit analyses, standard” (1). but to provide feasibility and cost-of-compliance estimates to counter the cost estimates and claims Secondly, in an industry challenge to OSHA’s of infeasibility made by the opponents of particu- regulation of vinyl chloride, the Second Circuit lar regulations (47). Court of Appeals upheld a “technology-forcing” standard. By this decision, a standard could be The first major dispute concerning the use of considered feasible even if the technology neces- economic criteria involved a more stringent stand- sary for compliance was not already widespread ard for asbestos exposure, which was issued in in the regulated industry. All that was necessary 1972. At that time OSHA lowered the permissi- was that the technology was “looming on today’s ble exposure limit for asbestos from 12 fibers per horizon” and could be brought into widespread cubic centimeter to 5, with a further lowering to use. A standard would be considered technologi- 2 fibers per cubic centimeter by 1976. The delayed cally infeasible only if meeting the standard was effective date for the 2-fiber limit was designed shown to be “clearly impossible” (454a). “to allow employers to make the needed changes the Secretary is not restricted by the status for coming into compliance.” OSHA was sued by . . . quo. He may raise standards which require. . . the Industrial Union Department of the AFL-CIO improvements in existing technologies or which for, among other issues, having considered eco- require the development of new technology. nomic factors in setting this limit. The Industrial Union Department argued that the phrase “to the At about the same time as these judicial deci- extent feasible” in section 6(b)(5) should be inter- sions, President Ford issued the first Executive Or- preted to mean only technological feasibility— der requiring inflationary impact statements. In i.e. whether or not the technology to control ex- following years, OSHA came under pressure from posures was available. the Council on Wage and Price Stability to base its decisions on the results of cost-benefit analy- The D.C. Court of Appeals ruled in this case sis. A proposed standard concerning coke oven that OSHA could take account of the costs of emissions was the basis for the first clash between complying with a new standard. Thus, in this OSHA and CWPS. CWPS participated in the decision, “feasibility” under the OSH Act was rulemaking proceeding and argued that OSHA defined to include both technology and econom- had overstated the expected benefits of the pro- ics. According to the court, a standard would be posed standard. Using its own estimates of the ex- considered economically feasible if compliance pected number of lives saved and two estimates with it would not threaten the viability of an in- of the “value of life” from the economics literature, dustry as a whole, even if individual firms might the Council suggested that the proposed stand- close because they could not meet the standard ard was not worthwhile. Finally, CWPS recom- (223). mended that OSHA consider allowing the use of respirators to comply with the standard (290). Two subsequent decisions refined this two- pronged definition of feasibility. The AFL-CIO Between the time of proposal and of final pro- challenged OSHA’s decision to relax a regulation mulgation, a new Assistant Secretary for OSHA, Ch. 14—Decisionmaking for Occupational Safety and Health: The Uses and Limits of Analysis ● 283

Morton Corn, was appointed. About the use of publicized case, Charles Schultze (then chairman cost-benefit analysis he has written (122): of the Council of Economic Advisers), his staff, and CWPS became involved in the cotton dust After arriving at OSHA, I engaged in an in- depth consideration of cost-benefit analysis, ap- rulemaking process in 1978. They suggested plying the methodology to the coke-oven stand- changes in OSHA’s draft standard. After an ap- ard. . . .With the dose-response data at our dis- peal to President Carter, the major requirements posal, various assumptions were used to ring in for engineering controls in the standard were not changes on different methodologies for estimat- changed, although some features of the standard ing benefits. The range in values arrived at, based were modified. The intervention by Schultze and on the different assumptions, was so wide as to CWPS, however, was viewed by many as an at- be virtually useless. The conclusion I reached after tempt to reduce the cost and the protectiveness this exercise was that the methodology of cost- of that regulation (135,290,479). benefit analysis for disease and death effects is very preliminary, and one can almost derive any In 1978, OSHA issued final standards for ben- desired answer. zene, DBCP, arsenic, cotton dust, acrylonitrile, In October 1976, when OSHA actually issued and lead. Four of these six final standards were its regulation concerning coke oven emissions, its challenged in the courts. The cases concerning the statement of reasons clearly rejected the use of benzene and cotton dust standards are particu- cost-benefit analysis. In part, this position was a larly relevant to the evolution of the use of eco- reaction to the arguments of CWPS, whose inter- nomic analysis at OSHA. vention was perceived as an attempt to reduce the level of worker protection. In the preamble to the OSHA’s more stringent standard for occupational exposure to benzene was challenged by the final regulation, OSHA based its rejection of cost- petroleum industry. The Court of Appeals for the benefit analysis on the difficulties of accurately Fifth Circuit ruled in this case that the phrase “rea- estimating the expected benefits of the new stand- sonably necessary or appropriate” contained in ard and the lack of “an adequate methodology the definitional section of the act (section 3(8)) to quantify the value of a life” (620). meant that OSHA could issue a more stringent This attitude toward cost-benefit analysis con- regulation only if it estimated the risks addressed tinued during the Carter administration. In 1977, by the standard and determined that the benefits Eula Bingham, as Assistant Secretary for OSHA, of the standard bore a “reasonable relationship” expressed concern about proposed procedures to the costs. This ruling, in effect, erected a cost- concerning economic impact analysis (56): benefit decision rule for the agency to follow. The court invalidated the standard because it con- While one can argue over the specific role of economics in establishing regulations, it is clear cluded that there was insufficient evidence that to me that economics should not be a paramount this more stringent standard would have any “dis- consideration in setting safety and health stand- cernible benefits” (13). ards. The overriding purpose of the OSH Act is to protect workers, tempered by considerations This decision was appealed to the Supreme of feasibility, Accordingly, I would agree that Court, which in the summer of 1980 upheld the some economic impact analysis should be per- lower court’s decision to vacate the standard, al- formed to provide a basis for evaluating indus- though it did not follow the same reasoning. In try representations as to economic impacts and fact, while the Court voted 5 to 4 to strike down possibly to influence the length of the compliance the standard, the majority could not agree on a period allowed for a given standard. I do not common set of reasons. Five separate opinions believe that policy decisions impacting worker were issued by the Court, but no single opinion safety and health can or should be subject to a had the support of more than four justices. Jus- formalized benefit-cost test. tice Stevens presented the views of four of the CWPS continued to participate in OSHA’s justices who had voted to strike down the stand- standard-setting proceedings and generally was ard. In that opinion, the issue of whether the OSH very critical of OSHA’s proposals. In a widely Act required the agency to follow a cost-benefit z&f . Preventing Illness and Injury in the Workplace

rule was not addressed. Instead, this plurality gation to find that a significant risk is present declared that the agency had not made a “thresh- before it can characterize a place of employment old finding” that risk presented by benzene ex- as “unsafe. ” posure was “significant” (224): In a footnote, the Court noted that the ultimate By empowering the Secretary [of Labor] to pro- decisions of the Agency concerning the acceptable mulgate standards that are “reasonably necessary level of risk “must necessarily be based on con- or appropriate to provide safe or healthful em- siderations of policy as well as empirically veri- ployment and places of employment,” the Act im- fiable facts” (224). plies that, before promulgating any standards, the Secretary must make a finding that the work- Justice Marshall filed a dissenting opinion for places in question are not safe. But “safe” is not the four Justices who voted to uphold OSHA’s the equivalent of “risk-free.”. . . a workplace can benzene standard. In that opinion, Marshall ar- hardly be considered “unsafe” unless it threatens gued that the plurality’s review of the record was the workers with a significant risk of harm. “extraordinarily arrogant and extraordinarily un- Therefore, before he can promulgate any per- fair” because the plurality had improperly made manent health or safety standard, the Secretary its own findings concerning factual issues and had is required to make a threshold finding that a place of employment is unsafe-in the sense that unfairly described OSHA’s analysis of these significant risks are present and can be eliminated issues. Moreover, this dissent argued that the re- or lessened by a change in practices. quirement to demonstrate a “significant risk” was “a fabrication bearing no connection with the acts After an extensive review of the record in this or intentions of Congress and is based only on case, the plurality ruled that because OSHA had the plurality’s solicitude for the welfare of regu- not made this threshold finding and because the lated industries” (224). record before the agency did not contain “substantial evidence” to support such a finding, The issue of whether the OSH Act required a OSHA had exceeded its authority in issuing the cost-benefit test in addition to a finding of signif- more stringent standard for benzene exposure. icant risk was taken up in a case concerning a The Court did not rule on the issue of whether more stringent standard for exposure to cotton the OSH Act required a cost-benefit test in addi- dust. Textile industry representatives argued that tion to this requirement to demonstrate “signifi- OSHA had exceeded its statutory authority be- cant risk.” cause it had neither conducted a cost-benefit analysis nor explictly determined that the benefits of The Court provided only limited guidance as the standard justified the costs of compliance. La- to what was meant by “significant risk, ” words bor representatives and OSHA argued that the that do not actually appear in the language of the OSH Act did not require such a cost-benefit deci- act. To quote the plurality opinion (224): sion rule. Instead, after determining that exposure First, the requirement that a “significant” risk to cotton dust presented a “significant risk” (as be identified is not a mathematical straitjacket. required by the decision in the benzene case), the It is the agency’s responsibility to determine. . . agency was required to issue the most protective what it considers to be a “significant” risk. Some standard subject only to the constraint that com- risks are plainly acceptable and others are plainly pliance with the standard be technologically and unacceptable. If, for example, the odds are one economically feasible. in a billion that a person will die from cancer by taking a drink of chlorinated water, the risk The Supreme Court affirmed the OSHA cot- clearly could not be considered significant. On the ton dust standard by a vote of 5 to 3. The major- other hand, if the odds are one in a thousand that regular inhalation of gasoline vapors that are two ity opinion held that cost-benefit analysis was not percent benzene will be fatal, a reasonable per- required by the OSH Act. Instead, Congress had son might well consider the risk significant and erected a requirement for “feasibility analysis. ” take appropriate steps to decrease or eliminate it. Citing dictionary definitions that “feasible” means Although the agency has no duty to calculate the “capable of being done, ” the Court ruled (17) that exact probability of harm, it does have an obli- section 6(b)(5) of the OSH Act Ch. 14—Decisionmaking for Occupational Safety and Health: The Uses and Limits of Analysis ● 285

... directs the Secretary to issue the standard Current OSHA Criteria that “most adequately assures. . . that no employee will suffer material impairment of health, ” lim- OSHA now uses a four-step process for mak- ited only by the extent to which this is “capable ing decisions about health standards, as expressed of being done. ” In effect. . . Congress itself by former Assistant Secretary Thorne Auchter. defined the basic relationship between costs and First, the agency determines that the hazard in benefits, by placing the “benefit” of worker health question poses a “significant risk.” Second, OSHA above all other considerations save those making determines that regulatory action can reduce this attainment of the “benefit” unachievable. Any standard based on a balancing of costs and bene- risk. Third, it sets the regulatory goal (for health fits by the Secretary that strikes a different bal- standards, this is the permissible exposure limit) ance than that struck by Congress would be in- based on reducing this risk “to the extent feasi- consistent with the command set forth in sec. ble.” Finally, OSHA conducts a cost-effectiveness 6(b)(5). analysis of various options to determine which will achieve this chosen goal in the least costly In a footnote, the Supreme Court also endorsed manner (434,638). the definition of economic feasibility that had been suggested by OSHA. According to this, to prove However, it is not clear what criteria apply to a standard economically feasible, OSHA must safety standards and other regulations issued by show “that the industry will maintain long-term OSHA. In the cotton dust decision, the Supreme profitability and competitiveness” (17). This def- Court left this issue unresolved, but did state that inition is consistent with the earlier courts of ap- it is possible that Congress could have set different peal rulings on the standards for asbestos, coke criteria for health standards than for safety stand- oven emissions, and lead (10,223,654). ards. It also noted that the “reasonably necessary or appropriate” language of section 3(8) would The legal battles concerning OSHA standards apply to safety standards (224). were not just about the details of economic analysis and risk assessment. In part, these battles took “Significant risk” on symbolic meanings. That is, people on both sides of these issues took positions based on what To determine whether a hazard poses a “sig- they believed these cases symbolized for Govern- nificant risk of material health impairment, ” ment regulatory activity. Some supported the OSHA now generally uses the techniques of quan- standards issued by OSHA because these stand- titative risk assessment. In several publications ards were believed to represent strong efforts to since the Supreme Court decisions concerning control occupational health and safety problems benzene and cotton dust, OSHA has presented after decades of neglect. They viewed cost-benefit quantitative risk assessments, These risk assess- analysis as a technique that was being introduced ments have calculated the estimated risk at the to weaken governmental protections. Others op- currently permitted exposure levels and the an- posed these standards and supported legal restric- ticipated risk at the new, lowered exposure levels, tions on OSHA, including requirements for cost- in order to show that its proposed standards are benefit analysis, because they believed that the addressing “significant risks” and that they will agency had gone too far in imposing regulatory serve to reduce the risk of occupational disease. costs on businesses.1 As discussed above, the Supreme Court provided only limited guidance on what occupational risks should be considered significant. The plurality opinion appears to have indicated that a l-in- ‘Table A-3 in appendix A presents a list of the legal cases con- 1,000 risk of death is “significant” while a one- cerning OSHA standards. For further discussion of OSHA’s legal in-one-billion risk is not. However, the example obligations concerning the development of standards, see (333,408). Mintz (307) provides excerpts from primary source documents rein this opinion refers to a one-in-one-billion risk lated to this history of standard-setting, including legal briefs and from a single drink of water and a l-in-1,000 risk court opinions. from regular inhalation of gasoline vapors. If this 286 ● Preventing Illness and Injury in the Workplace is adjusted for the total amount of water the aver- Feasibility age person consumes, the resulting risk estimates As indicated earlier, for both technological and for these hypothetical examples concerning inhala- economic feasibility the general requirement is tion of gasoline and consumption of chlorinated water are about the same. (McGarity has reached that OSHA must show that compliance with the a similar conclusion by calculating the total num- new standard is possible, that it is “capable of being done.” Technological feasibility refers to the ber of cases expected in the exposed populations availability of technologies and methods to com- for these two examples (297a).) ply with the new standard. To prove this, OSHA OSHA’s “significant risk” determinations have can, of course, refer to plants and technologies principally relied on comparing the estimated risks that already meet the new standard. But OSHA for a particular hazard to the I-in-l, (X)() guide- is not bound just to the status quo. The courts line. For additional support, OSHA has also in- have ruled that the agency has the authority to cluded comparisons with the risks of fatal occupa- require technological improvements. The agency tional injury (derived from the data of the BLS must present substantial evidence to prove that Annual Survey) and with the quantified risks of it is reasonable to expect that efforts by industry several occupational health hazards. For exam- will lead to compliance, even if the exposure re- ple, OSHA has argued that a particular level of ductions are greater than those that have previ- arsenic exposure presents a “significant risk” be- ously been achieved. In the words of one court cause the estimated death rate at that level (8 (654), deaths per 1,000 workers) is OSHA’s duty is to show that modern technol- . . . 1/4 to 1/2 the death rate [from injuries] in ogy has at least conceived some industrial strate- the riskiest occupations, 2 to 5 times higher than gies which are likely to be capable of meeting the the risks in occupations of average risk, and 10 PEL [permissible exposure limit] and which the to 100 times the risk of the low risk occupations. industries are generally capable of adopting. It is also 1/3 of the maximum permitted radia- To meet this duty, OSHA uses the information tion cancer risk but about 3 times higher than the cancer risk which 95 percent of radiation work- provided in public comments and the public hear- ers are under (638). ings. A feasibility determination usually relies on the reports and opinions of expert consultants, the Finally, OSHA has compared the estimated risks availability of control technologies, descriptions of exposure at currently permissible levels to the of plants and companies already achieving the estimated risks for exposure that were regulated new standard, and general comments submitted in previous years. For instance, in proposals con- to OSHA. cerning new standards for ethylene dibromide and asbestos, OSHA compared the estimated risks for Economic feasibility refers to the economic ca- those substances to the risks of exposures to cot- pability of the regulated industries to afford the ton dust and coke oven emissions (642,647). technologies needed for control. The Supreme Court has ruled that an analysis of economic fea- In one case, OSHA has exempted one group sibility does not mean cost-benefit analysis. of employers from the OSHA commercial diving Rather, OSHA is required to show that compli- standard because the agency determined that the ance with a standard is affordable by the regu- estimated risks of injuries for this group were lated industry us a whole. A standard can be con- below those for industries with low injury rates. sidered feasible even if it adversely affects profits In explaining its decision, OSHA relied on a cal- and causes some employers to go out of business culation showing that the injury rate for scientific and educational divers was lower than the injury rates for a number of industries, including This calculation, howwer, substantially understates the risk dur- 2 ing diving operations because none of these employees is actually banking (634). underwater for 2,000 hours per year. In fact, the actual underwater exposure time for these divers is usually less than one-tenth of the ‘This calculated injury rate was based on the number of reported time assumed by OSHA (249a). The actual risk estimates for this deaths and injuries divided by 2,OOO hours per year, the equivalent group of employees while engaged in diving would, therefore, be of full-time employment, 40 hours per week for so weeks per year. over 10 times greater than the risk shown by the OSHA calculation. Ch, 14—Decisionmaking for Occupational Safety and Health: The Uses and Limits of Analysis . 287 rather than comply with the standard. However, have improperly influenced at least two OSHA OSHA must show that the long-term profitabil- regulatory proceedings, concerning hazardous ity and competitiveness of an industry will be communication and commercial diving, after pri- maintained. vate meetings with representatives of the industries affected by the regulations (519,520). To analyze economic feasibility, OSHA generally estimates the costs of compliance for a In several cases concerning proposed standards, standard. These compliance costs are usually pre- OSHA has published alternatives suggested by sented in several ways: for example, average cost OMB. These include, for example, the 1981 proper affected firm, average cost per exposed posal for hearing conservation (637) and the 1984 worker, compliance costs as a percentage of in- proposal on grain elevators, which included dustry sales, and compliance costs as a percent- OMB-suggested alternatives (348). Two disputes age of total payroll costs. OSHA has not set forth between OMB and OSHA concerning proposed any mathematical formula for determining when standards have been appealed to the Presidential these costs would be considered economically in- Task Force on Regulatory Relief. In both cases feasible. Rather, it presents the costs, and ex- (the cotton dust and hazard communication stand- presses a judgment about whether or not they ards), OSHA was allowed to publish its proposals would impose a substantial burden or have a sig- (637). nificant impact on the market structure of the af- OMB suggestions have been incorporated in at fected industries. least two final standards. For the hearing conservation amendment, several technical requirements Cost-Effectiveness Analysis were altered (637). In another case concerning a Finally, OSHA uses the techniques of cost- final standard for ethylene oxide exposure, OMB effectiveness analysis to evaluate alternative meth- expressed “reservations” about part of the OSHA- ods of achieving the health protection goal that drafted standard (in particular, the provision es- has been selected on the basis of the risk and fea- tablishing a short-term exposure limit). In re- sibility analyses. Cost-effectiveness analysis at this sponse, OSHA removed that provision from the point is applied in a relatively narrow way. It is final, published standard, and requested addi- not used to judge the “worth” or desirability of tional public comment (649). OSHA considers the standard, but only to select among alterna- OMB review to be “akin to internal review,” after tive approaches for meeting that standard. which all final decisions are made by OSHA and the Department of Labor (637). Regulatory Impact Analyses The General Accounting Office has studied Executive Order 12291, issued in early 1981, re- agency compliance with Executive Orders 12044 quires that OSHA prepare Regulatory Impact and 12291 and has collected information on the Analyses. These consist of detailed discussions of costs of preparing economic analyses. For ON-IA, the quantified benefits, compliance costs, and eco- these analyses cost an average of $338,000 (510). nomic impacts for alternative standards consid- Although it is difficult to determine the value and ered by OSHA. effects of these analyses, the amount of resources used by OSHA to develop them has been sub- The order also requires that all OSHA regula- stantial. tory actions be reviewed by OMB, which, to the extent permitted by law, requires regulatory agencies to demonstrate that their regulations are cost- Effects of Decision Rules beneficial. Generally, the results of the OMB review and agency responses need not be made pub- Questions remain, however, concerning the ac- lic. Thus, it is difficult to determine if OSHA deci- tual application of the current OSHA criteria. sions have been altered by OMB’s cost-benefit How large does a risk need to be in order to be review (510). It has been argued that OMB and considered significant? How is risk to be meas- the Presidential Task Force on Regulatory Relief ured? How is technological feasibility determined? 288 . Preventing Illness and Injury in the Workplace

How costly can a standard be before it threatens Would OSHA’s decisions have been different the viability of an industry and thus is considered under a cost-benefit decision rule? One contract economically infeasible (277)? Indeed, there have report (262) prepared for this assessment suggests been and continue to be disputes about the ap- that for at least some of the major OSHA health plication of any set of decision criteria by a regu- standards, the use of a cost-benefit decision rule latory agency. In practice, decision criteria may would have led to less stringent standards. only define a range of allowable decisions and will often not mandate a particular result. Judith and Lester Lave (262) applied several different decision frameworks to four health stand- The positions adopted by the participants in ards issued by OSHA--those for coke oven emis- OSHA’s regulatory proceedings, general political sions, benzene, vinyl chloride, and cotton dust. considerations, and the personal judgments of They compared the regulations actually issued by OSHA’s Assistant Secretaries and its staff have OSHA using its criteria of technological and eco- been and continue to be important to OSHA’s nomic feasibility and those that the Laves believe decisionmaking. Moreover, OSHA has always would have been issued under a cost-benefit deci- balanced various factors before issuing standards. sion rule. In three of these cases-for coke oven For instance, OSHA stated in its preamble to the emissions, benzene, and vinyl chloride-a cost- vinyl chloride regulation that its “judgments have benefit analysis would have led to a less stringent required a balancing process, in which the over- regulation. For cotton dust, the conclusion of an riding consideration has been the protection of analysis is very dependent on the discount rate employees.” These more informal judgments may chosen and the value placed on preventing addi- be even more important than the formal decision tional cases of byssinosis (see below for a discus- rules used by an agency. sion of these issues). This case, they concluded, is one in which “reasonable analysts can choose In addition, the court battles about OSHA deci- values within the range accepted by the econom- sions and more general discussions of “regulatory ics profession and wind up with opposite conclu- reform” were not only about what the decision sions about the desirability of a stringent standard.” rules should be, but also about who should be empowered to interpret and apply those rules. The conclusions about the fates of the vinyl Should the Assistant Secretary for OSHA have chloride and cotton dust standards are note- the authority to make these decisions with only worthy because in both cases improvements in limited review by the courts? Or should review- productivity accompanied compliance with a ing judges become extensively involved in analyz- stringent OSHA standard. In part, those improve- ing the factual record and OSHA’s judgments? ments may have been spurred by the necessity to And to what extent should outside agencies, such comply with OSHA’s stringent requirements. If as CWPS and OMB, be involved in examining a cost-benefit decision rule had been in effect for and approving OSHA’s decisions? OSHA, these gains in employee health and in economic productivity might not have occurred. Because judgment and the identity of the deci- It cannot be conclusively shown that the ap- sionmaker are important, and because there are plication of a cost-benefit decision rule would many uncertainties in estimating effects (discussed have changed OSHA’s decisions and the nature below), the adoption of a cost-benefit rule may of technological change in affected industries. not lead to decisions that are different from those Lave and Lave had to rely on currently available adopted under a feasibility test. DeMuth (136) has information. Had a cost-benefit rule been in ef- suggested that “just as the Corps of Engineers be- fect, OSHA might have prepared additional quan- came adept at demonstrating that every dam that titative information, especially on the benefits of could be built would pay for itself, so the regula- its standards. tory agencies will learn to demonstrate, with increasing analytical verve, that every new regula- However, their conclusion does support the tion is cost-beneficial. ” concern of many participants and observers that Ch. 14—Decisionmaking for Occupational Safety and Health: The Uses and Limits of Analysis ● 289 cost-benefit analysis in practice would not be a of cost-benefit analysis will prevent regulatory neutral decision rule, but one that is biased agencies from carrying out their responsibilities against improvements in worker health and safety to protect health and safety and will lead to ex- (43,65,115,194,411). For example, Baram (43) has tensive delays in an already slow regulatory proc- concluded that in many cases “[c]ost-benefit anal- ess— in other words, to a “paralysis by analysis. ” ysis is a ‘numbers game’ that is used to oppose Boden (65) has expressed concern that under cur- regulatory actions that have been proposed to rent circumstances, the use of cost-benefit analy- protect public health and the environment .“ Con- sis “may bias political decisions against even those nerton and MacCarthy (115) believe that the use regulatory decisions that are cost-effective. ”

THE USES AND LIMITS OF ECONOMIC ANALYSIS

Value of Economic Analysis Lave and Lave (262) have suggested that the use of economic analysis “sharpens the questions, Economic analysis—including cost-benefit anal- clarifies the implications of policies, and gener- ysis, cost-effectiveness analysis, and the feasibil- ally manages to attain solutions at lower cost. ” ity analysis performed by OSHA—can provide In addition, when the necessary data are avail- decisionmakers with important information on able and when the quantification of intangible ef- the problems and alternatives they face and the fects is not a problem, cost-benefit analysis can consequences of various courses of action. In “The shed light on the implications for economic effi- Implications of Cost-Effectiveness Analysis of ciency of alternative projects. Medical Technology” (539), OTA described 10 general principles of analysis that are applicable As one element of the decisionmaking process, to the conduct of both cost-effectiveness and cost- as a decision-assisting tool, economic analysis can benefit analyses. These principles are: provide guidance and information. In addition, economic analysis can provide support for deci- ● Define problem. sions or actions that may be taken on other ● State objectives. grounds. For example, in some cases, decisions ● Identify alternatives. based on noneconomic grounds will also be sup- ● Analyze benefits and effects. ported by the results of economic analysis. In ● Analyze costs. those cases, most people would support the use ● Differentiate perspective of analysis. of economic analysis. Arguments arise when the ● Perform discounting. analysis supports less stringent standards than ● Analyze uncertainties. those chosen for other reasons. ● Address ethical issues. ● Interpret results. Difficulties in Implementation

Following these principles can lead to the devel- The limitations of these techniques are particu- opment of clear and useful analyses. But the proc- larly evident when they are considered as deci- ess of collecting information, ordering it, and sion rules. In 1980, OTA concluded that cost-ef- analyzing it can be just as important as the final fectiveness and cost-benefit analyses exhibited too results of an analysis. many methodological and other limitations to justify sole or even primary reliance on them in Performing an analysis of costs and benefits can making decisions (539). That conclusion is still be very helpful to decisionmakers because the applicable for the analysis of measures to improve process of analysis gives structure to the problem, occupational safety and health. allows an open consideration of all relevant effects of a decision, and forces the explicit treat- A number of the analytical principles-defining ment of key assumptions (539). the problem, stating objectives, identifying alter- 290 . Preventing Illness and Injury in the Workplace natives, differentiating the perspective of the anal- trols, but there is only limited information avail- ysis, analyzing uncertainties, and interpreting able on the actual workplace effectiveness of such results—are relatively uncontroversial, in large devices (ch. 8). part because they are components of any proc- After identifying and quantifying the benefits ess of rational decisionmaking. However, the to be expected from a given action, a cost-benefit quantification of benefits and costs and applica- analysis (but not a cost-effectiveness analysis) re- tion of a discount rate to them are distinctive quires that these be converted to units that can features of these types of analyses. They are also be directly compared with the costs of the action. the features that present the most difficult meth- Analysts almost invariably choose monetary units odological issues and arouse the most contro- for this. versy, especially when applied to governmental regulation of worker health and safety. There are two major approaches to placing a value or a price on human lives or lifesaving pro- Benefits Analysis grams. The first is to consider the value of a life to be the present discounted value of the person’s The benefits of various alternatives must be future income. Because this method, known as the identified, and if possible, quantified. Quantifica- human capital approach, assumes that the value tion bears the danger that the effects that can be of a person’s life is equal to their expected income, measured will receive more attention than those it implies that “women are valued less than men, that are not quantified, even if the unquantified blacks less than whites, retired people less than are believed to be more important. Lave (263) has workers, and low-paid textile workers less than called this a “Gresham’s law of decisionmaking. ” higher-paid steel workers” (115). In addition, it On this danger, Mishan has written (308): cannot include the value that other people attach . . . the outcome of all too many cost-benefit to saving a particular person’s life. studies follows that of the classic recipe for making horse and rabbit stew on a strictly 50-50 basis, The other major approach has been to evaluate one horse to one rabbit. No matter how carefully lifesaving programs on what people are “willing the scientific rabbit is chosen, the flavor of the to pay” for them. This approach, too, runs into resulting stew is sure to be swamped by the horse- problems. It is difficult to find out exactly what flesh. The horse, needless to say, represents those people are “willing to pay.” A survey could be other [unquantified] considerations . . . . used, but the interpretation of the results is difficult. The uncertainties of any quantification of benefits begins with uncertainties concerning the rela- Another measure can be obtained from analy- tionship between exposures and health hazards. sis of the additional pay that workers may receive Epidemiologic studies are often limited by small for taking unsafe jobs. (These have been termed sample sizes, a lack of information on past ex- hazard premiums or compensating wage differen- posures, and the presence of confounding vari- tials. See ch. 15 for a discussion. ) It is also possi- ables. The results of animal testing present prob- ble that consumers are willing to pay “extra” for lems in determining the applicability to human less hazardous consumer products. Some econ- populations and in extrapolating from the high omists, using the techniques of statistical analy- doses often used in such studies to the lower doses sis, have attempted to measure the extent of these found in the workplace (see ch. 3 and 542). “revealed preferences. ” These studies are subject to a number of technical problems in the meas- Even after identifying the risks involved, an urement of risk levels and in adjusting for other analysis of the benefits of controlling them must factors that influence wages and prices. The results discuss the effectiveness of the various technol- of these studies have been used to calculate the ogies that could be applied. Often, information “value of a life. ” crucial to that analysis is lacking. For example, personal protective equipment is often suggested Table 14-1 shows the wide range of implied as a cost-effective alternative to engineering con- values for human lifesaving derived from such . ——

Ch. 14—Decisionmaking for Occupational Safety and Health: The Uses and Limits of Analysis ● 291

Table 14-1 .—Willingness-to-Pay Estimates of the a ing programs. Moreover, the value of life derived Value of Life from this approach remains a function of income Value per and wealth. statistical life (thousands of Cost Analysis Method 1977 dollars)b Survey approach: The estimation of costs is often thought to pre- Acton (1973) ...... 38 sent fewer difficulties than the analysis of bene- Jones-Lee (1976) ...... 8,440 Landefeld (1979) ...... 1,200 fits. Nevertheless, the costs of controls are often Revealed preference: hard to estimate accurately. In part this is because Labor market: many control technologies involve changes in the Dillingham (1979) ...... 277 Thaler and Rosen (1975) ...... 364 actual productive process or have multiple uses. Viscusi (1978) ...... 1,650 For example, what portion of the costs of install- Smith (1976) ...... 2,045 ing duct work in a new plant should be attributed Olson (1981) ...... 5,935 Consumption activity: to the need to dilute an air contaminant? And Dardis (1980) ...... 101c what percentage should be listed as the cost of Ghosh, Lees, and Seal (1975) . . . . . 260 providing heating and air conditioning to a Blomqist (1979) ...... 342 Portney (1981) ...... 355 plant—an ordinary cost of doing business? %/here a study included a “central” or “most reasonable” estimate, that is shown, where only a range was ~iven, the lowest value is presented In addition, company officials are usually in the bvalue~ were convefled t. 1977 dollars using the U S Bureau of Labor Statistics Consumer Price Index best position to know what needs to be changed cit IS unclear from the Dardis study what year’s dollars apply, although the in their plants to comply with a proposed stand- estimate presented here appears to be based on an average value for the period 1974-79 ard. But these people also have a vested interest SOURCE. (262) in the regulatory proceeding. Moreover, line managers and plant engineers would rather overstate studies. Two different types of studies are cited. than understate expected costs in order to ensure that they will be given a sufficient budget within The first, labeled the survey approach, involves the company to pay for the controls (140). Finally, surveying groups of people and asking them what when OSHA “forces” the diffusion of technology, they think should be spent on lifesaving programs. there is little or no experience on which to esti- The second type, revealed preference, uses data concerning either work and employment (labor mate the costs of widespread use of a particular new technology. Consequently, the costs of pro- market) or consumer purchasing (consumption ac- posed regulations are often overestimated. (For tivity) to calculate either workers’ or consumers’ examples of this, see 195. ) “willingness to pay” for risk reduction. The wide range in these values creates difficul- Uncertainties ties for analysts attempting to use the “willingness to pay” approach to place a value on the benefits Estimates of both costs and benefits are usu- of programs. Besides this practical problem of ally surrounded by uncertainty. The combined effect of these uncertainties and of assumptions choosing a figure for valuing benefits, there are also several conceptual problems with using the made by the cost-benefit analyst, both of which “willingness to pay” approach to assess lifesav- are found at every stage of the estimation of costs ing programs. (See 277 and 297 for a discussion. ) and benefits, can produce large differences in the analyses conducted by different people. Thus it Recently a third approach has been suggested is possible for one analyst to take a high estimate (259). Called “adjusted willingness to pay,” it at- of costs and a low estimate of benefits and con- tempts to combine the two traditional approaches clude that the program should not be undertaken, by estimating what an individual should be will- while another analyst can take a lower cost esti- ing to pay to avoid the financial losses associated mate and higher estimate of the benefits and con- with premature death. However, it still is not able clude that the program is worthwhile. Often these to include the willingness of family, friends, co- disputes cannot be resolved. (For a discussion of workers, and strangers to contribute to lifesav- several kinds of uncertainty, see 175. ) 292 ● Preventing Illness and Injury in the Workplace

Discounting ible damage to their health or well-being while the costs to employers involve increased expenses Finally, these analyses require that costs and and, possibly, reduced profits. This further com- benefits be made commensurate over time. This plicates the comparison of costs and benefits. usually involves adjusting future costs and benefits at a specified discount rate to calculate the pre- Alternatives to Aggregated Analysis sent value of the costs and benefits. The general justification for discounting derives from the fact Some analysts believe that ultimately solutions that resources can be invested to earn interest over to these problems can be found. OTA (539) has time. Thus, in order to compare costs and bene- suggested that the alternative of “arraying” the fits that occur in different years, all future effects various effects of a program or proposal should are discounted and expressed in terms of “present be investigated, rather than trying to reduce all values. ” effects to a single “bottom line. ” Ashford, et al. In practice, there is considerable disagreement (33), for example, have suggested “trade-off analysis.” This approach would involve a comprehen- over what discount rate should be used (539). sive description of the expected effects of an Moreover, although the logic of discounting is agency’s actions on three “flows”: economic, en- derived from several basic propositions of eco- vironment/health, and legal. To avoid the prob- nomic theory, the discounting of future costs and lems of monetization and valuation, all effects are benefits has two effects that create some left in their natural units. To reduce discounting controversy. problems, the time pattern of the effects is pre- First, the effects (both costs and benefits) on sented. Finally, the analysis provides a matrix of future generations are almost completely ignored effects and actors, to present a clear picture of who with most discount rates. Second, the process of gains or loses what from the regulatory action. discounting implies that risks that manifest themselves in the very near future are to be prevented Ethical Considerations before risks 10, 15, or 20 years in the future. This means that, all other things being equal, the risks The use of cost-benefit and cost-effectiveness of occupational injury should be reduced before analyses also raises ethical considerations. Sup- reducing the risks of occupational cancer. Or, for porters of the greater use of these techniques point example, that OSHA should act to reduce the risks to the limited resources available for improving of exposure to beta-naphthylamine (which has health and safety. This, they believe, implies that caused cancer in some workers after a latent the only moral course of action is to use those re- period of less than 5 years) before it acts to resources in a way that maximizes net social bene- duce the risks of exposure to asbestos (which fit. As a moral doctrine, this belief derives from causes various types of cancer with latent peri- the traditions of utilitarianism (246). The ad- ods ranging from 15 to 30 years). vocates of cost-benefit analysis believe that the use of formal analysis will help achieve the great- Distributional Effects est possible level of human welfare (304,446). Although cost-benefit and cost-effectiveness . . . estimating benefits and costs is often diffi- analyses were designed to evaluate economic effi- cult, especially . . . where the benefits may be ciency, they are not very well developed for the in terms of lives saved or pain and suffering evaluation of distributional implications. This is avoided. Some say this means putting a value on an important problem for the application of these human pain, suffering, and death, which is not only ludicrous, but downright immoral. If any- techniques to programs designed to improve oc- thing, we would argue, the reverse is true. Since cupational health and safety. Usually these pro- resources are limited, we cannot avoid the need grams are aimed at benefiting a group of work- to identify—and, in some way, to estimate-ben- ers, while the costs of the programs fall largely efits and costs. The more compassion we have for on employers. In addition, the benefit to the our fellow human beings, the more important this workers often involves the prevention of irrevers- becomes (304). — .—

Ch. 14—Decisionmaking for Occupational Safety and Health: The Uses and Limits of Analysis 293

In disagreement, MacLean and Sagoff (281) dis- severely limited: How can cost-benefit analysis cuss the philosophical justifications used to support claim to be either neutral or comprehensive if it cost-benefit decisionmaking and conclude that cannot deal with a wide range or moral, cultural, these justifications fail. Cost-benefit analysis is aesthetic, and political concerns? There may be not, in their view, a neutral decisionmaking rule some issues that raise few important cultural or because it is unable to take account of many com- moral issues; for example, the commodities mar- monly held ethical principles and values. In or- kets may be left to determine the prices of hog der to include concerns for equity and justice, a bellies or potash. This does not show, however, formal cost-benefit analysis would either have to that markets or market analysis can give us an find some way to assign a price to these concerns adequate policy for public safety and health. On or ignore them. But justice and equity are not the contrary, where moral, political, and cultural merely matters of consumer preference; they de- values—not simply economic ones—are at stake, pend on political and moral arguments: “Equity we need to make moral, political, and aesthetic is a matter of right or wrong; it is to be thought judgments. Cost-benefit analysis does not replace over and argued about. It is not a consumer serv- these “subjective” judgments with “objective” or ice or a fungible good” (281). “neutral” ones. Rather, it distorts or ignores the noneconomic values it cannot handle (281). (See On examining the cost-benefit decision rule pre- 205,246,277, and 297 for additional discussion of sented in President Reagan’s Executive Order the ethical implications of cost-benefit analysis. ) 12291, MacLean and Sagoff conclude that its unitary yardstick of positive net benefits is

SUMMARY

Improving occupational safety and health in- action. As decision roles, however, formal anal- volves the identification of hazards, the develop- ysis is considerably more controversial and sig- ment of control techniques, and the decision to nificantly more limited in its uses. control. The issue of decisionmaking—the question of who is to make decisions and on what In 1980, OTA concluded that cost-effectiveness basis—is important because interested parties can and cost-benefit analyses exhibited too many differ greatly about the nature of occupational methodological and other limitations to justify hazards and the best means to reduce or elimi- sole or even primary reliance on them in making nate them. Employers’ decisions tend to follow decisions. That conclusion is still applicable for the dictates of the competitive market system; the analysis of measures to improve occupational public decisions can consider a number of other safety and health. factors. These limitations include difficulties in quan- Various techniques of economic analysis includ- tifying the magnitude of the expected benefits, in ing cost-benefit and cost-effectiveness analyses valuing these benefits, in calculating the expected have been developed to assist private and public costs of improved health and safety, in perform- decisionmakers. As decision-assisting tools, these ing discounting, and in analyzing the distribu- techniques can help policymakers to reach sound, tional implications of alternative policies. The use well-informed, reasoned decisions about the of cost-benefit and cost-effectiveness analyses also management of workplace hazards. There is wide- involves ethical considerations. The advocates of spread agreement, at least in principle, that deci- cost-benefit analysis believe that the use of for- sionmakers need to have some minimal under- mal analysis will help achieve the greatest possi- standing of the important features of the problem ble level of human welfare. Critics of formal anal- to be addressed, the factors involved in the deci- ysis, however, argue that cost-benefit analysis is sion, and the implications of various courses of limited because it is unable to take account of 294 Preventing Illness and Injury in the Workplace many commonly held ethical principles and OSH Act does not require that OSHA base its values. decisions on the results of cost-benefit analyses; instead the agency must base its decisions on deter- The history of decisionmaking at OSHA is in- minations of “significant risk” and “feasibility. ” tertwined with debates over the allowable use of cost considerations. Some of that debate has been OSHA now uses a four-step process for mak- over the specific requirements of the OSH Act and ing decisions about health standards. First, the whether OSHA must base its decisions on a cost- agency determines that the hazard in question benefit analysis. When it passed the OSH Act, poses a “significant risk.” Second, OSHA deter- Congress brought the Federal Government into mines that regulatory action can reduce this risk. the field of occupational health and safety. Con- Third, it sets the regulatory goal based on reduc- gress, however, was less clear about the precise ing this risk “to the extent feasible. ” Finally, decision rules for OSHA to follow when setting OSHA conducts a cost-effectiveness analysis of health and safety standards. Because of the con- various options to determine which will achieve cern of a number of Congressmen about costs and this chosen goal in the least costly manner. economic effects, OSHA was required to consider OSHA also prepares Regulatory Impact Anal- the “feasibility” of its standards. But the final bill yses to comply with the requirements of Executive also included the goal that “no employee will suf- Order 12291. Because the results of the OMB refer material impairment . . . “ In addition, dur- view of these analyses are not made public, it is ing the 1970s, a series of Executive Orders and difficult to determine if OSHA decisions have other procedural requirements have affected been altered by OMB’s cost-benefit review. In ad- OSHA’s standard-setting activities. dition, one contract report prepared for this A number of legal challenges to OSHA stand- assessment suggests that for at least some of the ards have brought the courts into this arena. Two major OSHA health standards issued in the 1970s, of these challenges were decided by the Supreme the use of a cost-benefit decision rule would have Court. The Supreme Court has ruled that the led to less stringent standards. 15. Incentives, Imperatives, and the Decision to Control Page

297 ‘ 301 302

● ✎✌✎☛✎✎✎ 313 ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ 320 ...... ***..*.*...*.. 322 TABLE Page Safety Clauses,

... .., ... ,,, ,O ...... 315 FIGURE Page & Lawsuits 1970-82 . **....*.. ● ● . ● ● ● ● ● ● ● ● ● 310

15 Incentives, Imperatives, and the Decision to Control

OTA has identified a number of incentives and not been applied in many of the Nation’s 4.5 mil- one major imperative for the implementation of lion workplaces. Examining incentives and im- control technologies. As used here, an “incentive” peratives can assist in understanding the decisions is something that encourages an employer to im- to implement controls and can outline areas for plement a control. Because of the special value improvement. Most incentives and imperatives placed on health and safety, many people believe can be used together and, in some cases, they in- that society should, by law and regulation, require teract and build on each other. In some other employers to take the steps necessary to prevent cases, however, various historical circumstances work-related illnesses and injuries. This belief have led to compromises that bar the use of some underlies the basic approach of the Occupational incentives (e.g., workers’ compensation laws gen- Safety and Health (OSH) Act of 1970, which is erally prohibit employees from suing their em- an “imperative” for implementing controls. ployers), This chapter presents a description and assessment of these incentives and imperatives. A great deal that is known about controlling the causes of occupational illness and injury has

ASSESSMENT OF EXISTING INCENTIVES Voluntary Efforts After being informed of or discovering the existence of job hazards, some employers will take action to reduce, minimize, or eliminate those hazards. They do so either because of altruism toward their workers or out of enlightened self- interest. Strictly speaking, pure altruism implies that the employers take these actions only because of concern for their workers without thought of the ultimate implications for the firm in terms of worker good will, productivity, profits, or future sales. Although this will often be true for the personal motivations of health and safety profes- Photo credit OffIce of Technology Assessment sionals, most decisions concerning company pol- Signs are often used to provide information to workers. icy will consider carefully the potential effects on This one is a gentle reminder about the profits. use of safety shoes Probably more common than pure altruism are an enhanced corporate image or the perception voluntary actions out of enlightened self-interest. that a given firm is a “good place to work. ” In These actions are taken because a firm perceives addition, voluntary efforts may be undertaken to that voluntary efforts, although not necessarily solve a particular problem before the Government profitable in the short term, will benefit the firm or other groups become involved. (Of course, in the long run. This long-term benefit could be there may be other benefits to the firm in terms

297 298 Preventing Illness and Injury in the Workplace of reduced workers’ compensation costs, reduced employee health and safety is the commitment of capital costs if accidents damage plant and equip- top management. Two studies (441,443) have ment, or a reduced threat of potential liability or found that low-accident-rate plants tended to have an Occupational Safety and Health Administra- “greater management commitment and involve- tion (OSHA) fine. ) ment in plant safety matters” (443), although both of these studies involved relatively small numbers The pressures of the competitive marketplace of companies. In addition, a number of com- will, however, substantially limit the ability of in- panies, particularly large ones, have created de- dividual companies to improve employee health partments to handle issues of occupational and and safety. As described in chapter 14, if a com- environmental health and safety, have hired pro- pany spends resources on improving workplace fessional staffs with technical expertise concern- conditions and its competitors do not, this coming health and safety, and have established inter- pany can easily find itself at a disadvantage be- nal review mechanisms to ensure compliance with cause its competitors can use the resources thus health and safety standards. (See 179 for a dis- saved to expand or improve production. Unless cussion of some of these arrangements, and box the company can save money in some other way, P.) it will have lost money by attempting to do the “right” thing. The important role played by management There are several ways in which the company commitment to the goals of occupational safety may receive a financial return on its health and and health must be stressed. In the United States, safety investment—through reduced workers’ employers are responsible for the organization, compensation premiums, for example, or an im- design, and management of workplaces. If improved labor relations atmosphere, or reduced provements in employee health and safety are to vulnerability to fines for violating Government occur, they will have to involve decisions by man- regulations. In this case, the action is not exactly agement. It is therefore not surprising that man- “voluntary, ” but results from some other in- agement commitment has been called the single centive. most important ingredient in effective health and safety programs. Voluntary actions may result in a firm saving on both the direct and indirect costs of occupa- The other incentives and imperatives described tional accidents. The direct costs include work- in this chapter can be thought of as ways of ob- ers’ compensation payments for lost wages and taining the commitment of employers when vol- medical care, while the indirect costs include loss untary efforts are insufficient to correct occupa- of productivity, disrupted schedules, equipment tional health and safety hazards. These incentives and property damage, administrative time for ac- and imperatives generally reward or penalize com- cident investigations, and training of replace- panies. It has also been suggested that the comments. Estimates of the size of indirect costs range pensation of individual line managers should be from 4 times to 20 times the direct costs of ac- linked to improvements in product safety, pollu- cidents (82,380). The National Safety Council has tion control, and occupational health and safety estimated that the average total cost (both direct (139). and indirect) to an employer of a lost-workday In addition, there are a number of voluntary accident is about $9,400 (324). Sheridan (436) organizations and associations that have been ac- reports that one large firm has estimated the total tive in the occupational safety and health field. direct and indirect cost to be $14,000 for a lost- These include professional societies, voluntary work-time injury, $100,000 for a fatality, and standards organizations, trade unions, and em- $200,000 for injuries involving permanent dis- ployer associations. Companies and their employ- ability. ees may participate in these voluntary organiza- As noted in chapter 10, it has frequently been tions and these voluntary organizations are said that the most important variable in determin- sometimes involved in the development of some ing whether a firm will be generally protective of of the other incentives and imperatives. This is Ch. 15—incentives, imperatives, and the Decision to Control ● 299

Box P.—Example of Voluntary Control Efforts: Du Pont Effective company-run safety and health organizations require strong impetus and initiative on the part of management. At E. I. du Pont de Nemours and Co., this is reflected in the company’s “nine safety principles” (described inch. 6). The commitment of Du Pont to worker safety dates to the 19th century, when it produced gunpowder. In the early days the company’s mills blew up so regularly that they were built with three sturdy stone walls and one flimsy wooden wall, facing Brandywine Creek. When the inevitable explosion came, the wooden wall would blow out. The rest of the mill would remain standing and could be economically restored to service. “Going across the creek” became a company euphemism for being blown to bits (284). In fact, two du Pont family members were killed in these explosions. Du Pont requires all levels of management and all employees to be responsible for health and safety on the job. Not only does this attention improve employees’ working conditions, but whether dealing with 19th century gunpowder or 20th century petrochemicals, this attention reduces the chances of de- stroying valuable plant and equipment and incurring expensive downtime. In addition, Du Pont integrates safety with the management of the firm and uses the safety records of managers in making decisions about promotions. Du Pont established the Corporate Environmental Quality Committee in 1966 to carry out “top management’s commitment to environmental quality, including safety and health.” It meets weekly and is the overseer of four supporting committees, including an “Occupational Safety and Health Committee.” At the plant level, Du Pont has Central Safety Committees consisting of plant staff managers and assistant managers, the superintendent of safety and health, the environmental control manager, and the plant physician or medical supervisor. This committee directs the plant safety and health program, and meets at least once a month. Plant subcommittees, headed by first-line supervisors, vary in number and function, depending on plant size and needs. Anywhere from 6 to 10 workers serve on subcommittees; there are no requirements, however, for minimum employee representation on the subcommittees. Individual employees are selected by the plant manager to serve on subcommittees, which are considered the focal point for plant involvement in safety policy. Their findings are reported to the Central Safety Committee. About 95 percent of all safety hazards are corrected by line organization, and the resolution of individual safety hazards rarely requires involvement of a subcommittee. The minority of complaints that are referred to the committee concern issues that potentially involve plant-wide policy changes. It has been noted that because of its low injury rate, Du Pont annually saves millions of dollars for workers’ compensation compared with what its costs would be if its injury rate equaled the national average for all manufacturers (284). But the chemical industry as a whole, of which Du Pont is a part, generally has had a lower injury rate than that for all manufacturers, although Du Pont’s injury rates are better than even the chemical industry average. Little information is available on what it costs Du Pont to achieve this savings in workers’ compensation costs. Finally, while Du Pont is also a widely recognized leader in providing a comprehensive occupational health program for its employees, data on the incidence of occupational illnesses are limited (see ch. 2). Thus it is impossible to quantify Du Pont’s occupational illness rates and compare them with averages for the chemical industry or for all manufacturers. 300 Preventing Illness and Injury in the Workplace especially true for OSHA regulatory proceedings, deliberations of these committees. Small busi- in which both trade unions and employer asso- nesses may also be underrepresented. ciations are often extensively involved. There are several reasons for this. First, participation involves the commitment, at the very Voluntary Standards least, of staff resources. Oftentimes, unions and These voluntary activities may include coop- public interest groups lack sufficient staff re- erative efforts to develop voluntary standards, in- sources to participate. Second, standard-setting dustry standards, or consensus standards. How- groups have generally been created and staffed ever, comparatively few of these standards are by manufacturers and employers, in part because concerned with occupational health and safety. these groups began the process in order to stand- ardize equipment and designs (the “nuts and bolts” Some voluntary standards are purely advisory; standards). Labor and public interest groups are others, because they specify certain product at- generally not interested in those issues and do not tributes or dimensions, are ignored by firms only participate. This historical nonparticipation may at their peril. For example, some voluntary stand- have carried over into the safety and health standards specify the design and dimensions of nuts and ards activities. Third, the membership of the com- bolts. If a firm wishes to manufacture nuts to fit mittees is often unbalanced. For example, the the bolts of other manufacturers, the firm must committee that drafted the ANSI standard for follow the “voluntary” standard. Another exam- abrasive blasting operations consisted of 14 man- ple would be the various standards concerning the ufacturer and trade association representatives, electronic components, such as stereo components 6 people from professional organizations, 5 rep- and computers. The use of voluntary standards resentatives of government agencies, 3 individ- in these areas enables consumers to purchase sev- ual members, and 1 person representing a labor eral pieces of equipment from different manufac- organization. Such unbalanced representation not turers and then hook them together into a smooth- only provokes questions about the degree of pro- ly functioning system. tection afforded by voluntary standards, it also Still other “voluntary” standards have been makes labor and public interest groups hesitate adopted by Government agencies and now have to participate for fear that their involvement will the force of law. For example, the National Elec- imply approval of a voluntary standard that they trical Code, developed under the auspices of a had only minimal influence on. voluntary organization, the National Fire Protec- In addition, voluntary standards are just that, tion Association, is widely used as the basis for “voluntary, “ and probably cannot be enforced. the mandatory building codes of many localities Enforcement by a trade association or other orga- (203). Similarly, most of the existing OSHA safety nization may violate the antitrust laws. The Su- standards, which now have the force of law be- preme Court has ruled that companies can ex- hind them, began as “voluntary” standards de- change safety-related data and set standards to veloped under the auspices of the American Na- protect the public’s health and safety or to pro- tional Standards Institute (ANSI). (See discussion tect themselves from product liability actions. The in ch. 12. ) Court has also cautioned, however, that the The standard-writing work of technical orga- standards must not be used as a guise for ex- nizations, professional organizations, and trade cluding competitors or for facilitating price fix- associations is often delegated to subgroups or ing. The Federal Trade Commission, which shares committees. OTA has not precisely determined authority for the enforcement of antitrust laws what fractions of the participants in these volun- with the Justice Department, has advised that tary standard-setting groups represent employers, compliance with such a standard must remain manufacturers, labor unions, public interest voluntary. There appears to be a limited excep- groups, government, and others. It is clear, how- tion to this general rule-trade associations can ever, that labor and public interest groups have require members to abide by a standard that is been and continue to be underrepresented in the established to provide the legitimacy of something Ch. 15—Incentives, Imperatives, and the Decision to Corftrol . 301

(e.g., a breed of cattle) or to establish a network Although they can be updated more quickly in which the rules or standards are fundamental than OSHA has tended to rewrite its regulations to its functioning (e.g., the arrangements sur- (see ch. 12), unions and public interest groups rounding the purchase of flowers by telephone) question whether voluntary standards are suffi- (203). ciently protective. The existence of these standards might bring the practice of companies to a It is not likely, therefore, that a trade associa- common level, but, in general, voluntary stand- tion would try to enforce a health or safety standards and the voluntary approach will often lead ard by excluding an offending company from to only limited changes in the health and safety membership, organizing a boycott, or taking some conditions of the workplace. other punitive action. It may make efforts to encourage companies to comply, but the possibility of running afoul of the antitrust laws will make Provision of Information it stop short of requiring companies to comply (203). Thus even if a company follows a recom- The availability of information about occupa- mendation concerning worker health and safety tional hazards and their control is a necessary first made by a trade association or standards orga- step for many improvements in workplace health nization, the employer has no assurance that com- and safety. It may be provided by private sector petitors will follow suit. (Voluntary standards also organizations, including professional societies, play some role in liability actions and the enforce- voluntary standards organizations, insurance ment of the “general duty” clause of the OSH Act companies, employers’ associations, trade unions, (203).) universities, and individual experts. Or it may be provided through the research and dissemination Moreover, there is concern about the adequacy efforts of Federal and State Governments. The of voluntary standards, even if they are fully availability of information can combine with the adhered to by employers. Beyond the common other incentives to prompt employer action. For lack of labor and public interest representation example, a company management committed to on the committees that write these consensus improving job safety and health would use avail- standards, some people object to the standards able information to analyze job conditions and because they often involve a compromise among to make improvements. But without both com- the various interests, which may reduce the level mitment and information, no actions will be of protection afforded by these standards. In ad- undertaken. dition, the employer and manufacturer representatives on these committees are likely to agree to The provision of information through research a standard that each of them can already adhere and dissemination is an important activity unlike- to. They are unlikely to adopt a standard that ly to be met by private parties because informa- would require large changes in their current tion, once disclosed, becomes a “public good. ” operations, even if some industry leaders have This means that the developer of the information achieved higher levels of protection. Thus these cannot capture its full economic benefit because standards are likely to represent the “lowest com- that person cannot always charge for the infor- mon denominator” of performance within an in- mation. Furthermore, as illustrated by the descrip- dustry. tion of a company’s use of computer conferenc- ing (ch. 10), a company may elect to hold private Voluntary standards are an important source its safety and health information or to release it of information for employers, workers, govern- only under its own terms. ment agencies, and others involved in the health and safety field. They represent hours of effort State and local “right to know” laws, and the by many practitioners and professionals in this development of an OSHA regulation concerning field and can often be useful sources of technical “hazard communication” (labeling) promise to information, especially for defining terms and provide more information to workers and em- standardizing certain technical aspects of meas- ployers. The impetus for three laws has been pro- urement and control. vided by coalitions of unions, health and safety 302 . Preventing Illness and Injury in the Workplace

professionals, and public interest groups who terbalance other factors that affect employers’ believe that workers and their doctors have a right decisions about whether or not to implement con- to be informed of the identity of substances they trols (box Q), which may be more important than work with and the potential hazards of those sub- a lack of information in explaining why many stances. Economists have also suggested that one employers have not invested in health and safety way in which the existing market system may improvements. (In addition, as described in ch. have failed is by not providing workers with suf- 12, the OSHA consultation program also provides ficient information about hazards. In either case, information on hazards and controls to employers. ) one possibility is to require that information be provided to workers (628). Workers’ Compensation and Insurance The new OSHA “hazard communication” reg- Employers may take actions to improve job ulation requires, among other things, that con- safety in order to reduce the costs of workers’ tainers of “hazardous” chemical substances bear compensation and property insurance. The most a label with information on the name of the sub- important purpose of the workers’ compensation stance and the precautions to be taken. Both the system is to provide workers suffering from work- laws and the OSHA regulation enhance the displace injuries and illnesses with medical treatment semination of information to individual workers in expectation that this will lead to improvements in on-the-job health and safety conditions. In addition, labels provide important information to “down-stream” employers who previously had been ignorant about hazards in the materials they purchase. One important issue in this area concerns manufacturers’ desire to limit the information provided in order to preserve trade secrets, such as the chemical composition of a product. A second issue concerns which substances are deemed “hazardous” and who determines this. A third concerns the coverage of the laws and regulations. Do citizens and communities, in addition to workers, have access to this information? And what industries are covered? Fourth, there are questions about the relationship between the new OSHA regulation and the State and local “right to know” laws. In particular, Federal OSHA is arguing that, in general, its “hazard communication” regulation preempts State and local “right to know” laws. All of these issues are being considered in the recent legal challenge to the OSHA regulation. (For a general discussion of these issues, see 44. ) The increased availability of information on workplace hazards may serve as an incentive for companies to introduce controls. However, while the provision of information through both research and dissemination is an important first step, it is not sufficient by itself to guarantee improvements in health and safety. It may not coun- Ch. 15—Incentives, Imperatives, and the Decision to Control ● 303 and to compensate them for the income lost be- empirical evidence concerning the effect of this cause of those injuries and illnesses. But the im- incentive is thin. provement of occupational safety and health is There are, of course, difficulties in estimating also a goal of workers’ compensation. To the ex- what injury rates would have been in the absence tent that employers, through this system, pay for of workers’ compensation. In theory, various sta- the costs of medical treatment and lost wages, tistical techniques can be used to analyze the ef- there is a monetary incentive to reduce those costs. fects of particular programs, after adjusting for Thus, employers may install control technologies other factors that influence injury rates. One study in order to save on their workers’ compensation (105) found a decrease occurred in the number of premiums. In addition, the insurance companies certain occupational fatalities at the same time and the state agencies that provide workers’ com- that workers’ compensation was created. Several pensation insurance often provide services to im- other studies (53,104,106,107), however, have not prove health and safety conditions in their client found a favorable effect on injury rates from companies. workers’ compensation. Similarly, actions to reduce the costs of prop- The economic incentive regarding occupational erty insurance may coincidentally benefit worker injuries is diluted to some extent because many health and safety. For example, preventing fires employers pay premiums that are based on the and explosions in a factory may result in lower average experience for their industry or line of property insurance costs, as well as reducing the business (so-called manual rates). These rates number of workers injured. apply, it was estimated in 1972, to the 85 percent The history of workers’ compensation reveals of companies that employ about 15 percent of the different motives and goals for the various inter- work force (317). These firms are so small that ested groups (described further in ch. 11). Pro- year-to-year injury rates vary widely purely by gressive reformers sought to alleviate the loss of chance. Thus, in order to ensure year-to-year cer- income suffered by accident victims and their fam- tainty of payout by the insurance carriers, as well ilies and to encourage prevention. Businesses as to minimize administrative costs, these com- wanted to stabilize the uncertainties inherent in panies are grouped and pay rates determined from the liability system, to limit the growth in the size the manual. of awards, and to restrict more sweeping social At the other extreme are firms large enough to changes. They were also interested in prevention, predict with a high degree of confidence their ac- in part as an additional means to control or re- cident rates from year to year. Except in States duce costs. The National Association of Manu- that prohibit it, these firms generally insure them- facturers, for example, was very impressed with selves. It has been estimated that less than 1 per- the preventive effects of the German compensa- cent of firms self-insure, but that those firms tion system. In their view, the causes of accidents employ 10 to 15 percent of the workers included needed to be given equal consideration with the under the compensation system (317). consequences (274). Nearly every contemporary observer includes prevention as one of the goals Finally, firms in the middle are “merit-rated,” for workers’ compensation (30,46,106,131,317, generally using either “experience-rating” or 656,657), “retrospective-rating,” which are methods for ty- ing a firm’s premiums to its actual loss experience. Under experience-rating, insurers modify the man- Workers’ Compensation and ual or class rate for a firm by its actual accident Occupational Injuries experience for the most recent three years. Thus, Although many employers and insurance com- successful efforts to prevent injuries in the cur- panies believe that workers’ compensation is an rent year will lead to premium savings in the fol- incentive for prevention of occupational injuries, lowing three years. As the firm’s size increases, the precise circumstances under which this is true more weight is given to the company’s actual ex- are fairly complex (see, e.g., 656,657). Moreover, perience and less to the employer’s class or indus- 304 Preventing Mess and Injury in the Workplace

try average (409,656). Under retrospective-rating, There have been only a few empirical studies the employer pays for losses up to specified ceil- of the effects of experience-rating on injury rates. ing limits and the insurer pays for losses above Russell (409) found that large firms, which are the ceiling. Thus this plan “provides the firm with generally experience-rated, had lower injury rates a combination of insurance and self-insurance” than medium-sized firms. But she also found that (656,657). small firms had low injury rates as well, even though they, as a group, are not experience-rated. In addition, premium discounts are given to Two other published studies have not been able large policyholders and some insurance companies to isolate any measurable effects for the experi- pay dividends to their policyholders (682). These ence-rating system used in workers’ compensation dividends can be either “flat rate” (the same to (109,464). all policyholders) or “sliding scale” (higher rates paid to employers with better loss records and to A second limitation on the safety incentive pro- larger policyholders). The sliding scale plans, be- vided by workers’ compensation is found in the cause they tie dividends to the loss experience of benefit levels. Employers’ incentives are directly firms, may provide safety incentives (657). related to the degree that the workers’ compen- Although the rates for small companies are gen- sation system provides for the full social costs of erally not directly based on their experience, if injuries. Generally speaking, workers’ compen- such a firm has a particularly bad record it might sation pays the medical expenses associated with be placed in an “assigned risk pool,” with a cor- the injury but only a portion of the employee’s respondingly higher premium. Small firms with lost wages. very good records, however, generally do not re- The States generally replace two-thirds of lost ceive a reduction in their premiums (317). wages up to a maximum amount or ceiling. This A primary goal of insurance is to spread risks replacement, however, is usually based on the em- among employers, thus preventing, for a given ployee’s wages just before the injury and are often employer, a very large or catastrophic loss in any not adjusted for potential increases in the em- one year. The losses to be paid out are made, ployee’s earnings over his or her career. In addi- through insurance arrangements, into predicta- tion, many States have mandatory waiting peri- ble and regular annual payments. Moreover, ods or minimum lengths of time that a disability when the pooling of risks places safe employers must last before any payment will be made. Al- in the same group with less safe employers and though workers’ compensation benefits are not both have the same premium rate, the less safe taxable, the system does not replace lost fringe will not have any incentive to improve perform- benefits, which have become an increasingly ance. Thus, the goal of insurance or loss spreading larger portion of employee wage packages in the acts to dilute the incentive provided by workers’ last decade. Moreover, the ceiling on payments compensation. is frequently so low that many workers receive much less than the theoretical two-thirds replace- Although this effect has been often noted, the ment. It was estimated that during the late 1960s, degree of dilution is unclear because, although workers’ compensation had a median wage-replace- most employers do not appear to be experience- ment rate of only 50 percent (52). More recently, rated, most employees work for employers who a group of researchers found earnings-replacement are. Moreover, as Victor (656) has shown, the size rates of 46 percent, 59 percent, and 75 percent in at which a firm becomes eligible for experience California, Florida, and Wisconsin (81). But it is rating varies dramatically among industries, not clear to what extent this applies to other largely because of the differences in injury rates States. among industries. Finally, although most employees work in firms that have some form of Furthermore, as a general rule, workers’ com- experience-rating, it is difficult to determine what pensation replaces only lost earnings. It does not portion of all occupational injuries and illnesses usually compensate for “pain and suffering” or occur in these firms. even the loss of physical capabilities that do not Ch. 15—Incentives, Imperatives, and the Decision to Control 305 directly result in loss of earnings (260). “In [workers’ compensation appeared to provide only 22.5 ers’] compensation . . . the only injuries com- percent of the income maintenance for those pensated for are those which produce disability totally disabled by occupational injuries. The re- and thereby presumably affect earning power” maining three-quarters of their support came from (144). One example of this are injuries that are Social Security, employer/union retirement and limited to damage to the workers’ reproductive disability funds, veterans’ benefits, private insur- system. The States generally do not allow com- ance, welfare, and other sources. Workers’ compensation for this because such damage does not pensation should not necessarily be “charged” reduce the workers’ earning power (144,260). In with replacing all the lost income of injured work- these cases, injured and ill workers are not fully ers if other factors contributed to the total compensated. disability. It is, however, surprising that most of the income support for this group comes from However, in nearly all of the States, workers other sources, Employees and other public wel- with certain kinds of permanent partial injuries fare programs thus may be bearing much of the (e.g., loss of a hand or leg) receive compensation cost of occupational injuries (437). payments based on schedules of fixed dollar amounts for the part of the body affected. (For These results may occur because the disabled a list of these, see 484. ) For the same injury, all workers did not apply for workers’ compensation workers receive that same payment, even if their benefits, because no benefits were ever awarded wage levels differ. Thus, compensation for per- or because the workers’ compensation benefits ran manent partial injuries is often not directly tied out while the disability remained. Further research to an individual’s lost wages, although the pay- is needed to determine the factors that contrib- ments may ultimately be based on the average ute to this apparently inadequate compensation. wages of all workers. But from the standpoint of evaluating the incen- A third limitation is that some occupational in- tives of workers’ compensation for prevention, juries (and most occupational illnesses) are not the conclusion is the same. To the extent that these compensated at all because the worker fails to file costs do not enter the workers’ compensation sys- a claim, Some of these workers become depen- tem, employer premiums will not rise, and em- dent on other social insurance programs (e.g., ployers do not face the full financial incentive to Social Security) or pension programs that provide reduce the incidence of injuries and illnesses. disability benefits. Although this is mainly a prob- Finally, one other factor may influence the lem for occupational diseases (see discussion in safety incentives provided by the workers’ com- the next section) it may also be a factor for oc- pensation system. As is discussed later in this cupational injuries. chapter, under certain assumptions it is possible Based on data from the 1972 Survey of Disabled that employers may pay workers additional wages and Non-Disabled Adults, a Department of La- in order to attract them to hazardous jobs. If these bor report (596) concluded that only 43 percent additional wages or hazard premiums do exist, of people severely disabled by work-related in- the creation of a workers’ compensation system juries received workers’ compensation payments. may lead only to reductions in the hazard premi- Severe disability was defined as complete inability ums. In effect, compensation would shift from to work. The work-related disabilities were deter- before the accident to after the accident, and from mined by analysis of survey responses and thus all exposed workers to those who incur injuries. the results of these surveys need to be interpreted However, this shift may not result in any cautiously. change in injury rates (131,151,300). Three recent An analysis of a similar survey conducted in studies (83,150,151) have found, in fact, that in- 1978 found that only 33.1 percent of those whose creases in workers’ compensation costs and bene- main disability was due to an on-the-job injury fits are associated with decreases in employee were currently receiving workers’ compensation wages. Two of these studies (150,151) found that benefits (437). This analysis also found that work- increases in workers’ compensation costs were, 306 Preventing Illness and Injury in the Workplace

at least for nonunion workers, completely offset Barth and Hunt (46) describe a number of scien- by decreases in employee wages. This result im- tific, legal, and regulatory barriers that impair the plies that for these employees, workers’ compen- certain and timely compensation of occupational sation has no net effect on employer safety in- illnesses. They note that the system provides a bi- centives. furcated response to disease claims. Those that are readily connected to workplace exposure and Workers’ Compensation and that are relatively inexpensive (e.g., acute derma- Occupational Disease toses) are compensated much like accidental in- As opposed to the situation regarding work- juries. Disease claims that involve serious disabilities that are less clearly linked to workplace place injuries, most observers agree that many exposures (e.g., chronic respiratory disease) are cases of work-related disease fail to enter the workers’ compensation system. It is clear that any marked by extended controversy and long waiting economic incentive provided by workers’ compen- periods between a claim being filed and first payment. Moreover, these cases create a dispropor- sation will be reduced substantially if only a few tionate amount of administrative costs for the sys- occupational illnesses are compensated. However, representatives of the insurance industry claim tem. They note that “[f]or such claims the system that only a few occupationally related diseases go retains many of the undesirable features of the tort uncompensated (285,286). system that workers’ compensation was supposed to supplant” (46). There have been several estimates of the number of cases of occupational disease that are com- For occupational illnesses that manifest them- pensated. Data collected by the Bureau of Labor selves only after a latent period of years there may Statistics in the Supplementary Data System sug- not be a strong economic incentive for preven- gest that only 3 to 3.7 percent of all first reports tion. Firms contemplating an investment that will of workers’ compensation concern an occupation- reduce their workers’ compensation payments 20 al disease. Barth and Hunt (46) report that the per- to 30 years from now (or even payments for prod- centage of all compensation cases that concerned uct liability, as discussed later) can invest the occupational diseases ranged from 0.1 percent to money elsewhere for a better return. That alter- 5.5 percent for the 12 States for which data were native investment may be more profitable than available in 1975. Half these States fell in the range the possible reduction in future compensation between 1 and 2 percent. They also report the costs. Moreover, the firm may not even be in ex- results of a large survey conducted by Cooper & istence in 20 or 30 years, and its managers will Co. of 44,066 workers’ compensation cases in the almost certainly have changed. Thus a firm may fall of 1975. About 0.8 percent of the cases con- fail to take actions to prevent occupational illness. cerned on-the-job heart attacks and about 2.1 per- On the other hand, the threat of having an oc- cent were related to other occupational diseases. cupational disease disaster similar to that associated with asbestos may outweigh this financial Because there are no firm estimates of the total calculation (286). number of occupational disease cases, it cannot be said with certainty that this range of 2 to 4 per- A number of State statutes and interpretations cent means the compensation level for disease is of them impede compensation for occupational too low, too high, or just right. However, as men- diseases. For example, some States have restric- tioned in chapter 2, the number of all work-related tive definitions that make it difficult for disease cancer cases currently reported to worker’s com- victims to receive compensation. Most States have pensation agencies is substantially less than the abandoned or gone beyond the “schedules” or lists number of cancer cases estimated to be caused by of occupational diseases, which were often unduly occupational asbestos exposure. In addition, be- restrictive (such as textile producing States that cause of the difficulties faced by anyone filing a did not list byssinosis on their schedules, or coal claim for occupational disease compensation (dis- mining States that did not compensate coal work- cussed in this section), it is probable that many ers’ pneumoconiosis). Many States, however, disease cases go uncompensated. deny compensation for “ordinary diseases of life” Ch. 15–incentives, imperatives, and the Decision to Control ● 307 and will compensate only those that are “pecu- cases” for the compensation of disease by the liar or particular” to some line of work. This rule workers’ compensation system. has been applied even for occupations that face an increased risk of contracting an “ordinary dis- A study (431) of a group of insulation workers ease of life. ” Legal strictures concerning time who died from asbestos-related disease found that limitations and the requirements for proving of those who stopped working because of their causation also create difficulties for compensating terminal illness, two-thirds never filed for people with occupational diseases (46,261). disability benefits from workers’ compensation before their deaths. Of the claims that were filed, The Surveys of Disabled and Non-Disabled nearly half were still pending at the time of the Adults provide some information concerning the workers’ deaths. When there was a surviving sources of income for people disabled by disease, widow, claims for death benefits were filed in though, as mentioned earlier, the self-identifica- fewer than half the cases. tion of “job-related” diseases means care must be taken in analyzing the results. The 1978 survey In addition, workers’ compensation was the sole showed that of those who attributed their disabil- or primary source of medical benefits for only ity to bad working conditions, only 21.8 percent about 4 percent of these workers. The workers had ever applied for workers’ compensation, as and their families appear to have relied on union opposed to 64.4 percent for on-the-job injuries. welfare funds, Medicare, other private insurance This application rate was static between the 1972, plans, and their own savings to pay for the med- 1974, and 1978 surveys (437). ical costs of asbestos-related disease. A tort liability suit was filed in fewer than one-fifth of the The 1972 survey found that of those who cases for which data were available. In only 9 per- thought their disabling illness was due to work- cent of the cases did the worker or spouse file both place conditions, only 3 percent were receiving a claim for workers’ compensation and a lawsuit. workers’ compensation. The 1974 figure is essen- In over half the cases (57 percent), neither a work- tially the same-5 percent (596). By 1978, of those ers’ compensation claim nor a liability suit was citing “bad working conditions, ” about 13 percent filed. The most important factor in explaining the were receiving workers’ compensation benefits, failure to file for workers’ compensation appears compared with 33 percent for those with job- to be “ignorance, either of the source of the dis- related injuries (437). Although about 23 percent ease or of the legal rights of survivors to compen- of total income replacement for job-related in- sation. ” juries comes from workers’ compensation, for disabilities due to “bad working conditions, ” the Johnson and Heler (236) have calculated the ex- figure is about 12 percent. Nearly half the income pected monetary losses for the families of these maintenance for this disease-disabled group comes workers. They estimated that the gross loss due from Social Security Disablity Insurance. These to disability and death from asbestos-associated estimates have been criticized for various reasons disease amounted to over $250,000 per family. related to the design of the survey, the size of the Half the widows they studied received no bene- survey population, and the use of self-reporting fits at all. The other half received a variety of ben- to describe both health conditions and the work- efits. About 28 percent of these benefits came from relatedness of those conditions (214,285,680). workers’ compensation, while 16 percent came from tort suits and settlements. The remainder However, more detailed studies of two well- came from Social Security, private pensions, and known occupational hazards, asbestos and cot- veterans’ benefits. Johnson and Heler also calcu- ton dust, support the general conclusion that most lated the net financial losses for these families. On of the income support for workers disabled by oc- average, the widows who received benefits had cupational illnesses does not come from the work- approximately one-third of their losses replaced. ers’ compensation system (217,431). Because both asbestos and cotton dust have been clearly linked In theory, through the workers’ compensation to occupational disease and have received wide- system, employers bear the costs of occupational spread public attention, they represent the “best disease. But for this group of workers, about 85 308 ● Preventing Illness and Injury in the Workplace percent of the gross wage loss was borne by the In fact, many insurers provide loss-control serv- families of the affected workers. Of the small por- ices to their policyholders. The results of one tion of lost wages that were replaced, less than survey imply that private insurers and State work- half were paid for by employers through workers’ compensation funds provided a total of 1.6 ers’ compensation and by producers through tort million such visits to policyholders in 1974. Pri- suits and settlements. vate insurers provided 1.5 million of these visits (472,473). More recently it has been estimated that Hughes (217) has similarly studied income re- the insurers who are members of the two major placement for a group of workers exposed to cot- trade associations (the Alliance of American In- ton dust who developed byssinosis, He found that surers and the American Insurance Association, workers’ compensation replaced only 6.9 percent or AIA) employed about 8,600 loss-control spe- of the estimated lost income for this group. Forty percent of lost income was replaced by funds from cialists in 1983, while independent firms added about another 1,000, for a total of 9,600. The Social Security, 3.5 percent from Veterans Admin- istration benefits, and 2.6 percent from private 8,600 specialists working for the members of the two associations conducted about 1.5 million pension plans. The total income replacement visits to policyholders. It is also estimated that amounted to about 53 percent of expected earn- about 177,000 samples of suspected toxic sub- ings. “Even with Social Security funds in crisis, stances were analyzed in 1983, while approx- it is apparent that a massive shift of costs has been imately 40,000 policyholders participated in train- made from the employers to the general taxpay- ing programs provided by the Alliance and AIA ing public, due to an almost nonexistent work- member companies (286). ers’ compensation system—a form of public subsidy to disease-producing industries . . . “ (217). Some of the expenditures for “loss control” are expenses for inspecting workplaces in order to de- Role of Insurers termine the nature of the operations and to set Beyond the incentives faced by private employ- premiums (30,473), rather than to suggest or man- ers, insurance carriers might have an independ- date preventive actions. Little statistical informa- ent financial incentive to prevent injuries and ill- tion is available to determine the percentage of nesses. If an insurance company can improve the insurer visits that are primarily for collecting in- loss experience of its policyholders, it maybe able formation for rate-setting and the percentage of reduce the amount that must be paid out in claims. visits that provide safety advice (473). In practice, however, this incentive is limited, too. Low benefit levels reduce the incentives to Aside from this rate-setting activity, insurers insurers just as they reduce the incentives to em- have offered loss-control services as one way of competing in an industry that until recently was ployers. In addition, if the merit-rating system is subject to detailed price and service regulation by working properly, insurers will have little incentive to improve the loss experience of firms that State agencies. In general, that regulation did not are fully merit-rated. For such companies, the ben- allow insurers to compete by charging different rates. It is not clear what effect various efforts to efits of reduced claims will be received by the em- “deregulate” this industry will have on the pro- ployer. vision of loss-control services (268). Insurers will receive an independent benefit only if they can cut losses in the time period before However, although insurers do provide consul- the premiums are adjusted by the experience fac- tative services to their policyholders, they gener- tors. Insurers will also benefit if they can reduce ally do not grant rate decreases to employers who the losses for firms that are not fully merit-rated. accept such services. It is likely that the advice However, insurance industry representatives provided by insurers has a positive effect on argue that reduced rates are important for attrac- safety, but this effect may be limited by fear that ting and holding customers, and that maintain- the employer, a valued client, will simply change ing customers is a powerful incentive for reduc- insurance companies rather than make a large ex- ing claims and therefore rates (286). penditure for health and safety controls. ——

Ch. 15—Incenfives, Imperatives, and the Decision to Control ● 309

To the extent that occupational diseases do not producers and employers should bear the costs enter the compensation system, there is no finan- of occupational disease (14). However, the use of cial incentive to prevent them, either on the part Federal funds to supplement an occupational dis- of the employer or the insurer. Compensation for ease fund may dilute this incentive. If Congress illnesses will also be discounted substantially be- takes action concerning occupational disease com- cause of the latent period between exposure and pensation, this action could include a requirement disease manifestation. The insurance industry, that companies take concrete steps to prevent however, maintains that it is taking steps to avoid future cases of disease. future occupational disease disasters similar to the asbestos situation (286,675). Tort Liability Changes in Workers’ Compensation The effects of court-enforced tort liability on In practice, the workers’ compensation system employer practices concerning health and safety only provides a limited economic incentive for has been highlighted by the large number of law- prevention, especially for occupational illnesses. suits concerning worker exposure to asbestos (box Changes have been suggested to improve eco- R). In particular, attention has focused on the nomic incentives provided by this system. These well-publicized case of one supplier of asbestos, include increasing the degree of experience-rating the Manville Corporation (formerly Johns Man- in the compensation system (317), instituting em- ville), which has filed for a corporate reorganiza- ployer deductibles or copayments for the first $500 tion under the bankruptcy laws because of the of compensation expenses (449), and changing the burden of paying numerous liability suits. Fac- Federal income tax deductions allowed for working the threat of potentially costly lawsuits and ers’ compensation premiums (228). However, be- large awards for damages, employers and manu- cause workers’ compensation is currently admin- facturers may take action to improve workplace istered by the States, the first two suggestions health and safety. would involve changes in each of the States or As discussed in chapter 11, before the passage creation of a single Federal system. of workers’ compensation laws in the early part Moreover, the effects of the suggested changes of this century, injured workers could sue their are not completely clear. For example, Victor employers for damages. They encountered sub- (656) suggests that the creation of employer de- stantial difficulties in winning these suits, how- ductibles might increase the incentives faced by ever. Workers’ rights to sue their employers were some employers while decreasing the incentives greatly restricted with the creation of the work- faced by other employers. The net effect on in- ers’ compensation system. For most cases of oc- juries is unclear. In addition, increasing the eco- cupational injury or illness, workers are not nomic incentives of workers’ compensation could allowed to sue their employers for such compen- increase employers’ incentives to contest claims, sation. Instead, the workers’ compensation sys- as well as their incentives to prevent illness and tem, in theory, provides a specified level of ben- injury. Finally, many of the limitations of injury efit that is awarded to pay medical costs and taxes (discussed in ch. 16) also apply to these sug- compensate for lost wages. gested changes in the workers’ compensation In general, the law of torts provides the oppor- system. tunity to sue for monetary compensation when There has been considerable recent discussion property has been damaged or a person has been concerning the possibility of creating a Federal injured, but precisely defining the field of tort law system to compensate for occupational exposures is difficult. One noted scholar has written: to asbestos (517). Prevention should be a consid- A really satisfactory definition of a tort has yet eration in any changes in compensation. A com- to be found. . . . Included under the head of torts pensation system should be designed to “inter- are a miscellaneous group of civil wrongs, rang- nalize” the costs of disease. In other words, ing from simple, direct interferences with the per- 310 ● Preventing Illness and Injury in the Workplace

Asbestos Use I 19s0. .-, * textile fibers, and polypropylene fibers.

son . . . or with property . . . up through vari- uses. Whether or not the seller was legally negli- ous forms of negligence. . . The law of torts . . . gent is irrelevant in this situation because the very is concerned with the allocation of losses arising fact of the defect indicates a breach of the war- out of human activities. . . So far as there is [a ranty (203). general principle] . . . it would seem that liability must be based upon conduct which is socially Under strict liability, the manufacturer is lia- unreasonable . . . or unreasonable interference ble for injuries resulting from a defective prod- with the interests of others . . . (Presser. Quoted uct that is unreasonably dangerous, without rein 379). gard to fault or contractual limitations. Here, the Baram (379) notes that controversies involving degree of diligence or care in preventing injuries risks to human health have traditionally involved is immaterial (203). Strict liability has traditionally legal concepts grouped under tort law. Four areas been used for “ultrahazardous” or “abnormally of tort law have been used to resolve controver- dangerous” activities (42). In recent years, the sies that involve such risks: negligence, product courts have adopted in the field of product lia- liability, nuisance, and strict liability. Of these, bility law some of the concepts of strict liability. negligence, product liability, and strict liability In fact, product liability suits can be based on are applicable to issues of workplace health and negligence, warranty, and strict liability. Liabil- safety. ity is now applied in cases where: 1) a product was defectively designed, 2) a product was defec- Negligence has been defined as “conduct which tively manufactured, or 3) a product was sold falls below the standard established by law for without proper warnings concerning its use and the protection of others against unreasonable risk dangers. of harm. ” To maintain a suit based on negligence, four elements must be shown: the existence of a The “duty to warn” has been the basis for many legal duty or obligation to protect people from of the successful lawsuits concerning asbestos ex- harm, a failure to conform to that duty, a prox- posures, in which courts have ruled that the man- imate causal connection between that failure and ufacturer has an obligation to take reasonable ac- a resulting injury, and an actual injury (379). tions to discover the hazards associated with a Under the general doctrine of negligence, a per- product and to warn accordingly. This duty may son can be held liable for an injury if he or she include specialized testing (69): failed to act in a “reasonable” way to prevent the injury. The manufacturer is held to the knowledge and skill of an expert. . . [This] means at a minimum Product liability developed out of warranty he must keep abreast of scientific knowledge, dis- law. The courts have ruled that manufacturers coveries, and advances and is presumed to know and sellers implicitly warrant that their products what is imported thereby. But even more impor- are suitable and safe for all reasonably anticipated tantly, a manufacturer has a duty to test and in- 312 ● Preventing Illness and Injury in the Workplace

spect his product. The extent of research and ex- products, or when the proof for the causal periment must be commensurate with the dangers connection is only through statistical in- involved. ference. By far the most common use of tort actions in ● The long periods of time that often elapse be- occupational safety and health are suits against tween exposure and effect make the causal a third party—for example, the manufacturer of connection and the identification of products products purchased by the employer for use in and manufacturers much more difficult. the workplace. Thus, employees injured by a mal- ● Even if the manufacturer can be identified, functioning punch press can sue the manufacturer the firm may have gone out of business, be of the press, while employees who used asbestos currently unprofitable, or otherwise not able in their work can sue the manufacturer of the to pay damages. asbestos product. Because the ill or injured worker ● In some States, statutes of limitations bar is not directly employed by this third party, lawsuits for damages beyond a certain length of suits concerning work-related illness or injury are time (generally 3 to 10 years) after the last not prohibited by workers’ compensation laws. exposure to the hazard, even if the disease only manifests itself 20 or 30 years later (203). Although many employers are concerned about legal liability, in most cases it does not directly As noted earlier, State workers’ compensation affect them as employers. Rather, it affects the laws generally bar suits by employees against their products that they sell to consumers or to other employers, although there are several exceptions employers. In 1978 it was estimated that while to this. First, any employer that is not covered only 11 percent of all product liability awards in- by the State workers’ compensation law may be volved work-related injuries and illnesses, these sued. Some employment categories that com- cases accounted for 44 percent of the total dollars monly fall in this group are agricultural employ- awarded for product liability (228). ees, domestic servants, employees of very small businesses, railroad workers, and employees of This slightly roundabout approach of third- charitable organizations (407). party suits may lead ultimately to improvements in working conditions and in the health and safety Second, courts may grant employees “injunc- of the work force, but a number of factors limit tive and declaratory relief. ” In one case, Shimp its effectiveness. The first is that it does not apply v. New Jersey Bell Telephone Co., an employee to employers, even in cases in which the employer obtained a court injunction to require the employ- may have been in the best position to ensure that er to prohibit smoking in general working areas. the equipment and products were being used in The court ruled that such an injunction would not a safe and healthful fashion. In these cases, the be barred by the workers’ compensation law of manufacturer is still held responsible. New Jersey and that the employee, who was al- Other limitations on the usefulness of third- lergic to cigarette smoke, had a common-law right party suits are found in the traditional degree of to a healthful work environment. It is not clear proof demanded by the courts in liability actions. whether other courts will also adopt this reason- This burden of proof is often very difficult for the ing (407). worker, especially in cases of occupational dis- Third, in nearly all States workers can sue em- ease. The problems employees can encounter ployers for damages in cases of willful or inten- include: tional acts. This includes situations in which the It is difficult to prove (or even recognize) that employer actually intended that the employee be harm has occurred when a disease, such as injured, as well as cases that involve fraud or cancer, may be caused by many factors, in- deceit. by the employer. For example, if an em- cluding occupational ones. ployer deliberately conceals from a worker infor- The isolation of the product or products that mation concerning a work-related illness, employ- “caused” the harm is difficult, especially if ees may be successful in collecting damages from there are harmful interactions with other the employer. The West Virginia Supreme Court Ch. 15—Incentives, Imperatives, and the Decision to Control ● 313 — extended this rule to cases of willful, wanton, or Labor Market Forces, Collective reckless employer misconduct (407), although this Bargaining, and Workerss Rights was subsequently restricted by a new State law (2,342). To some extent, employers are motivated to improve employee health and safety because of pres- Fourth, in some jurisdictions an employer may sures from the labor market. Economists since be sued by an employee when acting in a “dual Adam Smith have hypothesized that employers capacity. ” The three main types of dual capacity would have to pay more to attract workers to jobs cases involving workplace health and safety issues with unsafe conditions or other adverse working are based on the obligations of the employer as conditions. In theory, if there is complete infor- land owner, medical practitioner, or seller of mation about workplace hazards and alternative products. Thus an employer, for example, who job opportunities, employers may find that they offers an onsite medical service and who negli- cannot attract enough workers. There are then gently treats an employee maybe sued for medi- two basic choices—raise the wages or reduce the cal malpractice. Similarly, an employer may also hazards. Thus, the possibility exists that this la- be sued by a worker for injuries or illnesses in- bor market pressure may induce improvements curred due to a product manufactured by the em- in employee health and safety. ployer. It should be noted that only a minority of jurisdictions currently allow such “dual capac- The payment of additional wages for occupa- ity” suits (407). tional risks can be seen in the existence of “hazardous duty pay” in certain high-risk occupations. It has been suggested that employees be allowed But these additional wages may also be built into to sue their employers. For example, Amchan (2) the general pattern of wages for an industry or has proposed that workers and their families be occupation. In this situation, they would not be allowed to sue employers if the worker is killed directly observable, but would be included as part or permanently disabled due to the “willful, in- of the total wage. The additional wages paid for tentional, or grossly negligent conduct” of the em- workers exposed to occupational risks have been ployer. Although such change might enhance em- termed “hazard premiums” or “compensating ployer incentives for prevention, it would also wage differentials. ” In theory, various statistical have a number of social, legal, and economic im- techniques could be used to separate the factors plications that need to be considered carefully. that determine wages, thus testing whether a “haz- The future importance of tort liability as an in- ard premium” existed and determining its size. In centive to control workplace hazards is unclear. practice, the data are difficult to analyze because Although much attention has been given to the of problems in measuring job risks and in adjust- circumstances of asbestos exposure, asbestos may ing for other factors that influence wages. not be typical. Currently the prohibition of most To date, the published studies on this question employee suits against employers severely weakens have generally found compensating wages for in- the incentive. In addition, the practical problems creased risks of death, but are inconsistent on of proving causation will tend to limit to just a whether there are also compensating wages for few hazards the lawsuits by workers against sup- increased risks of nonfatal injuries. In fact, sev- pliers. eral studies have found, contrary to expectations, Congress is currently considering legislation to that some groups of workers were not receiving change certain aspects of product liability law. As compensating wages. In some cases hazardous suggested earlier for changes in workers’ compen- work was associated with lower wages (138,192). sation it is important that the effect of these Moreover, even the studies that have found in- changes on incentives for prevention be consid- creased wages for hazardous work have not been ered carefully, consistent on the size of these increased wages or 314 Preventing Illness and Injury in the Workplace the implied “value of life.” (See ch. 14 for a table to health and safety. Apparently, emphasis on presenting this wide range of values. For reviews these issues at the plant level has increased dur- of this literature, see 191 and 448. ) ing the last decade. This may be because of a belief “that control and prevention of job hazards can In addition, employers may find that their be improved through the combination of more ef- workers are quitting soon after starting work be- fective OSHA rules, regulations, and enforcement cause of unsafe or unhealthful working conditions programs with trade union programs, including (658a). Efforts to reduce such turnover and to im- more effective collective bargaining contracts and prove employee morale may lead to investments their administration” (185). in health and safety controls. Of course, all these labor market pressures are The Bureau of National Affairs (BNA) has tab- vitally dependent on the existence of alternative ulated the major provisions in over 5,000 collec- job opportunities. In areas and times of high un- tive bargaining agreements and analyzed closely employment, this incentive to control hazards will a representative sample of 400 contracts (77). be substantially reduced. Moreover, other labor Table 15-1 illustrates the frequency of health and market imperfections, such as incomplete infor- safety provisions in collective bargaining language mation about job hazards, the costs of searching from 1954 through 1981. Until 1971, according for jobs, and unequal bargaining power also limit to the BNA, there was only a slight increase in this incentive. the prevalence of such provisions: for example, from 38 percent of contracts in 1954 requiring Collective Bargaining Agreements management to “take measures” to protect the workers to 42 percent by 1971. The only type of Negotiation and collective bargaining can also contract provision that showed more than a slight bean incentive. This requires that unions be com- increase was that for safety committees. mitted to the recognition of health and safety concerns and assign high priority to these issues in Passage of the OSH Act in 1970 coincided with collective bargaining. Other important union obli- an increase in the number of contracts that in- gations for the success of this strategy include be- cluded health and safety clauses, and the provi- ing fully informed about OSHA regulations, en- sions became increasingly specific. The overall fre- forcement procedures, and employee rights under quency of safety and health clauses between 1954 the OSH Act. In addition, the union must estab- and 1970 had remained between 60 and 65 per- lish a system to monitor employer actions, usu- cent, but by 1975, 82 percent contained such ally through safety stewards or trained local rep- clauses. There was also an increase of 11 percent- resentatives, as well as develop procedures for age points in clauses requiring employer com- hazard identification, management negotiation pliance with laws from 1971 to 1975. procedures, enforcement of committee findings, and rank-and-file feedback about these findings The general statement of responsibility in most (253). contracts states that the company must make “reasonable provision for the health and safety of the Labor unions can exert influence because their employees, “ which appears to be redundant with representatives can be present for all plant oper- the general duty clause of the OSH Act (see ch. ations, every working day. However, because the 12). Moreover, 29 percent of the contracts in 1981 collective bargaining process involves a process required the company to comply with present of negotiation and compromise, unions may make legal standards. These provisions enable local tradeoffs between economic benefits (including unions to use the grievance process (an internal greater wages, job security, and fringe benefits) dispute mechanism negotiated by labor and man- and more attention to occupational hazards. agement for the resolution of employee com- There are approximately 150,000 separate col- plaints) to change or influence health and safety lective bargaining agreements in the United States, conditions, in addition to filing an OSHA com- and 82 percent of these contain some reference plaint. Ch. 15—Incentives, /imperatives, and the Decision to Control ● 315

Table 15-1.—Percentage of Contracts Containing Health and Safety Clauses,@ 1954-81

Company obligation Employee obligation Other To meet To provide To provide Must wear provisions To take legal safety first-aid safety Must obey Safety Years Clauses measures standards equipment equipment equipment safety rules committees 1954 ...... 600/0 380/o 13 ”/0 270/o 11 ”/0 7% 17“/0 180/0 1961 ...... 65 34 11 32 16 7 11 28 1966 ...... 62 35 NA 28 12 6 8 29 1971 ...... 65 42 15 32 14 7 13 31 1975 ...... 82 50 26 36 21 NA NA 39 1981 ...... 82 50 29 42 21 NA NA 43 %cmsiderable overlap was noted in all categories NA—Not available SOURCE (77)

Joint Labor-Management Health and has been suggested that safety is such an “inte- Safety Committees grative” issue, leading to a capability for cooperative problem solving on safety issues through Joint committees vary in structure, organiza- health and safety committees in the context of an tion, and capacity for intervention. The role of overall bargaining relationship (253). the committee ranges from a limited monitoring of workplace routines to a strong source of pres- Collective bargaining often serves as a direct sure on workers, union, management, and OSHA. and immediate stimulus for setting up these com- The proportion of contracts with a clause con- mittees. However, it does “not guarantee that an cerning health and safety committees rose from active and ongoing committee will develop” (253). 18 percent in 1954 to 43 percent in 1981, with most Indeed, there are numerous instances where the of the increase following passage of the OSH Act presence of a contract clause requiring safety com- (see table 15-1). mittees has not resulted in regular meetings or useful recommendations by functioning committees. The growth in labor-management safety and health committees may represent an attempt on There has been only a limited amount of re- the part of both labor and management to resolve search assessing the impact on worker health and issues at the local level. An important factor in safety of these joint committees. Kochan, Dyer, a union’s ability to improve safety and health is and Lipsky studied union and management atti- rank-and-file concern about these issues and the tudes in 59 plants (253). For several reasons, they relative priorities that union members and their were unable to obtain data on actual injury and leaders place on health and safety in relation to illness experience at these plants. Thus their results other negotiable provisions, such as wages, work are generally based on subjective perceptions of hours, seniority, and grievance procedures. Ac- workers and managers from surveys and interviews. cording to Kochan and his colleagues, the most They attempted to describe the conditions important determinant of committee effectiveness under which labor-management committees from the management perspective is the attitude would have a high level of activity and would and commitment of the top management. It is also continue to function. These conditions occurred important that there be a balance of strengths be- when OSHA pressure was perceived to be strong, tween both bargaining groups. Acting as an ex- when the local union was perceived to be strong, ternal pressure, the presence of OSHA may place when there was substantial rank-and-file involve- a weak union in a more equal bargaining posi- ment in health and safety issues, and when man- tion with management (253). agement approached these issues in a problem- Collective bargaining is commonly associated solving manner. The committees that produced with opposition in interests and an atmosphere the largest number of recommendations tended of limited trust, but there may be “integrative” to have a high level of input from the local union issues over which parties share common goals. It membership, frequently reported back to those 316 Preventing Illness and Injury in the Workplace members, and were in workplaces with a high (leading to fewer worker complaints to OSHA) proportion of young workers. The degree of top and improved employer compliance with OSHA management commitment was also a very impor- standards (leading to fewer citations) (see also 335 tant variable for determining the success of these and 341 and box S.) committees. In particular, management tended to adopt a problem-solving approach to health and OSHA and Workers’ Rights safety issues when they were under pressure from OSHA to comply with particular regulatory re- The OSH Act itself created a number of oppor- quirements (253). tunities for worker participation concerning occupational safety and health. The act provided Cook and Gautschi (121) used data for 113 that workers could: manufacturing firms in Maine to study the effectiveness of OSHA and labor-management com- ● request OSHA inspections, mittees. They found a favorable effect on plant ● participate in the conduct of an OSHA in- injury rates from OSHA citations. In addition, spection, there was some evidence that union-management ● participate in any of the stages of a pro- safety committees were also effective in reducing ceeding before the Occupational Safety and injury rates. For the plants with 300 or more em- Health Review Commission, ployees, this effect was significant at the 90 per- ● contest the “reasonableness” of the abate- cent confidence level. However, for all the plants ment date set by OSHA, with 200 or more employees, the effect was not ● participate in standards development and the statistically significant. issuance of variances, and ● request a Health Hazard Evaluation from the California has created a program to encourage National Institute for Occupational Safety the formation of joint committees on construc- and Health. tion sites. Under this “Cooperative Self-Inspection Program, ” a site will be exempt from routine The right to participate in Occupational Safety OSHA inspections if a joint committee is set up and Health Review Commission proceedings and to perform regular inspections of the workplace. to contest OSHA citations has been the subject This program has been implemented on six proof several court cases. The result of these cases, jects. The California Division of Occupational interpreting the language of the OSH Act, is that Safety and Health reports that the injury rates at employees and their unions can participate in these worksites are “substantially lower” than these proceedings if the cited employer is con- both the rates for other California construction testing a citation. Employees and unions also have projects and the experience of these same com- an independent right to contest the reasonableness panies at other sites not included in the program of the prescribed abatement date. However, em- (337,347). ployees and unions do not have an independent right to contest an agreement between OSHA and In a study of survey responses from 127 firms the employer concerning the nature of the rein Massachusetts, Boden and colleagues (66) quired controls, the type of citation, or the pen- found that the “mere existence” of a joint com- alty amount. If OSHA and an employer agree on mittee in a workplace had no effect on either the these issues and the employer withdraws its “con- number of OSHA inspections prompted by worker test” before the Review Commission, the employ- complaints or the relative hazardousness of the ees can object only to the specific abatement dates. firm as measured by serious OSHA citations. The (See 307 for excerpts from several of the impor- researchers also conducted more detailed inter- tant cases on this issue. ) views with union and management representatives at 13 of the firms with labor-management com- In addition, the act created a mechanism in sec- mittees. The data from these interviews suggest tion 11(c) to protect employees from job discrimi- that the committees that were perceived as “ef- nation for having exercised any of the rights listed fective” apparently increased perceived safety above. However, the resources devoted to OSHA’s —

Ch. 15—lncentives, imperatives, and the Decision to Control 317

Box S.—Collective Bargaining Results: Worker Training and Health and Safety Committees Training Concerning Health, and Safety Approximately 420,000 of the total 800,000 General Motors (GM) employees are members of the United Automobile Workers (UAW). The UAW-GM contract provides for training of full-time union health and safety representatives for each GM plant with more than 600 employees (37). The full-time union safety representatives, selected first by the locals and finally by the UAW international headquarters, train alongside GM’s own safety representatives for 40 hours at the General Motors Institute. Training involves hazard recognition, OSHA complaint procedures, and OSHA regulations. In accord with the UAW-GM contract, the union representative accompanies his or her management counterpart twice monthly on inspections, and also walks with OSHA inspectors on their tours. The union member’s role also includes reviews of training and education programs and accident reports, and the person is empowered to shut down a hazardous operation, but only with joint approval by the plant safety officer. In addition, the union safety and health representative is involved in the initial stages of grievance resolution concerning health and safety issues. Some unions that represent workers who move frequently from worksite to worksite, such as the International Brotherhood of Painters and Allied Trades (lBPAT) and the United Association of Plumbers and Pipefitters, also place high priority on the training and education of their workers and union safety representatives. Part of this attention is directly related to securing desirable contract language concerning health and safety. IBPAT has concluded, for example, that its training program had a “statistically significant impact on collective bargaining, leading to more and better safety and health bargaining language in both local and district.” Labor-Management Committees One refinery plant of the Shell Oil Company that has 2,000 workers organized by the Oil, Chemi- cal and Atomic Workers (OCAW) also has contractual provisions for union-management committees. The current contract language was drawn up in 1973. Unlike most OCAW contracts, this one provides no trained hygienist to survey worksites regularly. It does, however, provide for a union-management health and safety committee that:

, ● is to be composed of an equal number of representatives from the hourly and staff [management] groups ., # and is to meet periodically to discuss health and safety matters and make recommendations to management. Where a recommendation made to the company is not accepted, an explanation will be made to the committee. $-,, Decisions by the company with respect to heaIth and safety recommendations shall not be subject to the griev- u.*. ance and arbitration procedures of the articles of agreement (emphasis added).

~ Company perceptions, however, differ. Shell has described the joint committee policies as having “dispelled rumors and improved relations” between the adversarial parties, and having “removed the >.’:.? q, mystery” with regard to company initiative and support of workplace health and safety. Management representatives felt that these issues were best handled in the joint committee setting, and contended ~’ that union desires to use the grievance procedure were often motivated by wanting to push other issues into the grievance setting “under the guise of safety.” ,,. -, .-

enforcement of this have been very limited. More- protections and obligations provided by the act over, employees do not have a right to pursue a and the standards that apply to their workplaces court-ordered remedy independently, but must (section 8(c)(l)). rely on OSHA to negotiate a settlement or file suit. Thus, one commentator has suggested that The OSH Act also mandates OSHA to issue the implementation of this provision “has been regulations requiring employers to maintain rec- seriously flawed” (406). ords of employee exposures, to provide employees and their representatives with the opportunity This protection from discrimination covers to observe employer-conducted exposure moni- workers who refuse to engage in imminently haz- toring, and to provide access to these records. ardous work. An OSHA regulation prevents em- Employers are also required to notify employees ployers from disciplining employees for such a re- who have been or are being exposed above per- fusal if there is insufficient time to eliminate the missible limits and inform them of “the correc- danger of death or serious injury through any tive action being taken” (section 8(c)(3)). Finally, other means. In Whirlpool Corp. v. Marshall, the the Act requires that standards issued by OSHA U.S. Supreme Court unanimously upheld this reg- “shall prescribe the use of labels or other appro- ulation (307,408). This right is limited to refus- priate forms of warning as are necessary to in- ing imminent injury hazards and does not extend sure that employees are apprised of all hazards to most chronic health hazards. to which they are exposed, relevant symptoms Under a provision in the National Labor Rela- and appropriate emergency treatment, and proper tions Act, employees who refuse to perform haz- conditions and precautions of safe use and ex- ardous work may be protected from disciplinary posure” (section 6(b)(7)). action if they are acting together and in good faith Several OSHA regulations give workers and (408). Recently, however, the National Labor Re- their unions the right to obtain information from lations Board ruled that this did not apply to one employers. For example, an OSHA regulation worker who refused a hazardous assignment be- issued in 1978 requires that the employer-main- cause he was acting only by himself (328). tained Log of Injuries and Illnesses be made avail- Of great importance to effective worker par- able to workers upon request, and employers are ticipation in labor-management activities is the required to post a summary of the log each year. availability of information to workers and their The OSHA Access to Records regulation and re- unions about plant-specific hazards. The OSH Act cent rulings by the National Labor Relations has several provisions requiring that employees Board also assist unions and workers in obtain- be fully informed about health and safety issues. ing information concerning exposures and em- The act provides that OSHA issue regulations re- ployer-held medical records. Finally, as discussed quiring that employers inform employees of the earlier in this chapter, various state and local Ch. 15–incentives, /imperatives, and the Decision to Control 319 right-to-know laws and the recent OSHA regu- tered by the Swedish trade unions, both at trade lation concerning hazard communication may union schools and in worker-organized “study also aid workers and their unions in learning circles” (64,242,677). about hazards. (For further discussion of employ- As described in chapter 10, employers, unions, ee rights, see 307, 333, and 408. ) universities, and other groups in the United States have developed and administered worker train- Possible Changes in Workers’ Rights and ing and education programs. The “right to know” Collective Bargaining movement in this country (discussed earlier) re- It has been proposed that workers’ rights con- veals the desire of many workers, unions, and cerning workplace health and safety be expanded. communities for complete information about Several observers have argued, based on the ex- workplace hazards and for greater involvement perience of other countries, that health and safety in health and safety issues. In addition to the improvements could be made by enhancing work- “right to know, ” Mazzocchi has suggested that ers’ rights to full information on job hazards, re- workers be afforded a “right to act .“ This would quiring worker participation in health and safety mean that, in each workplace, a worker would be “deputized” to represent fellow workers in seek- decisions, and permitting workers to refuse all work assignments they consider to be hazardous ing outside assistance to evaluate information ob- (64,242,677). tained under the “right to know” laws and the OSHA “hazard communication” standard (345). In Sweden, for example, since 1974, every es- However, direct application in the United States tablishment with five or more employees must of the experience of other countries will be im- have at least one worker-elected health and safety paired by differing legal, cultural, economic, and steward; larger workplaces typically have a steward for each area of the workplace with one political conditions. In addition, greater employee worker designated as the chief health and safety involvement in health and safety would represent steward. The health and safety stewards have the changes in the traditional authority of management to make decisions concerning all working authority to inspect the workplace for hazards, conditions. are involved in the design of the workplace, and have the right to examine company records. They It has also been suggested that greater use of also have the authority to order, for health and collective bargaining could improve occupational safety reasons, that a particular plant operation safety and health. Collective bargaining agree- be shut down, even over the objections of man- ments can detail procedures and investments that agement. In cases of a labor and management dis- are tailored for individual plants and firms, thus pute, the operation remains shut down until a decentralizing the process of implementing health government inspector visits the workplace and and safety controls. The result, it is claimed, can resolves the issue. be health and safety improvements that are both “more efficient and more effective” than those pro- In addition, all workplaces with 50 or more em- duced by the current system of national regula- ployees must have a labor-management health and safety committee. The committee discusses tory standards (37). plans to improve working conditions, is involved There are three significant factors that limit the with the design of new facilities and worker train- effectiveness of collective bargaining in improving, and supervises the company health and safety ing occupational safety and health. First, only staff. Although the workers have a majority on about 20 million workers, or about 20 percent of these committees, most decisions are reached by the U.S. work force of 100 million, belong to la- consensus of both labor and management repre- bor unions. Moreover, the percentage of the la- sentatives. Finally, a number of different types of bor force belonging to unions has been declining training courses have been developed for work- since the 1950s. This may represent the result of ers, supervisors, and the worker health and safety social and cultural patterns and the preferences stewards. These courses are commonly adminis- of U.S. workers and managers, but it should also 320 ● Preventing Illness and Injury in the Workplace be noted that U.S. law has not always permitted or coincidentally improve worker health and safe- collective bargaining. Because of this, one labor ty. Using the authority granted by several dif- union representative on the advisory panel for this ferent statutes, both the Environmental Protec- assessment suggested that changes in labor relation Agency (EPA) and the Consumer Product tions law have health and safety implications. and Safety Commission (CPSC) regulate products that may also be workplace hazards. For exam- Second, not all unions have the kinds of exper- ple, EPA regulates the production of pesticides tise in industrial hygiene, injury prevention, or and requires notification before other chemicals occupational medicine needed to negotiate and en- are manufactured in order to protect public health force agreements on occupational health and and the environment. EPA’s regulations, in many safety. A recent survey of 14 U.S. labor unions cases, also reduce worker exposures. shows that union spending on worker health and safety ranged from about 20 cents to over $15 per For some toxic substances, in fact, workers may member. These 14 unions employed only two full- be the only significantly exposed group. In light time physicians and one part-time one, while the of their overlapping authorities, EPA and OSHA majority of these unions employ few full-time in- have considered joint regulatory proceedings. The dustrial hygienists, public health professionals, or most recent public announcement of this has in- other health and safety staff (678). volved possible regulation of methylenedianiline, Third, by definition safety and health provi- although it now appears that the agencies have sions must compete for attention and resources decided that OSHA, not EPA, will take respon- with other bargaining issues. In periods of eco- sibility for regulating workplace exposures (346, nomic downturn, workers and their unions, con- 354). cerned about maintaining wage levels and preserv- Regulation of the products purchased by busi- ing employment security, may push safety and nesses may, in some cases, be more cost-effective health issues down the list of priorities. than requiring installation of industrial hygiene controls. For example, a number of workers ex- Government Regulation posed to formaldehyde are employed in establishments where the only source of exposure is the The final factor that influences employer deci- emission of formaldehyde from products supplied sions to control hazards is an imperative: the reg- by other industries. These include apparel man- ulations and standards issued by Government ufacturers using cloth treated with formaldehyde- agencies, mainly OSHA. As detailed in chapter based resins and office workers who are exposed 12, OSHA has been empowered to issue manda- to formaldehyde from particleboard or plywood tory regulations, to conduct inspections, and to products in their offices. In these cases, standards propose penalties and require correction when it concerning formaldehyde emission rates or prod- finds violations of those regulations. OSHA reg- uct content might be both less expensive and easier ulations are often the focus of health and safety to enforce than efforts to increase general ventila- discussions because they require response and tion in numerous small establishments in dispersed compliance. Nevertheless, they are limited as a locations (206). factor in health and safety because few regulations have been promulgated and enforcement is spotty. One way to issue standards for these hazard- Any changes in this, however, will occur only if ous products used in the workplace would be co- Congress and the executive branch act to increase ordinated regulatory efforts between OSHA and OSHA resources or change standard-setting and the EPA or the CPSC . This approach holds some enforcement procedures. promise for reducing the hazards of products purchased by small businesses. However, this ap- (As part of this assessment, Mendeloff analyzed proach will have only a limited impact on hazards some of the factors that affect compliance with related to the improper use of products in the several OSHA health standards. See box T.) workplace. In addition, there are difficulties in In addition to OSHA, other Federal agencies regulating products already manufactured and in may require employer actions that either directly use. Ch. 15—Incentives, Imperatives, and the Decision to Control . 321

Box T.—Factors Affecting Worker Exposures On contract to OTA, Mendeloff analyzed OSHA inspection data from 1973 through 1979 for four substances: lead, trichloroethylene, asbestos, and silica. Mendeloff developed several different measures of employee exposure and attempted to explain observed variations in exposure using other information included in the OSHA data. His analysis was limited because, first, not all inspection results were included in the OSHA computer system. Second, the results of his analysis explain only between 5 and 25 percent of the variation in the dependent variables. Thus, the variables he was able to test do not capture all the factors that determine success in controlling hazards (301). Mendeloff found that complaint and programed inspections were equally effective in discovering employer noncompliance with OSHA standards. Mendelhoff predicted that the presence of a union at an inspected establishment would contribute to lower exposures and fewer citations. However, this was not supported by the OSHA inspection data. He did find that, other things being equal, exposures tended to be higher at plants in nonmetropolitan (rural) areas compared with those in metropolitan areas. It might be speculated that workers in nonmetropolitan areas have fewer job choices and are thus more willing to tolerate higher exposure levels. Alternatively, it could be that information about hazards and controls reaches rural areas more slowly or is adopted more reluctantly there. Size of establishment was also examined. The results indicate that both large and small establishments were less likely to have overexposures than medium-sized establishments, paralleling the situation with injuries in which medium-sized establishments appear to have higher rates. However, in many cases the differences between the various establishment sizes were not statistically significant. Moreover, interpretation of the observed differences was hampered by ambiguities in the definitions of the employee exposure variables. Mendeloff also examined the possibility that employer efforts to control hazards depended on the cost of doing so. Using information from an OSHA report, he calculated the average costs per exposed worker of complying with OSHA’s proposed lead standard. The industries with the higher average compliance costs (primary smelting, secondary smelting, battery manufacturing, brass, bronze, and copper foundries) tended to have higher employee exposures. The industries with lower average compliance costs (newspapers, commercial printing, can manufacturing, gray iron foundries), whose costs were one- tenth to one-twentieth those of the higher cost group, had a substantially lower proportion of overexposures. These results support the claim that employees tend to be protected from exposures when protection is cheap and overexposed when it is expensive. Mendeloff reviewed the results of a number of recent inspections with OSHA inspectors. For firms in violation of OSHA standards, the compliance officers estimated the costs of reaching compliance. In some cases the compliance costs were modest, but for others, the costs were quite large. This group of cases may not be a representative sample, but for most of them, the costs of compliance would be substantially larger than both the average OSHA fine and the maximum OSHA penalty for serious violations ($1,000). Profit-maximizing businesses that take actions based on the “bottom line” will invest in health and safety only to the extent that such investments minimize their costs of doing business. If compliance costs are substantially higher than expected penalties, the profit-maximizing firm will not voluntarily udertake these health and safety investments. Jones (237) examined OSHA inspection data for asbestos exposures from 1972 to 1979. She found that an increase in engineering control costs of $l00 per employee was associated with an increase in the average asbestos exposure level of 0.7 fibers per cubic centimeter, while an increase in OSHA penalties of $350 per citation was associated with a decrease in average exposure level of 1 fiber per cubic centimeter. Thus employers appear to be sensitive to costs: as the costs of control go up, employers decide not to implement controls (leading to higher worker exposure levels), while as the expected costs of OSHA penalties go up, compliance improves. However, increased penalties were also associated with employers more frequently deciding to contest the OSHA citations, again consistent with the theory that employer decisions are sensitive to costs. Jones found that an increase in the total penalties of $1,000 was associated with a 27 percent increase in the probability that the employer would contest the results of the inspection. < 322 Preventing Illness and Injury in the Workplace

SUMMARY OTA has identified a number of incentives or tion is an incentive for prevention of occupational imperatives for the implementation of control injuries, although data supporting this conclusion technologies. These include: voluntary efforts by are limited. For occupational illnesses, the eco- employers and voluntary associations, the avail- nomic incentive provided by workers’ compensa- ability of information, the desire to reduce insur- tion is reduced substantially because few cases of ance or workers’ compensation losses, fear of tort work-related illnesses enter the workers’ compen- liability actions, labor market pressures and col- sation system. This appears to be true even for lective bargaining agreements, and government well-studied occupational diseases, such as those regulations. One or all of these may influence associated with asbestos and cotton dust. employers to install and maintain appropriate By far the most common use of tort actions in controls. However, all have significant limitations. occupational safety and health are suits against Management commitment has been called the a third party—for example, against the manufac- single most important ingredient in effective turer of products purchased by the employer for health and safety programs. After being informed use in the workplace. These suits may lead to im- of or discovering the existence of job hazards, provements in working conditions, but a number some employers will take action to reduce those of factors limit their effectiveness. The first is that hazards, although the pressure of competition will suit cannot usually be brought against employers substantially limit these voluntary employer activ- because workers’ compensation laws bar employ- ities. In addition, employers and their employees ees from suing their employers in cases involv- often participate in the development of voluntary ing work-related disease and injury. Other limita- standards, which are an important source of in- tions involve recognizing and proving causation, formation in the health and safety field. But be- and identifying the responsible manufacturers. It cause they are voluntary, these standards will is not now clear how important tort liability will often have only a limited impact on worker health be as an incentive, although it has probably en- and safety. couraged the development of substitutes for asbestos. Making information about occupational hazards and their control available is a necessary first To some extent, employers maybe motivated step for many improvements in workplace health to improve employee health and safety because and safety. Providing information through re- of pressures from the labor market. However, a search and dissemination- is an important govern- slack labor market, with relatively high unem- mental activity. The increased availability of in- ployment, and other market imperfections limit formation on workplace hazards may serve as an this incentive. incentive for companies to introduce controls, but Negotiation and collective bargaining can also it is not sufficient by itself to guarantee improve- bean incentive. Passage of the OSH Act in 1970 ments in worker health and safety. coincided with an increase in the number of con- Employers may take actions to improve job tracts that included safety and health clauses, and safety in order to reduce the costs of workers’ the clauses became increasingly specific. But this compensation and property insurance. The costs is significantly limited because, first, only a small of medical treatment and lost wages paid through percentage of U.S. workers belong to labor unions. the workers’ compensation system provide a mon- Second, not all unions have the kinds of exper- etary incentive to reduce those injuries and ill- tise in industrial hygiene, injury prevention, or nesses. In addition, insurance companies and State occupational medicine needed to negotiate and en- agencies that sell workers’ compensation insur- force agreements concerning health and safety. ance usually offer ‘loss-control services” to im- Third, by definition safety and health provisions prove health and safety conditions in their client must compete for attention and resources with companies. It is likely that workers’ compensa- other bargaining issues. Many workplaces have .—

Ch. 15–incentives, Imperatives, and the Decision to Control ● 323 joint labor-management health and safety com- However, there is still controversy about the re- mittees, but the research assessing the effective- quirements of these regulations. ness of committees in improving worker health The final factor that influences employer deci- and safety is limited. sions to control hazards is an imperative: the reg- The OSH Act itself created a number of oppor- ulations and standards issued by government tunities for worker participation concerning oc- agencies, mainly OSHA. In addition, some of the cupational safety and health. For example, State regulatory actions of other Federal agencies may and local “right-to-know” laws and the OSHA require employer actions that either directly or “hazard communication” standard will provide coincidentally improve worker health and safety. workers with more information about hazards. 16. Economic Incentives, Reindustrialization, and Federal Assistance for Occupational Safety and Health and ., ...... 337 —

16 ■ Economic Incentives, Reindustrialization, and Federal Assistance for Occupational Safety and Health

Economic incentives, tax programs, and finan- ble to integrate productivity-improving invest- cial assistance have all been suggested for stimu- ments in plant and equipment with the installation lating improvements in occupational safety and of control technologies that safeguard worker health, but have been little used in the United health and well-being. Finally, the Federal Gov- States. In addition, the process of industrial ernment could also establish new programs for change can itself be harnessed as a mechanism for financing research and training activities and for improving occupational safety and health. Dur- assisting small businesses. ing a time of reindustrialization, it may be possi-

ECONOMIC INCENTIVES AND FINANCIAL ASSISTANCE

The general notion behind economic incentives a 1977 memo to President Carter from his chief is to use the economic self-interests of individuals economic adviser, his budget director, and his do- and firms to accomplish social goals. For occupa- mestic policy adviser. In both cases, union offi- tional health and safety, the aim is to set incen- cials were outraged and the injury tax reference tives so that employers, while seeking to earn was either deleted or suggested as a supplement, profits, will also prevent injuries and illnesses. not an alternative, that should be studied by an interdepartmental task force (300). The task force In theory the workers’ compensation system later rejected the injury tax approach (228). provides an economic incentive for prevention, but in practice it is limited, especially for illnesses As opposed to the regulatory system, which (see ch. 15). Another possible type of economic penalizes firms for violations of regulations, an incentive would substitute an injury tax or an ex- injury tax system would levy direct financial pen- posure tax for traditional regulatory standards. alties on firms for each injury. Firms would be A third possibility involves the use of various gov- free to choose the least costly methods of acci- ernmental tax policies and financial assistance dent prevention. Under certain assumptions, this programs. would be the most “cost effective” way to reduce the number of injuries. Moreover, an injury tax Injury/Exposure Taxes is appealing because it is directly related to safety outcomes. (Similarly, it has been suggested that The idea an “injury tax” is generally well- effluent taxes or emission fees be used in the area received among economists as a substitute for of environmental protection. See, for example, health and safety regulations (see, for example, 251. ) 330,425,445,446). Although the idea has been criticized as politically infeasible (37,300), it has offi- The advocates of injury taxes believe them to cially surfaced at least twice—in a draft 1976 re- be better than regulatory standards, which are port of the Council of Economic Advisers and in considered inflexible and unnecessarily uniform

327 328 ● Preventing Illness and Injury in the Workplace across firms and industries, not necessarily related ers their usual wages if they must be removed to workplace or work-force characteristics that from lead-contaminated work areas have some cause injuries, and not necessarily the least costly characteristics of an illness/exposure tax. ) way to reduce injuries (445,446). In addition, an An injury tax, even if limited to cases of acute injury tax, if set at an appropriate level, might trauma, would encounter serious difficulties. The provide employers with a stronger incentive to first would be in setting an appropriate level for prevent injuries than the current Occupational the injury tax. Smith (445) estimated in 1970 Safety and Health Administration (OSHA) pro- dollars that a tax of $1,000 per disabling injury grams. It could also reward employers for efforts would lead to a decline in injury rates of about to go beyond current regulatory standards, as well 5 percent, while a $2,000 tax per injury would be as for conducting research on improved controls. associated with a 10 percent decline. Adjusting Smith (445,446) concluded that firms do re- his estimates to account for inflation since 1970 spond to financial incentives and estimated the yields injury taxes in the range of $2,500 to $5,000 effect that various levels of injury tax would have (in 1983 dollars) per lost-workday injury to on injury rates. To be effective, the taxes would achieve a 5 to 10 percent injury rate decline.1 have to be fairly large. Using 1970 data on in- Furthermore, if a tax for deaths or permanent juries, Smith estimated that a tax of $500 per in- disabilities were set equal to the lost earnings of jury would lead to a 2.2 to 3.2 percent reduction the killed or disabled worker, it would be ex- in the disabling injury rate (injuries involving lost tremely high-ranging from tens of thousands to work time as defined by standard Z-16 of the hundreds of thousands of dollars. These values, American National Standards Institute). A tax of large as they are, do not include the psychic costs, $1,000 would reduce injuries 4.4 to 6.2 percent, the pain and suffering associated with such in- while one of $2,000 would reduce injuries 8.8 to juries. To maximize the efficiency of the tax sys- 12.5 percent (446). tem it would be necessary to prohibit firms from Although Smith suggests moderate fines for insuring themselves against these large losses to each injury as a replacement for safety regulation, ensure that firms would pay the full tax. But such he does not extend that approach to health regu- large penalties would generate considerable po- lation. Because occupational diseases often man- litical opposition (64). ifest themselves only after a latent period, there A second problem would be ensuring the ac- would be considerable difficulty in determining curacy of the injury records on which any tax which of several employers was responsible for would be based. With a direct financial penalty the disease (446). Moreover, as discussed in chap- for each injury, employers have an incentive not ters 2, 3, and 15, there are considerable difficulties only to prevent injuries but also to underreport in distinguishing job-related diseases from non- them. (Ch. 2 and Working Paper #l discuss the occupational ones. controversy about the accuracy of current em- Taxing hazardous exposure levels rather than ployer-maintained records. ) Independent, firm- illnesses might be one possible way to affect job- by-firm audits to guarantee the accuracy of these related diseases. This could work like an effluent data would be quite expensive. Moreover, assess- or emission tax, under which firms pay a fixed ment of the results of an injury tax would be dif- amount for each additional unit of pollution they ficult because any declines in injury rates follow- add to the water or air. Nichols and Zeckhauser ing implementation might represent decreased (330) have suggested such a tax on occupational reporting, not increased prevention (64). In some noise exposure. But Smith points out that enforc- cases there would also be problems in distinguishing an exposure tax approach would require a “monumental inspection and monitoring pro- ‘This approximate adjustment for the effects of inflation is based gram” (446). The attending administrative costs on the two-and-a-half-fold increase in average hourly earnings from would probably offset the advantages of such a 1970 to 1983 and is consistent with the increases during that period in the most commonly used price indexes (the Consumer Price In- system. (To a limited extent, provisions of the dex, Producer Price Index, and Implicit Price Deflator for the Gross OSHA lead standard that require paying work- National Product). Ch. 16–Economic Incentives, Reindustrialization, and Federal Assistance for Occupational Safety and Health ● 329 ing occupational (and taxable) injuries from non- safer jobs than those in industries with high pre- occupational ones. These would include many vention costs. Job risks would thus remain un- sprains and strains, as well as many injuries to equally distributed. This, however, runs counter the back. to a commitment to the goal of equal protection for all (277). Third, an injury tax might increase the level of controversy in this field. Boden and Wegman (64) argue that it would be politically feasible only for Tax Programs and Financial Assistance the less severe injuries. Mendeloff (300) suggests When the topic of economic incentives is raised, that even a tax of $250 per injury might be more most employers think of changing the taxes they expensive to firms than current OSHA safety are most familiar with—the business income tax standards. In theory, an injury tax should apply system. Congress could modify the structure of to all injuries. But in practice, there would be con- business taxes to encourage investment in health troversies concerning application of an injury tax and safety control technologies or could assist in cases of claimed “employee misconduct. ” businesses in financing such investments. Fourth, the operation of such a system would A congressional decision to modify the tax be relatively invisible to workers. Employers structure or provide financial assistance can be would calculate the number of injuries and send thought of as providing some level of “social fund- the Government a check (64). If an injury tax sys- ing” for investments in occupational safety and tem replaced existing standards, and suspended health. The general rationale for this is to reduce unannounced inspections and workers’ rights to the costs of health and safety investments, thus request “complaint inspections, ” it would change encouraging firms to undertake them. Burstein dramatically the worker and governmental roles (531) states that tax policy can be used to lower in solving health and safety problems. business taxes on certain kinds of investments, Fifth, an injury tax might create an incentive thereby increasing the returns to business “to re- for larger firms to subcontract dangerous jobs to flect the external benefits provided by the activ- smaller, less financially solvent firms that lack the ity.” These external economic benefits, such as resources to prevent injuries as well as to pay the reductions in the cost of medical care due to im- injury tax. An injury tax might also lead to changes provements in employee health, are ones that in- in employment policies, for example, in hiring and dividual firms would not ordinarily receive be- firing decisions. The effectiveness of these policies cause companies do not shoulder all the costs of in reducing injuries has been questioned (64) (see ill health. Some of those costs are borne by em- also ch. 4), and such a response by employers may ployees, by other insurance policyholders, and by have other social and economic implications that the Government. Tax policy might be used to would need to be considered. reward the individual firm for actions that reduce these “social costs, ” Finally, injury taxes have been criticized on ethical grounds. It has been asserted that such a Many believe that society ought to assist busi- tax would be a “license to maim” or a “license to nesses in meeting certain social goals, such as re- kill.” By providing a system of “taxes” and “li- ducing pollution or improving worker health and censes, ” society would seem to be saying that it safety. This is especially true when, for various is all right for a certain number of occupational reasons, society changes the goals by, for exam- deaths and injuries to occur. But if an injury tax ple, increasing the stringency of applicable regu- reduced injuries while preserving other social lations. Thus these tax and financial assistance values, it would probably be considered an ad- programs lead to subsidies for businesses, but this vance because fewer workers were being harmed, may be appropriate to reach socially valued goals of environmental and worker protection. The operation of an injury tax would also lead to variations in the level of protection that de- Four kinds of programs are of interest: invest- pended on the costs of prevention. Workers in in- ment tax credits, accelerated depreciation, Gov- dustries facing low prevention costs would have ernment loan programs, and direct subsidies. All 330 ● Preventing Illness and Injury in the Workplace have already been used in the area of environ- argued, the Government has a role in providing mental protection and loan programs have been such financing. used, to a limited extent, for occupational safety Financial assistance can take a number of forms, and health. including Government guarantees of private An investment tax credit allows a business to loans, Government interest subsidies for private apply a certain percentage of the purchase price loans, direct Government loans (often at reduced of a capital good directly against the taxes owed interest rates), and the use of tax-exempt financ- by the firm, in addition to the normal deprecia- ing by private firms. In loan guarantee programs, tion of the investment over time. When firms have the Government promises to repay a private loan been allowed such credits, some of the invest- if the borrower defaults. This can be combined ments may have included controls for workplace with interest subsidies, under which the Govern- hazards or pollution. However, Federal tax pol- ment pays a share of the interest costs. Under di- icies have not created an investment tax credit rect loan programs, the Government acts as if it especially for employee health and safety invest- were a bank and loans the money directly at mar- ments, although such credits have been proposed. ket interest rates or at a lower, subsidized rate. Tax-exempt financing allows private business to Accelerated depreciation or rapid amortization take advantage of the lower interest rates on of investments is a second tax subsidy mechanism. bonds issued by States and local governments. This permits businesses to write off the costs of an investment more rapidly than the normal de- When Congress passed the Occupational Safety preciation rules would allow. Federal tax law has and Health (OSH) Act in 1970 it concluded that allowed 5-year amortization of pollution control the burdens of compliance would often dispropor- equipment (651). This accelerated depreciation for tionately affect small businesses. Rather than ex- assets that will last longer than 5 years has been empt such firms from the requirements of the act, permitted only for investments that lead to pol- Congress chose to amend section 7(b) of the Small lution control without also creating “significant” Business Act to allow the Small Business Admin- changes in other aspects of the facility. It has not, istration (SBA) to make or guarantee loans for however, been available for investments to pro- OSHA compliance expenditures. The require- tect worker health and safety (31). ments were, first, that the expenditures must be to comply with OSHA regulations and, second, A proposal to allow accelerated depreciation that the small business was “likely to suffer sub- for OSHA-mandated investments moved through stantial economic injury” without such assistance the legislative process to the conference commit- (OSH Act). tee stage in 1978. A compromise reached in conference was to request a Treasury Department fea- From September 1971 to August 1981, the SBA sibility study of such a change in the tax law (228). processed 261 OSHA-related loans for $72.8 mil- (The major conclusions of that study are discussed lion—about 26 loans per year, each averaging later in this section.) about $280,000. In fiscal year 1981, 9 loans were made for a total of $7.1 million. A third kind of Government program is to provide financial assistance, either directly or in- These loans constituted a very small part of directly. Financial assistance is defined here to SBA loans. Although exact figures are not readily mean programs that provide loans or other types available, in recent years the agency has been of financing (such as bonds) to assist businesses making or guaranteeing between 20,000 and in paying for health and safety investments. Busi- 30,000 loans annually for between $2.5 billion to nesses must repay the loans or bonds, but the cost $4 billion. It is possible that some firms have used of these obligations is often partially subsidized regular SBA loans to finance OSHA compliance and they provide a source of capital for invest- expenditures in addition to other investments, The ments that may not return a profit to the firm. dedicated OSHA loans, however, usually carried Private lenders are often reluctant to loan money a lower interest rate than the regular loan pro- for such “nonproductive” investments so, it is gram (233). —— ——

Ch. 16—Economic Incentives, Reindustrialization, and Federal Assistance for Occupational Safety and Health . 331

The OSHA loans program, as well as the SBA plant operations, not extra costs due to pollution loans for pollution control, were eliminated by control requirements. the Omnibus Budget Reconciliation Act of 1981 (Public Law 97-35, Section 1905). It is not clear Despite these restrictions, it has been estimated how effective this loan program was in improving that about 48 percent of all capital spending for health and safety in the handful of firms that re- pollution control has been financed with tax- ceived loans. In 1979, the General Accounting Of- exempt bonds (538). However, the IRS has not fice (GAO) criticized the program, in part because allowed tax-free financing for investments that few loans had been made. In addition, GAO con- protect employees from toxic substance exposure cluded, after an examination of SBA files concern- in addition to preventing environmental damage, ing a sample of loans, that it was not clear that apparently on the belief that the company would the loans had been granted only to businesses that have invested in worker protection in any case needed the loans or that the use of the loans ac- as part of a prudent personnel policy (651). In the tually resulted in the elimination of workplace last few years, Congress has also enacted several hazards (509). restrictions on the amount of tax-free financing that can be issued by State and local governments Tax-exempt bonds, issued by State or local gov- to private businesses. ernments to provide financing for private firms, are backed by the credit of the borrowing firm, Finally, the Government could assist businesses the revenue from the project financed, or the value by giving them direct subsidies or grants. Econo- of the facility. Defaults are borne by those who mists generally prefer direct subsidies to indirect hold the bonds, not the State or local government tax subsidies or loan programs because they create that issued them. Because the interest on such fewer market distortions, are simpler for the IRS bonds is exempt from Federal income taxes, to administer, and can often be more cost effec- lenders are willing to accept lower interest rates. tive. Direct grants do, however, enter directly into Thus, private firms that use these bonds are able the appropriations process and thus may be more to finance investments at interest rates lower than visible and hence more difficult to legislate. More- they would otherwise pay. In 1968, Congress over, a firm will probably incur a greater paper- limited the use of bonds that exceeded a certain work burden while applying for a direct grant size to those that finance pollution control and than it would with an indirect tax subsidy. certain public facilities (such as airports, conven- tion centers, parking garages, sports stadiums, As noted earlier, in the Revenue Act of 1978, Congress requested the Department of the Treas- etc. ) (492). ury to study the feasibility of tax incentives for The Internal Revenue Service (IRS) has had dif- occupational health and safety spending. The re- ficulty in administering this provision of the tax port, written by the Office of Tax Analysis of the code. More than 6 years passed before it published Treasury Department and published in January temporary regulations, and they encountered dif- 1981, was very critical of tax subsidies for OSHA ficulties in ensuring that this financing is limited and Mine Safety and Health Administration to pollution control and is not used by firms for (MSHA) compliance. It noted that such a subsidy “productive investments” (651). The IRS has program would be difficult to administer, mostly tended to limit “pollution control” to certain “end- because of problems in distinguishing health and of-pipe” technologies that have no productive safety expenditures from normal business costs. value, such as effluent water treatment. More fun- Second, the Treasury Department analysis ex- damental process changes that also reduce the pressed concern that subsidies for capital costs amount of pollution have generally not qualified only would encourage firms to adopt unneces- for tax-exempt financing (538). The IRS has also sarily capital-intensive compliance methods. disallowed tax-exempt financing for containment Third, they criticized special investment tax cred- devices for nuclear power plants, as well as spend- its, accelerated depreciation, and tax-exempt ing for the disposal of hazardous wastes, because financing on several more technical grounds, in- these were considered to be normal expenses for cluding the differential treatment of assets with 332 ● Preventing Illness and Injury in the Workplace

different lifetimes, the differential advantage in- implementation of controls, assist businesses in curred by profitable (as opposed to unprofitable) compliance, and possibly reduce the controversy firms and by capital-intensive (versus labor-inten- of regulatory proceedings because of the availabil- sive) companies, and the large benefit that tax- ity of sources of finance. However, there are sev- exempt financing gives to upper-income bond- eral disadvantages. holders (651). First, they would represent either a reduction The Interagency Task Force had previously ex- in Federal tax revenues or an increase in budget amined the issue of direct economic incentives. outlays. Second, a tax incentive program would In contrast to the negative comments by the Treas- also tend to increase the complexity of the tax law, ury Department report, they recommended, if while a direct assistance program would require general economic conditions permitted, both the personnel and resources for program administra- extension of the investment tax credit to noncap- tion. Third, these programs will often provide fi- ital expenditures for health and safety investments nancial benefits to firms that would have installed and the creation of a program of direct financial controls even in the absence of a subsidy program. assistance. This should, they suggested, take the Fourth, there would be difficulties in dividing the form of a direct subsidy rather than a tax credit. purchase price of equipment between features that The subsidy would be limited to high-hazard firms are health and safety controls and those that are within hazardous industries, and would only ap- normally part of the equipment for purely pro- ply to the firm’s health and safety spending that ductive reasons. Finally, each of these programs represented an increase over their spending in a has its own limitations, and would have differ- baseline year. The program could be administered ing effects on other aspects of business investment either through the Treasury Department or through behavior, as well as on the distribution of wealth, the regulatory agencies (OSHA and MSHA) (228). income, and the burden of income taxes. All of To date, these recommendations concerning fi- these would need to be considered before estab- nancial assistance have not been acted upon. lishing any program. OTA concludes that the use of tax incentives and financial assistance programs might spur the

REINDUSTRIALIZATION AND OCCUPATIONAL SAFETY AND HEALTH

Some recent discussions of the U.S. economy beneficial or an adverse impact on worker health have included references to “reindustrialization” and safety. Second, if the Federal Government and “industrial policy. ” Commentators have gen- stimulates economic revitalization through tax, erally focused on the international competi- expenditure, or financing programs, it may be tiveness of certain U.S. industries—principally in advantageous to incorporate health and safety manufacturing. Little of the discussion has been considerations into those policies. OTA, in this about occupational health and safety policy or assessment, is not advocating any form of indus- even about regulation. But many proponents talk trial policy. Indeed, industrial policies and more about using economic incentives—including tax general economic policies are areas beyond the law changes and Government financing—to im- scope of this report. But there do appear to be plement the new policies. Moreover, some have connections between these policies and possibil- suggested changes in health and safety regulations ities for improving the implementation of control in order to facilitate business reinvestment and technologies. plant modernization. Reindustrialization and Industrial Policy OTA considers the reindustrialization debate to be relevant to this report for two reasons. First, The terms reindustrialization and industrial pol- reindustrialization policies might have either a icy often have meaning only in the eye of the — -————

● Ch. 16–Economic Incentives, ReindustrializafionJ and Federal Assistance for Occupational Safety and Health 333 beholder. An enormous range of suggestions con- and efficiency, “ including changes in antitrust and cerning U.S. economic productivity and interna- trade policy, as well as programs to assist dislo- tional competitiveness have been made under cated workers, and changes in regulatory policy these labels (528). Some think the problem is that that would assist businesses in achieving economic the Government has interfered too much in the growth (493). market through existing tax and regulatory pol- A third strategy for industrial policy would in- icies. The solution they propose is to reduce the volve the creation of new institutions, among size of Government and limit its intervention to which could be an information and/or consen- providing certain “public goods,” such as national sus development agency, an executive branch co- defense. Others think that economic revitaliza- ordinating agency, and a Government financial tion can best be achieved by shifting Government institution. An information/consensus agency policies. These commentators want policies that could gather, synthesize, and disseminate infor- increase incentives to work and save and that re- mation on American industry, including assess- duce incentives to consume. Specifically, they ments of U.S. Government policies and of for- have advocated across-the-board changes in busi- eign activities. Several proposals would also ness tax laws to encourage investment. Some of include creation of a council, composed of rep- these suggestions were legislated in the Economic resentatives from business, labor, and Govern- Recovery Tax Act of 1981. ment, to develop a consensus on the goals of an Still another group wants to see explicit Fed- industrial policy. An executive branch coordinat- eral policies that will encourage the growth of ing agency could attempt to coordinate the pol- “sunrise” industries and that will ameliorate or icies of the Federal Government toward a particu- prevent the decline of “sunset” industries. Critics lar industry or group of industries, in order to of this position question whether the Government encourage growth and competitiveness. The pro- can correctly identify sunrise and sunset indus- posed Government financial institution would be tries and some wonder if this is an appropriate a national industrial development bank or sev- role for Government at all. (See 493 and 528 for eral regional development banks, often modeled a discussion of these views. ) after the Depression-era Reconstruction Finance Corporation (493,527). All three groups are attempting to create conditions that will lead to economic revitalization, As noted, environmental regulation and work- but generally only the third group is advocating place health and safety regulation have received selective or targeted industrial policies. These little attention in this debate. Magaziner and Reich analysts differ in their explanations of what has (283), for example, devote only one paragraph to gone wrong and in their prescriptions for new pol- the issue. In it they call for regulations that apply icies and institutional arrangements. (Some of the to “emerging” industries that are “harmonized as leading advocates of industrial policies include far as possible with the needs and requirements Reich (383,384), Magaziner and Reich (283), of other nations. “ “Declining” industries, on the Rohatyn (395), Bluestone and Harrison (62), and other hand, “should only be required to meet Etzioni (167). For contrasting views, see, for ex- standards that are appropriate to the remaining ample, Economic Report of the President (169) useful life of the industry. ” Bluestone and Har- and Schultze (426). ) rison (62) note that reindustrialization should be directed toward several goals, including the cre- The Congressional Budget Office (CBO) has ation of safer work environments, but they do not outlined three major strategies that have been pro- elaborate on this point. posed as alternative industrial policies. The first is to work with current policy instruments and Some discussions of industrial policy have, rely on economic recovery and private-market ad- however, advocated the relaxation of environ- justments to solve existing problems. A second mental and occupational safety and health regu- strategy is to “modernize existing policies . . . that lations as part of a plan for industrial modern- may now have become impediments to growth ization. For example, Etzioni (166) has suggested 334 ● Preventing Illness and Injury in the Workplace that the United States must choose whether rein- In addition to taxation and spending policies, dustrialization or improvements in the “quality Congress and the executive branch have already of life” will be made the Nation’s “top priority. ” created programs that might be considered forms These discussions assume that Federal regulations of industrial policy and have considered others. have seriously hampered the growth of the U.S. For example, there are already a large number of economy and improvements in productivity and Federal programs that provide grants, loans, loan international competitiveness. However, as dis- guarantees, and economic assistance for various cussed later in this chapter, the adverse impact purposes (512,529). on productivity of occupational health and safety The Steel Tripartite Advisory Committee, regulation is actually fairly small. Moreover, in appointed by President Carter, consisted of rep- some cases OSHA regulation has played a role resentatives from Government, labor, and busi- in inducing or facilitating several industrial inno- ness who discussed issues concerning moderniz- vations that improved both health and safety, as ation of the U.S. steel industry. Among other well as productivity. recommendations, they suggested delays in sev- OTA analyzed the role of technology in inter- eral environmental protection requirements in or- national competitiveness in a 1980 assessment on der to facilitate plant modernization (458). In technology and the steel industry and in a 1981 1981, Congress enacted the Steel Compliance Act, report comparing the competitiveness of three which postponed certain deadlines of the Clean industries—steel, electronics, and automobiles. Air Act for the steel industry. These reports examined the general issues of tech- Finally, a number of bills have been introduced nology and productivity in these industries, in- in recent sessions of Congress concerning indus- cluding the effects of regulatory policies. In the trial policy (see 534). Hearings have been held on case of steel, OTA found that several major new some of the topics addressed by these bills (513, steelmaking processes are not only more efficient, 515,525,534,549). but also create less pollution. Moreover, modest technological improvements have resulted from the “push” provided by health and safety, and by Reindustrialization and Occupational environmental regulations, including improved Health and Safety emissions controls and better door seals for coke ovens (538). More generally, the 1980 report out- The continuous process of industrial change (in- lined a number of policy options. cluding the replacement of plant and equipment) can lead to safer and more healthful workplaces. The most critical policy option may be that of In fact, a large portion of the improvements in a governmental steel industry sector policy, that worker health and safety during this century may is, for a coherent set of specific policies designed to achieve prescribed goals. , . . The lack of a sec- not have been the result of conscious decisions tor policy and the designation of a lead agency to add controls to existing processes, but may to implement such a policy has led to policies that have occurred coincidentally as new technologies, often conflict with one another, create an adver- new processes, and new industries were intro- sarial relationship between Government and induced. Unfortunately, it is extremely difficult to dustry, and fail to address critical issues. gather definitive data on this question. In its comparison of the competitiveness of Industrial and other policies that facilitate the steel, electronics, and automobiles, OTA outlined process of industrial change might simultaneously two prerequisites for industrial policy (543): improve occupational safety and health. More- over, regulations can have a favorable impact on mechanisms for reaching agreement on objectives that are acceptable to Government and the productive efficiency of an industry either be- various interest groups; and cause they directly spur innovations and changes improved analytical capability on the part of or because they provide an opportunity to change Government agencies concerned with eco- productive aspects of plant operations. But there nomic efficiency and competitiveness. is some danger that combining policies that are Ch. 16–Economic Incentives, Reindustrialization, and federal Assistance for Occupational Safety and Health 335 designed to improve productivity with those that may also present an opportunity to improve address employee health and safety will lead to worker health and safety. The various tax incen- an emphasis on the former at the expense of the tives, loan guarantees, and other subsidies that latter. have been suggested to improve industrial competitiveness might also be used as incentives for Industrial change does not automatically im- health and safety improvements. Such policies can prove worker health and safety, nor are new thus make compliance with health and safety reg- plants necessarily safer than old ones. What is true ulations easier. In addition, if desired, the oppor- is that it is generally cheaper and more effective tunity presented by reindustrialization could be to control any given health and safety hazard used to achieve greater levels of protection. when constructing new plant and equipment than it is to retrofit existing plant and equipment (see It has been argued that health and safety regu- box U). lation has hampered economic productivity by requiring expenditures for control technologies that By stimulating changes in plant and equipment are “nonproductive. ” Money spent for controls for productivity reasons, new industrial policies could have been invested in improving plant pro-

Box U.—The Costs of Add-On and New Process Controls It is widely believed, indeed it is pretty much common sense, that controls are cheaper and more effective when designed into new plants than when added or retrofitted. Nearly every control that could be included as a retrofit could also be included in the design stage. There is no reason to believe that it would cost more to purchase a control for a new plant of the same size with the same hazard than it would be to buy the control for an older plant, Second, in designing the new plant, the architects, designers, and engineers have a greater opportunity to make sure that everything “fits together.” For example, ventilation systems can be designed efficiently rather than having to wind their way around existing equipment, building structures, or other duct work. Work stations and tasks can be designed with ergonomics in mind (see ch. 7) to enhance worker productivity and to be less stressful. Thus the design can improve a number of different features, including employee health and safety. Third, installing retrofit controls always involves disruptions-cutting through existing equipment, temporarily closing down portions of the plant (or paying overtime rates to have the work done on weekends or at night), or taking machinery apart-that can be very costly. Fourth, retrofit controls can easily outlive the rest of the plant and equipment, especially in older facilities. Thus the plant’s life may end before the firm has reaped the full life of the control devices. For some devices there will be a salvage market, but for many there will not. OTA has found several examples to support this reasoning. Rollover protection for mining vehicles is more expensive when retrofitted than when purchased on new equipment (248). In the OSHA vinyl chloride rulemaking hearings, the Firestone Plastics Co. estimated that reaching a 100 parts per million standard would cost 34 percent more in a 25-year-old plant than in an 8-year-old one; meeting a 10 parts per million standard would cost twice as much in the older plant (38). Swedish research in the 1960s led to “designed-in” noise control for one company that cost only a fraction of what retrofit controls would have cost (228). Direct comparisons of the costs of control in new plants versus retrofitting old plants are often difficult, however, because other relevant variables change. For example, the new plant may be larger or have a different production process. A comparison of the experience of two plants in controlling radiation exposures illustrates this. Retrofitting an old uranium processing plant to meet new, more stringent radiation exposure standards cost approximately 30 percent as much as the original total cost of the plant. Later a new plant was built to meet the same radiation standards. The control costs for this plant amounted to only 8 percent of the new construction costs. But there were confounding variables: the new plant was larger and had at least one significantly different type of process equipment (97). 336 ● Preventing Illness and Injury in the Workplace

ductivity. In the aggregate, however, this diver- of the controls that were applied to reduce vinyl sion is relatively small. Employee health and chloride exposures also increased production effi- safety regulations have been estimated as having ciency. (See box N in ch. 12. ) led to a 0.4 percent decline in traditional measures of productivity (137). Of course, these meas- Ruttenberg’s case study of the OSHA cotton urements do not capture any of the health and dust standard, commissioned for this assessment, safety benefits of regulation (see additional dis- reveals a similar phenomenon. Although a direct cussion in ch. 13). cause-and-effect relationship is difficult to prove, it seems clear that the recent modernization of the But compliance with regulations can also pro- American textile industry was at least accelerated vide an opportunity to make changes that im- by the OSHA cotton dust standard. Compliance prove plant productivity. As one group of an- with the standard involved, in large part, the in- alysts has noted, “[b]ecause it is less expensive and stallation of new and more productive capital disruptive to make multiple changes simultane- equipment. Some “nonproductive” investments ously, rather than individually, businessmen nat- (such as for additional ventilation) have been re- urally take the opportunity of regulation to intro- quired, but process and equipment changes induce other improvements” (31). (See 228 for sev- cluded in new designs have substantially lowered eral examples for which productivity improve- cotton dust levels. This new equipment is also ments occurred at the same time as health and more productive because it consolidates several safety improvements. ) previously separate processes, reduces energy con- According to several studies, regulation can sumption, operates faster, and produces cloth of stimulate new research or, more likely, speed up improved quality. Although the new technology the tempo of existing research. One group of re- has some limitations, the textile industry has been searchers reports that 33 percent of their study’s able to raise productivity and improve worker respondents indicated process improvements spurred health and safety at the same time (413). by regulatory changes (231), while two other A recent report to OSHA concerning the costs studies reported similar conclusions (70,101). In of complying with the lead standard discusses the a five-country comparison of Government activ- potential for new technology simultaneously re- ities that affected innovation, the researchers con- ducing worker exposures, reducing the costs of cluded that regulatory requirements concerning controlling exposures, and improving productiv- environmental protection “may be more impor- ity in a primary lead smelter (see box V). It must tant” for inducing innovation than other programs be noted that this technology has not yet been im- that were designed explicitly to influence the in- plemented on a large scale and there are uncer- novation process (29). tainties about its adoption. But what is not seri- Ruttenberg (412) has pointed out a number of ously questioned is that process redesign is the instances in which regulation has been a stimulus most effective way of achieving both productivity for new markets, new jobs, and basic product and and employee health and safety goals. process innovation. “To a surprising degree, reg- As previously noted, industrial policies would ulation is the mother of invention, ” she noted. require not only agreement among the affected This occurs because in redesigning products and parties, but also analytical capability. The Na- processes to comply with health, safety, and envi- tional Institute for Occupational Safety and ronmental regulations, companies often funda- Health (NIOSH) and the Environmental Protec- mentally redesign the product or process through tion Agency have developed a methodology for the use of new technology. Second, the existence industry-specific research planning (220). The gen- of Government standards creates “assured mar- eral objectives of this work were to improve the kets” for the results of the research efforts. coordination of environmental and occupational In the OSHA arena, the stimulus of the vinyl regulations and to identify health and environ- chloride regulation accelerated a then-developing mental problems and solutions at an early stage. improvement in polymerization technology. Some Although the focus of this project was research — -..

Ch. 16–Economic Incentives, Reindustrialization, and Federal Assistance for Occupational Safety and Health “ 337

Box V.—Reducing Lead Exposures: Add-on Controls v. New Processes The OSHA lead standard sets a limit for oc- would, and the new machinery also achieves cupational lead exposures at 50 micrograms of some savings in labor and materials costs and an lead per cubic meter of air (50 micrograms/m3). increase in income from the sale of sulfuric acid In a study done for OSHA, Charles River Asso- (a byproduct of lead smelting). But at the same ciates (103) (discussed by Gobel, Hattis, et al. time, the capital costs of new production machin- (184)) examined work exposures and controls in ery are higher (table 16-1). a primary lead smelter with an annual capacity of 225,000 tons. The highest exposures, consid- The difference in costs might be reduced by current tax preferences for investments. Invest- ering both the number of workers exposed and ment tax credits and rules concerning deprecia- the exposure levels, occurred in two departments-the sinter plant and the blast furnace tion for capital investments might offset mom of the costs of new area. The Charles River Associates report com- machinery than of add-on con- pared the costs and emission reductions expected trols, but the quantitative impact of those pref- from using each of two methods of moving toward erences was not calculated. Moreover, changes compliance with the lead standard. in the tax laws to encourage health and safety investments might have an additional effect. The first method considered was conventional “add-on” or “end-of-pipe” control technology. If a new smelter is to be constructed, both None of the manufacturing process would be health and economic considerations would favor the new Machinery. Its construction cost and the changed, and no new production equipment cost of a would be installed. Instead, enclosure and ven- conventional smelter would be similar, tilation systems would be added to separate and its reduced emissions and reduced operat- workers from airborne lead or to dilute it to ing costs would sway the decision. But install- acceptable levels. The second method considered ing the new process in an existing smelter, given a number of new technologies, which involved the economic factors, would require a card new production machinery, to smelt lead. Install- weighing of tax advantages and the other capi- ing new production machinery would reduce tal and operating costs. worker exposures more than add-on controls 338 ● Preventing Illness and Injury in the Workplace planning for two agencies, the general idea could health and safety, industrial policies, and rein- be extended to other Federal financial and regu- dustrialization. First, if funds or tax incentives are latory policies. In particular, the methodology fol- created for the building or rebuilding of indus- lows the industry breakdown used in the Com- try, applications for those benefits might be re- merce Department’s annual report, U.S. Industrial quired to include a discussion of methods to be Outlook, and develops a series of indicators that used to control expected health and safety haz- could be used to coordinate Federal policies. ards. These funds could be extended to expenditures for control technologies to reduce those A Federal Interagency Task Force on Work- hazards. Second, companies receiving reindustri- place Safety and Health recommended in 1978 alization assistance might be required to design that OSHA take steps to identify hazards for health and safety into their new plant and equip- which engineering controls could be installed ment, either to meet existing health and safety along with the normal replacement of plant and standards, or to achieve lower exposure levels or equipment (228). Both types of analytical work safer processes. might be usefully considered in developing industrial policies. Two other options consider the relationship between OSHA regulatory actions and reindustri- OSHA regulations can also be designed to con- alization. Because of the potential for improving sider the capital and investment cycles of plants health and safety, as well as productivity, dur- and firms. To some extent, the use of delayed im- ing the process of modernization, OSHA regula- plementation schedules takes these concerns into tory actions could consider explicitly the capital account. Industries then have some flexibility con- and investment cycles of plants, firms, and indus- cerning the timing of engineering controls. Feasi- tries. Information could be developed concern- bility is an important consideration in such cases: ing the health, safety, investment, and produc- something that may not be feasible in 6 months tivity needs of various industries. In particular, may be feasible ins years. One suggestion, made if studies show that an industry is going to make in the OSHA proceeding concerning noise expo- major changes to improve productivity, OSHA sure rulemaking, is the possibility of different per- might consider delaying the required attainment missible exposure limits for new and old plants. of a standard through engineering means until the Such “grandfathering” may ease the compliance modernization is undertaken. Alternatively, burden in old plants but needs to be done cau- OSHA regulations might be used to spur the de- tiously lest it create an unwanted incentive for velopment of new technologies and accelerate the continuing the operation of inefficient, older process of industrial change. The history of plants. In addition, it tends to create two classes OSHA’s vinyl chloride and cotton dust regula- of protection, with workers in new plants being tions shows that, at least in some cases, employer protected more than workers in older facilities. efforts to comply with health and safety stand- As described in chapter 1, this assessment sug- ards can also be associated with productivity im- gests several options related to occupational provements.

FEDERAL AID FOR RESEARCH AND INFORMATION DISSEMINATION

Creation of an Occupational cently. Funding has been reduced for two areas Safety and Health Fund in particular—education and training programs, and research on workplace illnesses and injuries. The advisory panel to this assessment expressed One way to provide for more stable and enhanced concern about the large swings in occupational funding would be to establish an Occupational safety and health policy that have occurred re- Safety and Health Fund. This fund might also pro- Ch. 16—Economic incentives, Reindustrialization, and Federal Assistance for Occupational Safety and Health ● 339 vide a focus for the enhanced efforts in control of cotton dust. This project is funded by both technology research, education and training, and Government and industry, with oversight and di- information dissemination. rection provided by a group of labor, management, and Government officials. There have also The Work Environment Fund of Sweden offers been jointly administered research efforts and a possible model for a U.S. fund (468). Sweden, training programs that have resulted from collec- partly because of its commitment to providing ex- tive bargaining. tensive education and training, as well as funding for research on hazards and controls, is viewed Congress could consider several possible admin- by many as an international leader in occupa- istrative arrangements if it created a fund. It could tional safety and health. Two organizations are follow the Swedish model by creating a tripar- particularly responsible for these activities. The tite board to administer this fund, or it could del- first is the Work Environment Fund itself, estab- egate administrative responsibilities to NIOSH or lished in 1972, which is funded by a payroll tax OSHA. It could create this fund and its research on employers of 0.155 percent. The fund is admin- and training projects to exist alongside the existing istered, in typically Swedish fashion, by a tripar- projects and arrangements at OSHA and NIOSH, tite board composed of representatives of man- or it could consolidate with this fund all existing agement, labor, and Government. research and training, including NIOSH extramural research grants, NIOSH training grants, In its first 10 years, over 1,800 research and OSHA New Directions grants, and OSHA-funded development projects and about 1,500 for train- consultation. ing and information were funded. The research and development grants have included funding Financing could be through a payroll tax on of epidemiological and toxicological studies, employers or through a tax or surcharge based measurement techniques, and development of on the level of workers’ compensation premiums control techniques for a variety of hazards. In re- paid by employers (with some adjustments for the cent years, these have focused on the hazards of presence of health hazards in various industries). working with various chemicals, including sol- For example a 0.1 percent employer tax on the vents, metals, minerals, welding and cutting prod- total U.S. payroll of $1.6 trillion (in 1982) would ucts, and rubber and plastics. The fund has also result in annual revenues of about $1.6 billion. sponsored work on physical agents, including A payroll tax of 0.01 percent would raise $160 noise, vibration, radiation, and in issues concern- million. A 1.0 percent surcharge on workers’ coming ergonomic design related to working postures pensation premiums (about $25 billion in 1980) and lifting requirements (468). would produce annual revenues of $250 million. The training and information projects have in- Assessments for health, safety, and environ- cluded the development of course materials for mental activities have been used in the United the introductory and advanced training of worker States in at least two cases. The Comprehensive safety stewards and supervisors. From 1972 to Environmental Response, Compensation, and Lia- 1981, nearly 400,000 individuals received some bility Act of 1980 created a “superfund” to pay form of training. The Joint Industrial Council, for activities related to releases of hazardous which was created by an agreement of Swedish wastes. Most of the funding for this is from taxes employers and unions in the 1940s, has produced on oil and chemicals (546). The recently enacted most of the educational materials concerning right-to-know law in the State of Washington in- workplace health and safety used for this train- cludes an assessment of 75 cents for each employee ing (468). in the State, to be paid by employers (351). One bill concerning asbestos compensation, now under There are only a few U.S. examples of cooper- congressional consideration, proposes establish- ative labor-management-Government research ing an assessment to be used for research (517). and training activities related to health and safety. One example is a set of experiments concerning Although creation of an Occupational Safety the use of “washed cotton” to control the hazards and Health Fund would enhance the commitment 340 . Preventing Illness and Injury in the Workplace to research and training in this area, there are dis- NIOSH and OSHA could also encourage the advantages to consider. The relationship between development of programs to provide industrial this fund and the existing agencies (OSHA and hygiene, safety engineering, medical surveillance, NIOSH) would need to be determined. Moreover, and worker health and safety training. These a new tax or surcharge, even though one of mod- could be established to service industries, regions, est size, runs against recent reductions in business and employers who do not offer such services. taxes. Especially helpful would be programs for servicing small businesses in particular areas. Federal Assistance to Small Businesses Because it is inefficient and impractical to re- Improvements in employee health and safety quire each small business to provide a full range can be difficult for smaller businesses, which often of safety and health services, programs to pro- lack technical expertise in industrial hygiene, vide shared resources might be cost effective. Fed- safety engineering, and occupational medicine. eral funding could be used to start such programs. Smaller businesses commonly face financial dif- However, there may be difficulties in sustaining ficulties as well. these programs after the startup period. Even though the price of shared programs should be As discussed above, from 1971 to 1981, the less than if a company were to purchase the serv- Small Business Administration issued a limited ices entirely on its own, some small businesses number of loans to assist small businesses in com- might find it beyond their means. The steps that plying with OSHA standards. In 1981 Congress are needed to aid those companies are not ad- eliminated authorization for this program. One dressed by providing shared resources. option would be to study the results of this loan program, especially to learn why it was used so As discussed in chapter 15, the regulation of infrequently. After such a study, Congress could the products purchased by small businesses may consider reauthorizing the loan program and pro- also be a way of improving, to some extent, the viding an adequate level of funding. health and safety of their work forces. This approach could also be applied to nonregulatory Other kinds of Federal assistance for small busi- testing programs. For example, NIOSH could con- ness could include providing technical assistance duct occupational safety and health performance and facilitating the creation of programs to protests of products used by small businesses and vide technical assistance. Chapter 12 discusses publish the results in a fashion easily accessible OSHA consultation, which is the major OSHA to small businesses, which could then use the re- activity for providing assistance to small busi- sults of these tests in purchasing decisions. nesses.

SUMMARY Several approaches to improving occupational for the regulatory system, which penalizes firms safety and health have not been extensively used for violations of regulations, an injury tax sys- in the United States. These include the use of eco- tem would levy a direct financial penalty on the nomic incentives, tax incentives, and financial firm for each injury. Such an approach appeals assistance. Although in theory the workers’ com- to many economists, but it presents a number of pensation system provides an economic incentive difficulties. to prevent worker injuries and illnesses, in prac- Congress could also modify the structure of tice it provides only a limited economic incentive business taxes to encourage investment in health for prevention, especially for occupational ill- and safety control technologies, or it could pro- nesses. vide direct assistance to businesses in financing Another possible type of economic incentive in- health and safety investments. Either decision volves injury or exposure taxes. As a substitute would provide some level of “social funding” for Ch. 16—Economic Incentives, Reindustrialization, and Federal Assistance for Occupational Safety and Health ● 341 investments in occupational safety and health, Health and safety regulation may have nega- thus reducing their costs and encouraging firms tive effects on productivity, by requiring spend- to undertake them. Four kinds of tax and assisting for controls that are “nonproductive,” com- ance programs could be considered: investment pared with traditional measures of productivity. tax credits, accelerated depreciation, Government But in the aggregate, this diversion is relatively loan programs, and direct subsidies. All have been small, and may, in many cases, be offset by reg- used to stimulate adoption of pollution controls, ulation-induced changes that improve plant pro- and loan programs have been used, to a limited ductivity as well as employee health and safety extent, for occupational safety and health. protection. Facing the need to redesign products and process to comply with health, safety, and Tax incentives and financial assistance pro- environmental regulations, companies may fun- grams might spur the implementation of controls, damentally redesign product or process through assist businesses in compliance, and possibly re- the use of new technology. For both the OSHA duce the controversy of regulatory proceedings vinyl chloride and cotton dust standards, em- because of the availability of sources of finance. ployer compliance was associated with improve- However, they would cause either a reduction in ments in productivity. For the lead industry, it Federal tax revenues or an increase in budget is possible that new technology could be intro- outlays and could increase the complexity of U.S. duced that would simultaneously reduce worker tax law or increase administrative burdens. More- exposures, reduce the costs of controlling ex- over, these programs can be inefficient, and would posures, and improve productivity. affect other aspects of business investment behavior, as well as the distribution of wealth, Financing for research and training activities income, and the burden of income taxes. and for assisting small businesses could be pro- Federal policies concerning reindustrialization vided by an Occupational Safety and Health might have either a beneficial or an adverse im- Fund. The Work Environment Fund of Sweden pact on workplace health and safety. It is possi- offers a possible model for a U.S. fund that could ble that many of the improvements in worker provide a new focus for enhanced efforts in con- health and safety during this century occurred co- trol technology research, education and training, incidentally with the introduction of new technol- and information dissemination. ogies, new processes, and new industries and not as the result of conscious decisions to add con- Small businesses face special problems in mak- trols to existing processes. A time of reindustriali- ing improvements for employee health and safety. zation offers opportunities to integrate productiv- They often lack technical expertise in industrial ity-improving investments in plant and equipment hygiene, safety engineering, and occupational with the installation of control technologies, as medicine, and they face financial difficulties as well as to exploit the fact that it is cheaper and well. Assistance to such firms could include Gov- more effective to control a hazard when design- ernment loan programs, consultation, and Gov- ing new plant and equipment than it is to retrofit ernment testing and regulations of products pur- existing operations. Reindustrialization also af- chased by small businesses. In addition, OSHA fords an occasion to achieve greater levels of pro- and NIOSH could encourage the development of tection. However, there is also the danger that programs to provide industrial hygiene, safety combining policies that are designed to improve engineering, medical surveillance, and worker productivity with those that address employee health and safety training to small businesses, health and safety will lead to an emphasis on pro- especially in regions currently lacking such ductivity at the expense of health and safety. services. —

7. Preventing Work-Related Injury and Illness in the Future

., Contents

Page Future Trends ...... 345 Opportunities for Prevention ...... 346 Energy-Related Technologies ...... 347 Modernizing Old Industries ...... 350 New Technology=...... 351 Rapidly Growing Occupations ...... , ...... ,+... O 353 Summary ...... 354 LIST OF TABLES Table No. Page 17-1. Potential Occupational Hazards in Coal Liquefaction Plants ...... 348 17-2. Standardized Mortality Ratios Among Coal Miners ...... 349 LIST OF Figure No. Page 17-1. Features of Stand d Fermentor . . . . ar ...... , ...... 352 17-2. Features of Contained Fermenter ...... ** 353 17-3. Frequency of Health Complaints in 55 Office Environment Investigations...... ,...... ,*O. 353

. 17 ■ Preventing Work-Related Injury and Illness in the Future

Looking ahead in occupational safety and The look at the future in this chapter is very health, except in the most general way, is as dif- much constrained by knowledge of the present. ficult and open to error as predicting the future The changes that are described are evolutionary, in any field. In general, changes in plant and man- not revolutionary. That is not intended to imply ufacturing techniques, shifts in types of jobs, and that revolutionary changes will not take place, but attention to controlling hazards are expected to it acknowledges that revolutionary changes are reduce work-related injury and illness. Some new far harder to predict. For example, the drafters and exotic hazards are certain to accompany new of the National Cancer Plan in 1970 failed to men- processes, but what has been learned from con- tion one of the most important biological ad- trolling or failing to control current workplace vances, recombinant DNA (544), for the simple hazards can be applied to recognizing and con- reason that it had yet to be “invented. ” trolling the new ones. The future will present a mix of the problems To some extent, the workplace of tomorrow of today and tomorrow. The mix, it can be said will be the workplace of today. Some worksites with some confidence, is bound to change, and and industries will change little, and even in those certain new, just emerging problems, are expected that change, seemingly mundane factors will con- to capture the attention and efforts of safety and tinue to contribute to workplace injury and ill- health professionals in the years ahead. This ex- ness. Inadequate guarding of machinery, poorly pectation is in keeping with the observation made designed hand and power tools, and inappropriate earlier in this report that concern about the new walking and working surfaces will not go away sometimes seizes the imagination and attention without attention to injury prevention. Some of expert and lay person alike. This is particularity long-known health hazards remain problems. For likely to happen when the old problems are con- instance, silica dust in plants, mines, and found- centrated in industries or trades that are seen to ries continues to cause ill health. The absence of be on the way out and there is little incentive to controls in such situations underlines the impor- invest in them. It is a sure bet, however, that old tance for policymakers to understand how incen- problems in old industries will not take care of tives for adopting controls work and do not work. themselves if attention is diverted from them.

FUTURE TRENDS

There are a few sources of information about ment of Commerce, estimates that the U.S. popu- expected trends in employment and other sources lation will grow throughout the 1980s at a rate for predicting where new technologies are likely to similar to that of the 1970s-0.9 percent per year. be used; careful consideration of trends by leaders The rate remains the same because the effects of in health and safety could lead to the develop- changes in fertility, immigration, the age of the ment of controls to accompany the new tech- population, and mortality roughly cancel out. The nologies. predictions for increases in fertility rates and legal and illegal immigration and decreases in mortal- The 1984 edition of U.S. Industrial Outlook: ity would, by themselves, lead to an expected in- Prospects For Over 300 Industries (554), prepared crease in the population growth rate. Balancing by the Bureau of Industrial Economics, Depart- those is the fact that women born during the baby

345 346 . Preventing Illness and Injury in the Workplace

boom of the 1940s and 50S are passing out of their dustries. The injury rate in those industries— most fertile years, reducing the number of women computer, banking, legal, and medical services having children. Reduced mortality and aging of and related activities—is lower than the all-indus- the baby boomers will produce an increase in the try average. Nevertheless, the number of people average age of the population. employed in them means that attention to safety and health in those industries will be increasingly The labor force increased by 23 million (26 per- important in the future. cent) between 1972 and 1982. Part of the growth resulted from the unprecedented increase in the The Environmental Protection Agency (EPA), number of women entering the work force, the in cooperation with the National Institute for Oc- remainder from the entry of young men. This cupational Safety and Health (NIOSH), has con- spectacular rate of growth will not continue dur- sidered the production of a document that would ing the next decade, when 17 million people (a concentrate on environment and occupational 16 percent increase) are expected to enter the work health and safety. As planned, the outlook would force. These projections indicate that prevention have two parts. In the first, data available from of injury and illness in the next 5 years will be Federal sources about current and expected em- affected by an aging work force including an in- ployment levels, injury and illness rates, com- creasing number of women. pliance with Occupational Safety and Health Administration (OSHA) regulations and EPA air Even this very general information can be of and water quality regulations, solid waste dis- value to health and safety professionals as they posal, and energy and water use would be pre- look ahead to the next decade. For instance, older sented for each four-digit Standard Industrial workers may respond differently to prevention Classification code. Insofar as possible, informa- training than younger workers. Women will re- tion about occupational and environmental meas- quire different considerations in the ergonomic de- ures would be presented for each industry cov- sign of jobs and personal protection equipment. ered in the Department of Commerce Industrial Information about which industries are likely Outlook. A very useful feature of the EPA/ to grow and which are likely to decline is useful NIOSH document is that all the data are entered for directing research in control technology and in a computer network so that users with personal for encouraging companies to consider injury and computers can carry out their own analyses. The illness prevention before a new plant is built and second part of the document is a narrative, pro- new technologies installed. Anticipation of prob- filing each of the industry groups in terms of envi- lems in the planning stages can help prevent seri- ronmental safety and health, highlighting trends ous problems later on. Gaining industries cited in the work force, the use of new technology, the in the 1984 report include the motor vehicle and handling of hazardous substances, and other re- electronic-product-related groups. However, the lated matters (220). most rapid growth was expected in the service in-

OPPORTUNITIES FOR PREVENTION Many new technologies should be inherently • Modernizing old industries safer for workers than those of the past, but new Steelmaking processes will require modification of old or fab- Automobile manufacturing rication of new controls. Eight kinds of work- • New industries places are discussed here as examples of changes Semiconductor manufacturing that are expected: Biotechnology industries ● Rapidly growing occupations ● Energy-related industries: Office work Synthetic fuel production Coal mining In terms of expected numbers of workers, the Off-shore oil drilling industries vary. Synthetic fuels may or may not Ch. 17—Preventing Work-Related Injury and Illness in the Future . 347 become important in the future and even if they ing new energy sources and new technologies for do, large-scale production will take time to conserving energy. Production of liquid fuels from achieve. Should that happen, a new work force coal, oil shale, and tar sands will involve new will have to be trained, and employment in the technologies and new hazards, and new controls industry could take off. Coal mining and offshore will be needed. Increased coal use will stimulate oil drilling are employing more workers now. the purchase and use of new mining machinery With the introduction of larger, more efficient ma- that will require appropriate controls. chinery in mining, the number of miners is expected to plateau and then decline. Oil drilling Synfuel Production employment will increase or at least remain constant so long as the search for oil fields is prof- Synthetic fuel (synfuel) production methods abun- itable. Steel and auto manufacture, despite some break down complex molecules of relatively dant and naturally occurring carbonaceous ma- recent resurgence, are not expected to employ as terial such as coal or oil shale to produce simpler, many as they did in the 1970s. cleaner, more efficient fuel. Although these tech- Jobs in the semiconductor-related technologies nologies have been used before on a small scale, are expected to increase. Biotechnology is, per- they have never been economically competitive haps, at a point in its development analogous to with fuel production from crude oil and have re- that of the semiconductor business a decade or mained commercially undeveloped. These tech- two ago. If some of the newly created biotech- nologies, though, could eventually be adopted by nology firms are successful, there should be more the United States to achieve energy independence. jobs in this field, but the maximum number of Congress established the U.S. Synthetic Fuels workers is likely to be small and they will be Corporation (SFC) to fund projects that would highly trained. lead to production of clean and safe energy. The Technologies affecting office work crosscut all SFC has proposed guidelines to monitor emissions sectors, both public and private, and are rapidly in the workplace and to the surrounding environ- growing as the U.S. economy becomes increasing- ment (653): ly information-oriented. It is expected that the Any contract for financial assistance shall re- number of workers in offices or at least doing quire the development of a plan, acceptable to the work that is now associated with offices will in- Board of Directors, for the monitoring of envi- crease over the next few decades. ronmental and health-related emissions from the Consideration of future employment oppor- construction and operation of the synthetic fuel tunities is directly related to health and safety. Not project (486). only do different jobs carry different risks, but Plans requesting financial assistance are to in- more important, work is essential to good health. clude details of a monitoring system, listing sub- A number of studies have shown that medical stances to be monitored; the frequency, location, services are more heavily used during periods of methods, and durations of monitoring; and work- unemployment, that mental health problems in- er exposure and health surveillance programs. A crease in number and severity, and that suicides worker registry is required to integrate worker ex- increase. Working is good for health, and proper posure data, medical records, demographic infor- attention to identifying and controlling hazards mation, and job classification codes so that any in the workplace can prevent injuries and illnesses. trends in work-related injury and illness can be identified. Energy-Related Technologies Production of liquid fuel from shale oil involves liberation of some chemicals that are carcinogenic, The fuel crises of 1973 and 1978 led to dramatic and studies of pilot plant workers in this country changes in the production and use of energy in report dermatitis, eye irritation, and thermal the United States. The Department of Energy was burns from job exposures. Fire and explosion are formed, and resources were marshaled for the Na- hazards because of the operating temperatures and tion to become energy independent through find- pressures necessary for synfuel production. 348 . Preventing Illness and Injury in the Workplace

Table 17-1 lists the potential hazards that could in future operations to prevent harm at all levels be expected in a coal liquefaction plant. Problems of operation (575). are encountered during coal handling and preparation, during the process itself, and during waste Coal Mining treatment. Estimates have been made of the chem- Most coal is used in the traditional way to gen- ical products and byproducts that might be found erate power directly in steam plants, and its use at each point in the process so that worker pro- has increased during the past decade of uncer- tection can be considered during design (575). tainty about oil supplies. Use is expected to in- Concern about these hazards has contributed crease further, to double, in fact, by the year 2000, to systematic analysis of plans of future synfuel to as much as two billion tons per year (536). plants in an effort to anticipate causes of injury Mining and moving coal are hazardous. Over or illness. For instance, NIOSH reports one ex- 100,000 coal miners have been killed since 1900), ample of the success of this approach. High pres- and coal workers’ pneumoconiosis (black lung dis- sure vessels for coal liquefaction operate at extremely ease) has taken a high toll. A mortality study of high pressures and temperatures and contain flam- 23,233 miners selected randomly from those eligi- mable material. Engineers, recognizing the poten- ble for United Mineworkers of America health and tial hazard from a high-pressure vessel used in a retirement funds as of January 1, 1959 and fol- bench-scale (laboratory-sized) coal liquefaction lowed through December 31, 1971 with 99 per- process, placed protective barriers around it. cent followup, reported excess mortality for stom- When the vessel unexpectedly exploded, harm was ach cancer, influenza, asthma, tuberculosis, and prevented. In another instance, reinforced con- accidents (392) (see table 17-2). Unless steps are crete walls between the liquefaction system and taken to assure prevention of work-related injury the operating control room protected operators and illness in coal mines, an increase in produc- from injury and operating controls from damage. tion may lead to excess morbidity and mortality. The unharmed operators were able to shut down the process, thus preventing further explosions Changes in technology are introducing new and fire. These same techniques could be applied hazards into the mine. For instance, the diesel en-

Table 17-1 .—Potential Occupational Hazards in Coal Liquefaction Plants

System, unit operation or unit process Potential hazards Coal handling and preparation system...... Coal dust, noise, fire, explosion, asphyxia (nitrogen and carbon monoxide gases), burns Liquefaction system ...... Phenols, ammonia, tars, thiocyanates, PAH’s, carbon monoxide, hydrogen sulfide, hydrocarbons, fires, explosions, burns, high pressures, noise, ash, slag, mineral residue, spent catalyst Separation system ...... Oils, phenols, hydrogen cyanide, ammonia, hydrogen sulfide, burns, fires Upgrading and gas purification ...... Light hydrocarbons, phenols, ammonia, hydrogen sulfide, carbon dioxide, carbon monoxide, burns, fire, explosion, high pressures Shift conversion ...... Tar, naphtha, hydrogen cyanide, fire, catalyst dust, burns, hot gases (carbon monoxide, hydrogen) Methanation a ...... Carbon monoxide, methane, nickel carbonyl, spent catalyst dust, fire, burns Waste treatment facilities . . . . Hydrogen cyanide, phenols, ammonia, particulate, hydrocarbon vapors, sludges, spent catalyst, sulfur, thiocyanates alndirect liquefaction SOURCE: (575), Ch. 17—Preventing Work-Related Injury and Illness in the Future . 349

Table 17.2.-Standardized Mortality Ratios Among Coal Minersa

Deaths Cause of death Observed Expected SMRb All causes...... 7,628 7,506.1 101.6 All malignant neoplasms ...... 1,223 1,252.2 97.7 Respiratory organs ...... 373 331.0 112.5 Stomach cancer ...... 127 91.9 134.9 b Major cardiovascular diseases ...... 4,285 4,501.2 95.2b Chronic and unqualified bronchitis ...... 26 29.0 89.7 Influenza ...... 28 14.8 189.6 b Emphysema ...... 170 118,3 143.7 b Asthma ...... 32 18.3 174.9 b Tuberculosis ...... 63 43.3 145.5 b Coal worker’s pneumoconiosis...... 187 — — Accidents ...... 408 283.0 144.2 b %al miners covered by the United Minewotiers of America health and retirement funds The vital statusof 22,998 miners was verified. bstandartjizedmortafity ratlo(sMR) is significantly different from 100at the5 percent confidence level SOURCE (392) gine is increasingly replacing electrically powered fer to the surface via conveyors running through mine equipment. The exhaust emissions from such the tunnel. Supports at the face protect miners engines include cancer-causing polynuclear aro- while allowing the roof to cave in behind the face matic hydrocarbons, respiratory system irritants, just worked, preventing unwanted roof fall. Long- and asphyxiating carbon monoxide and carbon wall mining may reduce fatalities but not neces- dioxide. sarily injuries or harmful dust levels (536). Control technologies for diesel engines are well known. Diesels operating under ideal conditions, Outer Continental Shelf Oil Production with proper maintenance and operation, produce Another energy area expected to grow is oil and minimal amounts of contaminants. In under- gas production on the outer continental shelf. De- ground coal mines, ventilation can be increased mand for energy has led to the development of to keep contamination at a minimum. Filters, re- technologies for deep water oil and gas explora- circulation of exhaust gases through the engine, tion in remote locations and under extreme envi- catalytic converters, and exhaust scrubbers can ronmental conditions. An increasing number of also be used to clean the exhausts (477). Such con- workers, already at high risk because of the nature trols are difficult and require high degrees of main- of oil extraction, will face even greater risks of tenance to assure that they are functioning properly. work-related injury and illness. In the period 1970- Other examples of a production technology that 79 employment in outer continental shelf oil and may increase risk of injury in the mines include gas exploration grew by 71 percent (to a work continuous miners and long-wall mining. Contin- force of 61,500) for an average annual growth of uous miners are machines that cut into a coal vein 3 percent. Moreover, there was a 20 percent per and transfer loose coal back to a conveying mech- year increase in two years, 1978 and 1979 (319). anism for transport to the surface. While they in- There are two areas of concern. The first is the creased productivity when they were phased into design and stability of drilling rigs. A combina- the mines from 1950 to 1970, the fatality rate rose tion of bad weather and inadequate structural slightly and the disabling injury rate continued strength has resulted in major catastrophes, such at a relatively constant level. Long-wall mining as the failure of a mobile offshore drilling unit in is a mining method in which a machine extracts 1982 in the North Atlantic that killed 84 workers. coal by moving back and forth across a face while conveying the coal to one of two tunnels dug par- The second area is the risk to workers from the allel to each other and at right angles with the coal work itself. Drilling is frequently continuous, be- face. The coal is transported to stations for trans- ing done in shifts 7 days a week, under difficult 350 Preventing Illness and Injury in the Workplace conditions. Drill space is cramped and pipes, tub- in the steel industry. Research in steel technology ing, tongs, and other material are heavy and cum- should continue to take into consideration fea- bersome. Walking and working surfaces are slip tures for protecting workers and improving the pery from drill fluids. The work is outside, so ambient environment both with respect to new weather is often a factor. The severity of these and to existing steel facilities. conditions is heightened on offshore drill rigs and The Working Group went on to recommend may be expected to be even more severe on off- Federal funding for research and development and shore units located in remote areas, where weather for demonstration plants. These recommendations may be extreme. were agreed to in recognition of the high risk of work-related injury and illness in steelmaking, and the opportunity to reduce costs through develop- Modernizing Old industries ment, demonstration, and adoption of control U.S. manufacturers are facing intense competi- technologies. tion from Japanese and Western European compa- New technologies for steelmaking are more pro- nies, and that sector of the U.S. economy may ductive and likely to cause fewer work-related in- be declining. The declines have been attributed juries and illnesses. For example, continuous cast- to inadequate investment in new plant, changing ing is used to produce 80 percent of Japanese steel market conditions, rising labor costs, and cost- and 32 percent of U.S. steel. It requires less generating Government regulations. Whatever the energy, costs less per ton of steel produced, and exact reasons—and they are a matter of argu- produces higher quality steel and less pollution. ment—it is agreed that the U.S. steel industry is It also eliminates soaking pits and reheating fur- no longer the first in the world. Changes in the naces and requires less coke making, three steps qualities of cars that induce people to buy an auto in the traditional process that produce emissions resulted in U.S. consumers buying more foreign- that are associated with disease. In general, bet- built models, and U.S. domination of auto sales ter working conditions exist in continuous casting is a thing of the past. Both these industries are steelmaking plants. retooling and retrenching to meet the foreign threats, and there will be opportunities to build Another new steel technology is direct reduc- in features to prevent work-related injury and tion of iron ore rather than the current process illness. of blast furnace and coke oven. Since direct reduction can be done without coke, risk of lung cancer from coke oven emissions can be elimi- Steelmaking nated. Other technologies that show promise for Steelmaking is one older U.S. industry that the 1990s and that can be made less likely to cause must change if it is to survive. Increased research, work-related injury and illness include formcok- development investment, and the use of new tech- ing (another process reducing coke oven emis- nology are all required to compete with steel- sions), direct casting of sheet and strip metal from maker from other countries (538). Unfortunately, molten steel, and one-step steelmaking directly this has become a chronic public policy problem. from ore. In the latter two processes, potentially As mentioned in chapter 16, the Steel Tripartite hazardous steps in the process are eliminated, thus Advisory Committee, made up of steelworkers, reducing risk of worker harm. steelmaker, and Government officials, was estab- The same changes that improve production and lished by President Carter in the 1970s to attempt reduce risks will cause further shrinking of the to revitalize this industry. work force. In the late 1970s, 450,000 workers The Committee’s Working Group on Techno- were employed in steel; in 1983, the number was logical Research and Development concluded that less than 250,000, and it continues to decline. (458): Automating environmental and occupational safety and health issues should be considered as an integral Automating includes many kinds of industrial part of technological research and development operations, ranging from metalworking in steel-

Ch. 17—Preventing Work-Related Injury and Illness in the Future ● 351 works and foundries, to forging and machining successful of the new industries is the semicon- of metal parts, to fabrication of plastic parts, and, ductor industry, and the most glamorous of the finally, to assembly. Hazards in the steelworks promising ones is biotechnology. and foundries include noise, heat, dust, and gas. Forging and machining are also hazardous. Auto Semiconductor Manufacture and Related Industry body painting presents special hazards due to the large volumes of paint and solvents used during The continuing demand for computer and video assembly. Stress-related illness may result from devices for industry and commerce, coupled with monotonous work and shift work. consumer electronics and computer technologies for microelectronic applications, is expected to Conventional controls are available for these fuel the growth of this industry. hazards. Local exhaust and dilution ventilation are widely used in all phases of manufacture and Microelectronics has been estimated to have a world market of more than $19 billion and em- can be expected to be improved in new processes. ploy a work force of 500,000 worldwide (258). Automated spray painting in booths reduces or eliminates worker exposure. The use of robots in The risk of work-related injury appears to be areas of high hazards can reduce risks to workers. lower and illness appears to be higher in this in- Robots or automated manufacture also intro- dustry. In 1981 the Bureau of Labor Statistics reduces hazards. There have been cases here and ported the injury incidence rate for semiconduc- abroad of workers being injured or killed by auto- tor and related devices as 4.6 cases per 100 mated machines. This points to the need for ma- full-time workers and lost-workday cases as 2.0 chines to stop when workers inadvertently come per 100 workers, compared with an all-industry in contact with them or to be installed where they average incidence rate of 8.3 cases per 100 work- keep workers out of danger zones while the ma- ers and lost-workday cases of 3.8 per 100 full-time chine is connected. workers (608). The California Department of In- dustrial Relations conducted a survey in 1980 that showed workers manufacturing semiconductors New Technologies had 1.3 illnesses per 100 workers compared with New technologies that are burgeoning or prom- a rate of 0.4 per 100 workers in the general man- ise to burgeon into full-scale industries present op- ufacturing industries. Lost time resulting from portunities to prevent work-related injury and ill- work-related illness was three times more com- ness in the early stages at low cost. The most mon among semiconductor workers (18.6 percent of all lost-workday cases) than in general manufacturing industries (6.0 percent of all cases) from 1980 to 1982. Almost half (46.9 percent) of all work-related illness among Californian semiconductor workers in this period was reported to result from exposure to toxic substances (258). Known health hazards include metal fumes from soldering, toxic chemicals such as epoxy resins and chloronaphthalene, silica flour used in making insulating materials and dielectrics, solvents for decreasing solder joints, and acids for etching printed circuits. The volumes of these chemicals used in California in 1979 were large: Over two million gallons of solvents; more than

Photo credit: NIOSH two million gallons of sulfuric, hydrofluoric, and other acids; more than one-half million gallons This paint spraying line has been automated using robots. Future developments in robotics may reduce of sodium hydroxide and other caustics; and over worker exposures in other hazardous operations one and one-half million cubic feet of arsine, phos-

352 ● Preventing Illness and Injury in the Workplace phine, diborane, and other toxic cylinder gases worker protection is monitored by the Research were used in processing semiconductors (258). Advisory Committee of the National Institutes of Health, and NIOSH has developed guidelines for Existing technologies, if applied, should be suf- medical surveillance of fermentation and biotech- ficient to control these hazards. Local exhaust ven- nology plant workers (259a). tilation is appropriate for soldering and etching operations. Substitution may be appropriate NIOSH industrial hygiene surveys of six indus- where solvents are found to be more toxic than trial laboratories using recombinant DNA tech- expected. Replacement of carbon tetrachloride nologies found wide variation in safety and health and trichloroethethylene (after they were shown practice among them. Practices ranged from “ex- to be carcinogenic among laboratory animals) emplary” plants with health and safety programs with perchloroethylene is an example of this. for workers to plants where workers were allowed to smoke and drink in the laboratory and to store Biotechnology beverages in laboratory refrigerators, and where procedures for biological waste-disposal were Recombinant DNA technologies, operating at undocumented—all unacceptable practices. moderate pressure and temperature, have fewer inherent physical hazards than traditional chem- Figures 17-I and 17-2 contrast a standard fer- ical processes, which sometimes operate at menter with a contained one, which would con- dangerously high pressures and temperatures. Fur- tain organisms and culture media used in biotech- thermore, to the extent that biotechnology uses nology. The contained fermenter provides for pinpoint production techniques to produce par- double filtering of exhaust gases to control emis- ticular chemicals, it will eliminate the currently sions, a special mechanical seal at the top to pro- encountered mixtures of chemicals, contaminated vide extra protection against loss of growth me- with unwanted toxic compounds, that are com- dium that contains bacteria, and an alarm to warn mon in conventional chemical synthesis. operators of ruptured seals, thus helping prevent possible loss of contaminated broth. These fea- Biotechnology’s hazards center on the possibil- tures of the contained fermenter provide increased ity that the microbes used in it or products pro- protection both to workers and the environment. duced from them will be harmful to human health or the environment. Since some of these organisms are “new,” in that they have been produced by genetic engineering, there is concern among some people that they present significant risks Figure 17-1 .—Features of Standard Fermenter (541,548). Most experts in the field see the orga- Exhaust nisms being used for or proposed for use as pro- to atmosphere duction organisms as “crippled” and unable to sur- unfiltered vive outside the laboratory or workplace. Dip tube Compressed air harvest and fill More of a problem are “new” micro-organisms inlet line developed to be released into the environment. Because they will have to compete with organisms that occur naturally, they cannot be crippled. For that reason, EPA is now considering regulating Cooling jacket the intentional release of such organisms, and Congress has expressed interest in a regulatory Agitator scheme.

In the area of using micro-organisms for pro- Air sparger duction, the United States appears to have studied questions of worker health more carefully than other countries. The U.S. voluntary approach to SOURCE: NIOSH

Ch. 17—Preventing Work-Related Injury and Illness in the Future . 353

Figure 17-2.—Features of Contained Fermenter tion has been adopted to reduce the relatively high Second costs of heating and cooling. Harmful air contami- back up Exhaust gas nants, aerosol can chemicals, tobacco smoke, and sterile filter filter system (1) Special double mechanical pathological micro-organisms may reach unac- seal with sterile fluid purge Primary ceptable levels. Some building technologies also (2) and loss of pressure air sterile contribute to the contamination. For example, the filter Dip tube irritating and potentially carcinogenic compound Compressed air harvest and fill inlet line (3) formaldehyde is emitted from plywood, and naturally occurring radioactive radon gas may be emitted from certain building sites. Even when the Other features exact causes of workers’ health problems remain (6) Collection of unidentified, the problems have been reversed contaminated et with increased ventilation. condensate and sterilization before Health effects have been related to indoor air release Agitator (7) System validation pollution. At high concentrations, indoor air con- before operation (8) Monitoring program taminants may cause irritation of sensitive tissues, (9) Operation with and acute or chronic illness. Many substances ap- Air sparger detailed protocol pear to act primarily as irritants at low exposure levels, inducing local inflammatory reaction in the

SOURCE: NIOSH eyes, nose, lung, or other sites (25). NIOSH made 159 health hazard evaluations in response to requests between 1971 and 1983. Table 17-3 shows Rapidly Growing Occupations that irritation of eyes and throat were reported in 81 and 71 percent of the cases respectively (661). Office Work Control of indoor air pollution maybe achieved Thirty-three million people work in office jobs by increasing ventilation rates, eliminating the in this country. The advent of new technology source of contamination, and air cleaning (25). such as computers using video displays and ad- Air cleaning is generally limited; although it filters vanced copying machines have changed the of- particles, it does not remove gases and vapors. fice. To some extent, machines can pace the work, introducing a new source of stress. Lighting and furniture, ignored in the installation of the first Table 17-3.-Frequency of Health Complaints in computers and word processors, are important 55 Office Environment investigations to worker comfort and probably to health and Percent of buildings safety. Symptom with complaint Eye irritation ...... 81 The energy crisis has resulted in office buildings Dry irritated throat ...... 71 being tightly insulated. Unfortunately, improved Headache ...... 67 insulation that keeps heat or cooled air in also pre- Fatigue ...... 53 Sinus congestion ...... 51 vents air exchange from cracks and other tiny Skin irritation...... 38 leaks and increases indoor air pollution. Because “Shortness of breath” ...... 33 building air is recirculated to conserve energy Odor ...... 31 Cough ...... 24 rather than directly exhausted to the outside, Dizziness ...... 22 many ventilation standards are now inadequate. Nausea ...... 15 Formerly unacceptable, the practice of recircula- SOURCE: (661). 354 ● Preventing Illness and Injury In the Workplace

Ventilation standards, if they are to guard against new concerns. Stress may be increased by a sense indoor air pollution under the new conditions, of isolation from the workplace and coworkers, need to be rewritten. and opportunities for advancement, which are a positive force in most workers’ lives, may not ex- There are reports that office work, especially ist. Also, to the extent that office machines bear work involving routine use of computers for doc- hazards, those are likely to go unrecognized in ument completion and filing, is being moved from the dispersed workplace and even more likely to the office to the home. While working at home go uncontrolled (25). is attractive to many people, it will also introduce

SUMMARY The ability to anticipate change is fundamen- potential of reducing exposures to inherently tal to preventing work-related injury and illness. dangerous chemical processes and potentially haz- Knowledge about the sectors of the economy ardous chemical mixtures. But care must also be where changes are likely to take place, what the taken in handling the micro-organisms used in changes might be, and how they might affect these processes. workers will help responsible officals carry out Office work is one of the most rapidly chang- their mandate under the Occupational Safety and ing occupations as new technologies are prolifer- Health Act and to improve occupational health ating. Since one-third of the work force are in of- and safety in other ways. fices, even low rates of work-related injury and Changes can be expected in energy-related in- illness can be of concern. Three areas where atten- dustries, modernization of older industries, indus- tion is needed in offices are indoor air pollution, tries relatively new to the economy, and in the stress, and the ergonomic problems associated rapidly growing area of office work. New tech- with VDTs. nologies are being introduced, some of which But the problems of the past will also remain. could cause work-related injury and illness. But It will be difficult to convince smaller firms to in- these changes also present opportunities for pre- vest in control technologies, especially as they will venting work-related illnesses and injuries. find the financing difficult. Companies relying on The production of synthetic fuels also creates older technologies, such as those found in the possible worker exposures to a number of differ- basic industries, remain reluctant to install con- ent hazardous substances, as well as a potential trols, especially when times are hard. The tortuous for fires and explosions. Coal mining has always process of court battles over standards is likely been hazardous, not only in terms of injuries, but to continue, leaving the public and workers some- is also associated with increased worker illness. what bewildered about the protection the Federal The use of diesel engines and new production tech- Government attempts to afford them. nologies introduce new hazards underground. The incentives for control that are discussed in Operations on offshore oil and gas rigs result in this assessment deserve careful attention to discern relatively high injury rates. how they work or do not work to encourage bet- Modernizing older industries such as steelmak- ter health and safety programs. It is by now clear ing and automating should provide opportunity that regulations are slow to emerge from OSHA, for additional protection of workers. Meanwhile, that enforcement will always be limited by the brand new technologies, such as electronics and small corps of inspectors, and that OSHA consul- biotechnology, are being introduced. Greater at- tation services cannot reach every workplace that tention should be paid to worker health problems could benefit from them. Continuing demands in the manufacture of semiconductors because of from a work force and a public less willing to ac- the use of toxic chemicals. Biotechnology has the cept risk of injury and illness will impose greater — . . .

Ch. 17–Preventing Work-Related Injury and Illness in the Future ● 355 pressures on workers’ compensation, liability law interest in the role of the general environment and and the courts more generally, and insurance com- the workplace in health has probably become part panics and voluntary associations interested in of the overall social consciousness of the Nation. health and safety. One of the measures of the importance of those ideas will be improvements in occupational health The great interest in environmental health that and safety in the years to come. developed in the 1970s remains alive and well, and Appendixes .

Appendix. A. —Supplemental Information on OSHA and NIOSH

Other Reports on OSHA and NIOSH enforcement activity, employee rights, and the history of State programs, The General Accounting Office (GAO) has exam- Three other reports on OSHA are of special inter- ined a number of aspects of OSHA and NIOSH oper- est. Two were prepared by Presidentially appointed ations. It has reported on standards-setting activities groups. The first, appointed by President Ford and of OSHA and the criteria-setting activities of NIOSH often referred to as the MacAvoy Commission, exam- in two different reports (494,501) and generally criti- ined OSHA’s safety standards and recommended that cized the slow pace of the development of new stand- OSHA issue performance standards(276). The second, ards, and the lack of coordination between the two an Interagency Task Force appointed by President agencies. GAO has examined emergency temporary Carter, made a large number of recommendations on standards (495,496) and the procedures used by OSHA OSHA inspection activity, creation of economic in- to grant employers variances from standards (497), centives for OSHA compliance, establishing cooper- and expressed concern that OSHA’s activities were not ative programs, and reforming regulatory activity sufficient to ensure worker health and safety. GAO (228). In addition, two academic economists, Richard has also criticized OSHA’s management of its consulta- Zeckhauser and Albert Nichols, studied OSHA regu- tion program (505), OSHA’s monitoring of State Pro- lation at the request of the Senate Committee on Gov- grams (500), as well as the administration of NIOSH’s ernmental Affairs, which published their report in 1978 HHE program (503). (685), GAO has reviewed OSHA’s health inspections (502), its safety inspections (504), and the procedures used for scheduling complaint inspections (507), and OSHA Standards Issued After Rulemaking was critical of several aspects of OSHA’s inspection activity. GAO has in two reports criticized OSHA’s As described in chapter 12, OSHA has the author- data collection efforts, pointing to inadequacies in data ity to issue new standards, and to modify or revoke on injuries and health hazards and OSHA’s failure to existing standards using procedures specified in the Oc- use the information it collects through accident inves- cupational Safety and Health Act (OSH Act). Tables tigations (499,508). A 1984 GAO report examined A-1 and A-2 present details of the rule-making pro- OSHA’s policies of encouraging the informal settle- ceedings that have resulted in final standards during ment of citations (511). OSHA’s first 13 years. These proceedings can begin Mary Jane Belle, of the Congressional Research with the receipt of a Criteria Document from NIOSH, Service, prepared a report in 1981 on OSHA reform the creation of an ad hoc advisory committee, or the (530). She has also written and updated a Congres- publication of an Advance Notice of Proposed sional Research Service Issue Brief on OSHA (533). Rulemaking. Although, in theory, both of these lat- Crisis in the Workplace by Nicholas Ashford (30) ter two might occur in the same proceeding; in prac- and Bitter Wages by Joseph Page and Mary-Win tice they have not. In fact, in recent years, OSHA has O’Brien (361), provide accounts of some of OSHA’s tended to use the Advance Notices, and has not used early history and present their evaluations of govern- ad hoc advisory committees. (The exceptions are mental activities. Other studies of occupational health standards involving the construction industry, for and safety regulation are Robert Smith’s The Occupa- which OSHA is required, by its own regulations, to tional Safety and Health Act (44 and John Mendeloff’s consult with the standing Construction Safety Advi- Regulating Safety (300). David P. McCaffrey, OSHA sory Committee. ) Moreover, in recent years, NIOSH and the Politics of Health Regulation (290) gives a his- has issued few criteria documents. Proceedings are tory and analysis of the health standards issued dur- now more likely to begin with a petition from an in- ing OSHA’s first decade, while Steven Kelman’s Reg- terested group, such as a union, for a standard, ulating America, Regulating Sweden (245) provides a The formal Notice of Proposed Rulemaking and comparison of OSHA and its Swedish counterpart. In publication of the Final Standard and statement of rea- his collection entitled OSHA: History, Law, and Po/- sons are necessary steps in order to issue a standard. icy (307), Benjamin W. Mintz provides numerous ex- A public hearing is not essential, unless an interested cerpts from primary source documents related to many party requests it. For major and controversial stand- of the important disputes about OSHA standards and ards, a hearing is invariably requested. 360 . Preventing Illness and Injury in the Workplace

Under section 6(f) of the OSH Act, “[a]ny person ard may be issued “repeated” violations. Fines of up who may be adversely affected by a standard issued” to $10,000 may be imposed for both willful and re- by OSHA can challenge the standard in any of the U.S. peated violations. OSHA’s largest penalties usually in- Courts of Appeal. A column in tables A-1 and A-2 involve an employer’s “failure to abate” or correct a haz- dicate if any challenge occurred, the circuit in which ard. The OSH Act authorizes penalties of up to $1,000 it was filed, and the date of the decision. Table A-3 for each day that the hazard continues beyond the day lists the names and citations for these cases. it was supposed to have been abated. In practice, these Finally, OSHA has for a number of its standards, have been limited to a maximum of 10 days or $10,000. taken formal steps to reconsider and revise standards The Act also authorizes criminal prosecution in sev- that had been issued in final form. The last column eral situations: First, a willful violation that results in of tables A-1 and A-2 list these actions. an employee’s death may be punished by criminal penalties including a fine of up to $10,000, or 6 months OSHA Enforcement Activity imprisonment, or both. For a second conviction, these maximum penalties are doubled. There have been only Tables A-4 to A-n present detailed information a handful of these cases under the Act. In addition the concerning inspection activity by OSHA since Fiscal Act provides for criminal penalties for OSHA officials Year 1973 and the State programs since Fiscal Year who give an employer unauthorized advance notice 1976. The data for these tables were provided by of an inspection, and against anyone who falsifies OSHA. Table A-4 provides the number of inspections, OSHA-required records, or uses force to interfere with both for safety hazards and for health hazards. Table the work of an inspector, although there have not been A-s presents these inspections according to OSHA’s any cases brought for these last three types. (For a priority categories—fatality/catastrophe investiga- more detailed discussion, see 307,333,408. ) tions, complaint inspections, programed inspections, In practice, penalties are substantially lower than and follow-up inspections. Table A-6 gives the num- the maximum penalty amounts outlined above, reflect- bers of inspections by major industry groups. ing, in part, OSHA’s discretion in setting penalties. In Tables A-7 to A-11 include information on the va- proposing penalties, OSHA considers the gravity of rious types of violations issued by OSHA. The OSH the violation, the good faith of the employer, the size Act specifies that penalties be imposed on employers of the business, and the employer’s previous history for violations of standards. Except in the case of de of compliance. minimus violations that have “no direct or immediate relationship to safety or health, ” and other-than- serious violations, OSHA must issue a citation, pro- Activities of Other Federal Agencies pose a penalty, and set a “reasonable” abatement period. OSHA and the 25 State Programs are directly re- A “serious” violation is issued for hazards that pre- sponsible for ensuring the health and safety of most sent a “substantial probability of death or serious phys- private sector workers in the U.S. However, workplace ical harm” to employees. A fine of up to $1,000 for health and safety for some private sector workers are each serious citation can be imposed. An other-than- the responsibility of other Federal agencies. In general, serious violation is not explicitly defined in the Act, health and safety conditions for most public sector but it falls between serious and de minimus violations. workers are not directly regulated by OSHA, although These violations have also been termed “non-serious State Programs, at least in theory, cover State and violations.” OSHA and OSHRC interpret other-than- local employees in States with State Programs. Finally, serious violations to involve conditions that have a di- the regulations issued by several other Federal agen- rect and immediate relationship to worker safety and cies also affect job safety and health, even though health, but without a substantial probability of death workplace conditions are not the primary focus of or serious physical harm. Although a fine of up to these agencies. $1,000 could be imposed for these violations, in prac- The constellation of governmental bodies with tice the proposed fines are substantially smaller. workplace safety and health responsibilities is sum- ‘Willful” violations are defined as those that are “in- marized in table A-12. The OSH Act directly regulates tentional and knowing, as distinguished from acciden- “employers,” who are defined as persons and busi- tal, and display a careless or reckless disregard or plain nesses who have employees and are engaged in inter- indifference to the Act or its requirements. ” (333). state commerce (Section 3(s)). This generally covers Employers will usually correct a hazard after being private sector employers, although anyone who is self- found in violation. Employers who subsequently are employed and who has no employees is not directly found to violate the same standard or a similar stand- subject to OSHA regulation. App. A—Supplemental Information on OSHA and NIOSH ● 361

In addition, the occupational health and safety of ing Federal employee labor organizations. OSHA also some private sector employees is regulated by other provides technical assistance to other Federal agencies agencies. Section 4(b)(1) of the OSH Act provides that concerning the health and safety of Federal workers. the OSH Act does not apply to “working conditions” The health and safety of State and local government for which other agencies “prescribe or enforce stand- employees is the responsibility of the States and ards or regulations affecting occupational safety or localities that employ them. Any State that establishes health. ” These exclusions are, in some instances, for a State Program must provide an occupational safety all aspects of occupational safety and health; in others and health program for state and local employees that only for certain hazards. For instance, the Mine Safety is “as effective as the standards” adopted for private and Health Administration (MSHA) is responsible for sector workers. But State and local government em- all safety and health hazards associated with mining. ployees in States without State Programs are not cov- The Nuclear Regulatory Commission, in contrast, is ered by this requirement. responsible for assuring that the workers under their In addition, several other Federal agencies can take jurisdiction are adequately protected from radiation actions that affect worker health and safety. The Envi- exposure only; OSHA is responsible for all other ronmental Protection Agency (EPA) regulates pesti- workplace hazards. cides under the Federal Insecticide, Fungicide, and The boundaries of authority are clear in some cases, Rodenticide Act and regulates toxic substances under while in others disputes have arisen. The Federal Avia- the Toxic Substances Control Act. In either case, EPA tion Administration (FAA) has requirements concern- actions to allow, limit, or prohibit the use of particu- ing the health and safety of flight crews, but coverage lar substances may affect employee health and safety. of aviation ground crews has been a focus of dispute In fact, in many cases, the exposed workers may be between the FAA and OSHA. the group most affected by these actions. This may Certain jurisdictional uncertainties have been re- also happen with actions taken by the Consumer Prod- solved by agreements between OSHA and other agen- uct Safety Commission in regulating hazardous con- cies. The Department of Energy, through a letter of sumer products. understanding, has responsibility to “prescribe and enforce occupational radiological and nonradiological safety and health standards” for the workers it cov- Comparison of Protective Levels ers. That 1974 agreement reaffirmed a 1964 letter of understanding between the then Atomic Energy Com- The NIOSH list in Summary of NIOSH Recommen- mission and the Department of Labor concerning dations for Occupational Health Standards contains responsibilities under the Walsh-Healey Act. recommendations for a total of 163 hazardous sub- Recently, Congress temporarily transferred jurisdic- stances and work conditions. There are 74 substances tion over stone and gravel quarries from the Mine which have no complications, and these are included Safety and Health Administration to OSHA for sev- on the comparison list. There are also 11 groups of eral months. Inspection authority for this industry has 71 separate substances for which NIOSH has made rec- now returned to MSHA. Current and future jurisdic- ommendations. Only 43 of these, however, were con- tional disputes may be resolved through letters of un- ducive to comparison. In addition, there are six sub- derstanding and inter-agency agreements, or through stances in three classes which OSHA or ACGIH treat congressional and court actions. separately, but NIOSH treats the same. These are cad- The employees of the Federal Government, as well mium, which OSHA separates into dust and fume; as of State and local governments, are not directly reg- PCBs, which are divided by the percent of chlorine ulated by OSHA. However, Section 19 of the OSH present; and the explosive nitro compounds, nitroglyc- Act requires that the head of each Federal agency pro- erin and ethylene glycol dinitrate. Finally, 10 NIOSH vide an occupational safety and health program for recommendations cover exposures to general catego- agency employees that is “consistent with” the stand- ries of toxic substances or harmful physical agents, ards issued by OSHA. Three different Presidents have while 5 others cover hazardous working conditions. issued Executive Orders concerning the health and These are described in chapter 12, but because most safety of Federal workers (Executive Order (E. O.) of them are not easily compared on a numerical basis, 11612, July 26, 1971; E.O. 11807, Sept. 28, 1974; E.O. they were excluded from this comparison. Thus the 12196, Feb. 26, 1980). There is a Federal Advisory total number of Protective Levels compared equals 74 Council on Occupational Safety and Health, appointed plus 43 plus 6 or 123. by the Secretary of Labor, that consists of 16 mem- Table A-13 presents the numerical Protective Levels bers--8 representing Federal agencies, and 8 represent- from OSHA, NIOSH, and ACGIH that were com- 3$2 ● Preventlng Illness and Injury in the Workplace pared. Alternative chemicals names are not used in recommendation or standard for a particular sub- table A-13. In most cases, the name used is the one stance, the word “NONE” is capitalized and present NIOSH uses. Abbreviations have been included in in both exposure limit columns. most cases for those substances which have them, and, Approaches differ among OSHA, NIOSH and in fact, some substances are seldom referred to by their ACGIH. For example, many of NIOSH’s recommen- chemical names, abbreviations being more convenient. dations are based on a lo-hour workday and not an In this table, all protective levels are listed in mg/m3 8-hour workday as are OSHA’s PELs. For this com- (milligrams substance/cubic meter of air). Generally, parison, it was assumed that this difference would have the protective levels in the actual recommendations only a negligible effect on the level of protection. and standards are given in ppm (parts per million) or For most substances, NIOSH recommends only one mg/m3 or both. For convenience and ease of compari- TLV (98 cases out of 131), either a TWA or a Ceiling son, all ppm concentrations were converted to mg/m3 Limit, but not both. OSHA has only one PEL, an 8- using the formula: hour TWA, for most of the substances it covers. On (MW X (X) ppm)/24.45 = Y mg/m3 the other hand, ACGIH recommends both a TWA and (at 250 C and 760 mm Hg pressure, where MW = Molecular a Ceiling TLV in over half of the cases included in this Weight). comparison (73/131). With differing specifications Table A-13 lists 123 toxic and hazardous substances concerning the type of Protective Level, it can be dif- and the corresponding Time-Weighted Average ficult to compare them. In addition, recommendations (TWA) and Ceiling permissible exposure limits for that no exposure be allowed for carcinogens is often each substance that are recommended by NIOSH and not reflected in the numerical levels recommended by ACGIH, and mandated by OSHA. The 123 chemicals an organization. included in the comparison are all those that appear There are also differences in defining specific sub- on the NIOSH list that also appear on either the OSHA stances since some descriptions are more inclusive than or ACGIH lists. The names of the NIOSH list sub- others. For example, ACGIH has four TLVs for as- stances that were left out for various reasons are listed bestos (one for each type), while NIOSH has a single in the Notes (No. 36). protective level. A similar problem occurs if the sub- When there is only one exposure limit in a protec- stances being compared are not exactly the same, or tive level the word “none” in small letters indicates if related substances are grouped differently, then the which exposure limit is not part of the standard. For standards limiting exposure will differ. An example of example, “none” under the NIOSH Ceiling Limit for this is the different exposure limits for soluble chro- carbaryl means that the NIOSH recommendation does mium, insoluble chromium, chromous salts, and not have a Ceiling exposure limit for carbaryl, but it chromic acid. These are detailed in the notes to table does have a TWA exposure limit. When there is no A-13. App. A—Supplemental Information on OSHA and NIOSH 363

I 364 . preventing Illness and Injury in the Workplace

\ —— . —

App. A—Supplemental Information on OSHA and NIOSH ● 365

Table A-3.—Court Cases Involving OSHA Health Standards*

Access to Employee Exposure arid Medical Records-Louisi- 5, 1983). Public Citizen’s Health Research, et al,, v. ana Chemical Association et al. v. Bingham et al.—Fifth Auchter, et al., 702 F.2d. 1150 (D.C. Cir., Mar. 15, 1983)– Circuit Court of Appeals remanded this case to the U.S. Public Citizen requested a court order compelling OSHA District Court for the Western District of Louisiana 657 to issue an emergency temporary standard. The District F. 2d. 777 (5th Cir., 1981). District Court affirmed the stand- Court decided to issue such an order. The case was ap- ard, 550 F. Supp 1136 (1982); Fifth Circuit Court of Appeals pealed to the D.C. Court of Appeals, which refused to or- affirmed, without opinion, the decision of the District der that an emergency temporary standard be issued, but Court (May 16, 1984). did order that OSHA expedite its section 6(b) rulemaking. Acryionitrile— Vistron v. OSHA (6th Cir., Mar. 28, 1978), emer- Fourteen Carcinogens —Dry Color Manufacturing Associa- gency temporary standard contested, request for stay of tion v. Department of Labor, 486 F.2d 98 (3d Cir., Oct. 4, standard was denied, 6 OSCH 1483. The petition for review 1973)—Vacated the emergency temporary standard for was then withdrawn. two of the fourteen carcinogens. Arsenic (lnorganic)–ASARCO Inc. et al. v. OSHA, 746 F.2d Fourteen Carcinogens-Synthetic Organic Chemical Chem- 483 (9th Cir., Sept. 13, 1984)-Court remanded arsenic ical Manufacturers Association v. Brennan, 503 F.2d 1155 standard to OSHA (Apr. 7, 1981). After OSHA developed (3d Cir., Aug. 26, 1974)–Affirmed standard for ethyl- a risk assessment to comply with the Supreme Court’s eneimine under the 14 Carcinogens standard (SOCMA l). ruling in the Benzene case, the Ninth Circuit Court of Ap- A petition for rehearing was denied Oct. 6, 1975. The Su- peals affirmed the arsenic standard. preme Court denied a request for review, 420 U.S. 973 (Mar. Asbestos—industrial Union Department, AFL-CIO v. 17, 1975). Hodgson, 499 F.2d 467, (D.C. Cir., Apr. 15, 1974)—Affirmed Fourteen Carcinogens-Synthetic Organic Chemical Manu- OSHA’s 1972 asbestos standard. facturers Association v. Brennan, 506 F.2d 385 (3d Cir., Asbestos— Asbestos Information Association/North Amer- Dec. 17, 1974)—Third Circuit Court of Appeals vacated the ica v. OSHA, 727 F.2d. 415 (5th Cir., Mar. 7, 1984)-Vacated standard for MOCA (1 of the 14 carcinogens) (SOCMA II). the emergency temporary standard issued on Nov. 4, 1983. The Supreme Court denied a request for review. Oil, Chem- Benzene–American Petroleum Institute v. OSHA, 581 F.2d ical and Atomic Workers International Union, AFL-CIO v. 493 (5th Cir., Oct. 5, 1978); Industrial Union Department, Dunlop, 423 U.S. 830 (Oct. 6, 1975). AFL-CIO v. American Petroleum Institute, 448 U.S. 807 (Su- Hazard Communication (Labelling--United Steelworkers of preme Court, July 2, 1980)-Both the 5th Circuit Court of America, Public Citizen, State of Massachusetts, Fra- Appeals and the Supreme Court vacated the OSHA ben- grance Materials Association, People of the State of II- zene standard, although for different reasons. iinois, Flavor& Extract Manufacturing Association, State Cancer Policy—American Petroleum Institute, et al. v. OSHA, of New York v. Auchter, Nos. 83-3554, 83-3561, 83-3565, et al., Nos. 80-3018, et al. (5th Cir., pending). 84-3066, 84-3087, 84-3093, 84-3117, 84-3128 (3d Cir., Coke Oven Emissions—American Iron & Steel Institute v. pending). OSHA, 577 F.2d 825 (3d Cir., Mar. 28, 1978)—Third Circuit Occupational Noise Exposure/Hearing Conservation Court of Appeals largely affirmed the Coke Oven Emis- Amendment —Forging Industry Association v. Sec. of La- sions standard. The Supreme Court agreed to review this bor No. 83-1232 (4th Cir., Nov. 7, 1984)-Fourth Circuit decision, but the request for review was withdrawn before Court of Appeals vacated the Hearing Conservaton the case could be heard. 448 U.S. 917 (1980) Amendment. Cotton Dust—AFL-CIO v. Marshall, 617 F.2d 636 (D.C. Cir., Lead–United Steelworkers of America v. Marshall, 647 F.2d Oct. 10, 1979); American Textile Manufacturers Institute, 1189 (D.C. Cir., Aug. 15, 1980)-The D.C. Court of Appeals Inc. v. Donovan, 452 U.S. 490 (June 17, 1981)-D.C. Court affirmed the lead standard in part, but directed OSHA to of Appeals and the Supreme Court both upheld the ma- determine the feasibility of engineering controls for 38 in- jor requirements of the cotton dust standard as applied dustries and occupations. The Supreme Court denied a to the textile industry. request for review Lead Industries Association, Inc. v. Cotton Dust–Cotton Warehouse Association v. Marshall, 449 Donovan, 453 U.S. 913 (1981) U.S. 809 (Oct. 6, 1980)-Supreme Court granted a petition Pesticides—Florida Peach Growers Association, Inc. v. De- for review and vacated the decision of the court of appeals partment of Labor, 489 F.2d 120 (5th Cir., Jan. 9, 1974)– with respect to the warehousing and classing segments The Fifth Circuit Court of Appeals vacated the emergency of the industry. temporary standard for pesticides. Cotton Dust—Texas Independent Ginners Association v. Mar- Vinyl Chloride-Society of the Plastics Industry, Inc. v. OSHA, shall, 630 F.2d 398 (5th Cir., November 14, 1980)-Vacated 509 F.2d 1301 (2d Cir., Jan, 31, 1975)—The Second Circuit cotton dust standard as applied to cotton ginning oper- Court of Appeals affirmed the vinyl chloride standard. The at ions. Supreme Court denied a request for review Firestone Ethylene Oxide—Public Citizen Health Research Group, et Plastics Co. v. U.S. Department of Labor 421 U.S. 992 (May al. v. Auchter, 554 F. Supp. 242 (D.C. District Court, Jan. 27, 1975).

Court Cases Involving OSHA Safety Standards Lavatories for Industrial Employment-Associated industries 109 (3d Cir., Dec. 31, 1975)–The Third Circuit Court of Ap- of New York State, Inc. v. Department of Labor, et al., 487 peals remanded to OSHA for a new statement of reasons F.2d 342 (2d Cir., Oct. 4, 1973)—The Second Circuit Court and then affirmed OSHA’s changes to the “no hands in of Appeals vacated the OSHA lavatory standard. die” standard. Mechanical Power Presses-AFL-CIO v. Brennan, 530 F.2d Commercial Diving Operations— Taylor Diving and Salvage 366 ● Preventing Illness and injury in the Workplace

Table A-3.—Continued

v. U.S. Department of Labor 537 F.2d 819 (5th Cir.,1976)— Court of Appeals for the D.C. Circuit remanded the rec- Court issued an indefinite stay of the ETS for commer- ord to OSHA with specific instruction to consult the Advi- cial diving. sory Committee. Commercial Diving Operations– Taylor Diving and Salvage Fire Protection-Fim Equipment v. Marshall 679 F.2d 679 (7th v. U.S. Department of Labor 599 F.2d 622 (5th Cir., July Cir., May 27, 1982)-case was dismissed for lack of stand- 16, 1979)—Vacated the medical requirements section (29 ing. Request for rehearing was denied (July 22, 1962). CFR 1910,411) of the final standard for commercial diving. Industrial Slings--Bethlehem Steel Corp. v. Dunlop 540 F.2d, Diving Exemptions-United Brotherhood of Carpenters and 157 (3d Cir., Feb. 11, 1976)—Vacated one paragraph of the Joiners of America, AFL-CIO v. U.S. Department of Labor, standard (29 CFR 1910.184) and remanded the standard No. 82-2509, D.C. Cir,, Apr. 4, 1984)-After oral argument, to the Secretary of Labor. court remanded case to OSHA for additional information. Marine Terminals--National Grain and Feed Association (D.C. Ground-Fault Protectlon--National Constructors Association Cir., pending). v. Marshall 581 F.2d. 960 (D.C. Cir., June 28, 1978)—The

“NOTES: F.2d—FederaJ Reporter, Second Series. U. S.–U.S. Supreme Court Reports. F. Supp.—Federal Supplement.

Table A-4.-Safety and Health Inspections

Faderal OSHA: Establishment Safety Safety Health Health Employees covered inspections inspections inspections inspections inspections by inspections Fiscal year (number) (number) (percent) (number) (percent) (number) 1973 ...... 48,409 45,225 93.4 ”/0 3,184 6.6% 5,440,303 1974 ...... 77,142 73,189 94.9 3,953 5.1 6,448,067 1975 ....., ...... 80,978 75,459 93.2 5,519 6.8 6,180,881 1976 ...... 90,482 82,885 91.6 7,597 8.4 6,601,729 1977 ...... 60,004 50,892 84.8 9,112 15.2 5,285,946 1978 ...... 57,278 46,621 81.4 10,657 18.6 4,522,582 1979 ...... , . . . . 57,734 46,657 80.8 11,077 19.2 4,262,749 1980 ...... 63,404 51,565 81.3 11,839 18.7 3,690,993 1981 ...... 56,994 46,236 81.1 10,758 18.9 2,672,129 1982 ...... 52,818a 43,609 82.6 9,209 17.4 2,235,823 1983 ...... 58,516 b 48,269 82.5 10,247 17.5 2,925,049 1984 (Oct. -Mar.) . . 30,606 c 25,086 82.0 5,520 18.0 1,552,120

State programs: Establishment Safety Safety Health Health Employees covered inspections inspections inspections inspections inspections by inspections Fiscal year (number) (number) (percent) (number) (percent) (number) 1976 d ...... 166,612 144,780 86.9 21,832 13.1 7,078,294 1977 143,469 130,643 91.1 12,826 8.9 6,000,009 1978 ...... 122,761 112,446 91.6 10,255 8.4 5,739,574 1979 ...... 107,636 99,509 92.4 8,127 7.6 4,932,303 1980 ...... 106,191 98,829 93.1 7,288 6.9 4,340,266 1981 ...... 108,376 99,303 91.6 9,073 8.4 4,404,364 1982 ...... 92,942 84,570 91.0 8,372 9.0 3,464,146 1983 ...... 103,879 e 93,406 89.9 10,473 10.1 3,818,287 1984 (Oct. -Mar.) . . 51,072 f 46,065 90.2 5,007 9.8 1,858,114 a-s not ~nclude 8,444 “ReCO@i Review” inspections in ftscal Year 1962. b-s not Include 1(),402 “Records Review” inspections in fiSCa\ Year 19M. cooes not include 4,9s4 ‘“R~ords Review” inspections during the first 6 months (Oct. -Mar.) Of fisCal year 1964. dNo data available prior to 19 76 estate data d~s not Include 2,554 “Records Reivew” inspections in fiScal Year f~. fstate data dogs not include 1,~ ‘, Records Review” ~ns~ctions during the first 6 months (Oct,-Mar.) of fiSCal year 1964,

SOURCE’ Office of Technology Assessment, based on data supplied by OSHA. — —

App. A—Supplemental Information on OSHA and NIOSH ● 367

Notes to Tables A-4 Through A-11.-State Program Data Table A-5-Types of Inspection

State programs Establishment Fatality/ inspection catastrophe catastrophe Complaint Complaint Programed Programed Follow-up Follow-up Fiscal year (number) (number) (percent) (number) (percent) (number) (percent) (number) (percent) 1976 ...... 166$612 4,278 2.6% 13,966 8.4% 119,120 71.5% 29,216 17.5% 1977 ...... 143,469 3,652 2.5 14,404 10.0 101,571 70.8 23,842 16.6 1978 ...... 122,761 4,609 3.8 15,467 12.6 81,762 66.6 20,923 17.0 1979 ...... 107,636 5,181 4.8 15,285 14.2 70,762 65.7 16,408 15.2 1980 ...... 106,191 5,264 5.0 13,823 13.0 72,899 68.6 14,168 13.3 1981 ...... 108,376 5,259 4.9 14,365 13.3 75,839 70.0 12,858 11.9 1982 ...... 92,942 4,663 5.0 10,721 11.5 68,100 73.3 9,455 10.2 1963 ...... 103.879 5.366 5.2 11.623 11.2 78.796 76.0 8.094 7.8 1984 (Oct.-Mar.).. 51,072 2,849 5.6 5,754 11.3 39,085 76.5 3,384 6.6 SOURCE:OtfkeofToch_A 8$e88mmt,ba8edoncW88upplledbyOSHA. App. A–Supplemental In formation on OSHA and N!OSH ● 369 Inapeotlonswith Eatdmtwnent lnapection8with tnapectionatith Inspectionswlth other~a In$pectionu inspections Seriousviolatfons willful violations reoe8tvtolations Yiolutions contested Flacal Year (numbed (number) (percent) (numberj (percent) (number) (percent) (number) (percant) (number) (percent) - 1973 ...... 48.409. 1,535 3.2% 0.0% 0.1% 23,814 49.2% 1,3 5 2.7% 1974 ...... 77;142 3-5 z 2 0.6 48#24 62.3 2,4 17 3.2 1975 ...... 80,978 :s 4.1 104 ::: 1,175 1.5 50,985 63.0 3,1 la 3.9 1976 ...... 5*W6 6.6 153 0.2 2a 2.4 59$)81 65.3 5,0 t7 5.5 1977 ...... ::E 11,0%? 18.5 169 0.3 2,356 3.9 31,126 51.9 4,2 0 7.0 1978 ...... 57~78 14,620 25.5 428 0.7 &191 3.8 25~7 U.1 ‘4 9.6 1979 ...... 57,734 16,624 587 1.0 2#3 3.9 25,068 43.4 :; 13 11.6 la m ...... 63,404 19,358 m.5 0.9 2s)21 3.2 27,366 432 7,3 n 11.7 1981 ...... 16~7 28.5 241 1,318 2.3 X$717 46.9 &5 t2 6.3 1882 ...... :’RI 12,852 24.3 86 :: 810 1.5 26,187 49.6 1,4 ‘o 28 1963 ...... $%6 14#86 25.4 105 0.1 1#132 $.7 30,472 52.0 1,1 k2 1.9 1964(oct.-Mar.) 8.156 26.7 65 02 610 2.0 15201 48.7 5 u 1.9

-~ Esttiishment htapectiata hwpections contested Fiaod year (number) (number) (-) 1876 ...... 166,612 &277 3.8% 10?7 ...... 143#68 5,024 3.5 1878...... 1~761 4,703 3.8 1979 ...... lo7,m 5,171 4.8 1860 ...... 106,191 CompanMe data for state programs ara not readily available 4,308 4.1 IWI ...... 166#76 4,452 4.1 1862 ...... %2,842 3* 3.5 1W3 ...... 103,878 3,322 1964 (oCt.-Mar.) 51,072 1,686 :; 80URCE: Offia of 1~ ~t, ~ on -s sw@iod by OSHA. App. A—Supplemental Information on OSHA and NIOSH . 371

. . ,. ,...... , ...... ,...... , ...... , ...... ,...... ,...... !. , ...... ,...... 372 Preventing Illness and Injury In the Workplace

...... ,, ...... ,. . . . .,,, . . ,. ..., . . . ..,...... , . . . App. A—Supplemental Information on OSHA and NIOSH 373

T-F 7-

U 374 . Preventing Illness and Injury In the Workplace

Table A-11.--Average Proposed Penalties

Federal OSHA Violations Other-than Serious Willful Repeat serious (average (average (average (average Average penalty Average penalty Fiscal year penalty) penalty) penalty) penalty) per inspection per violation 1973 ...... 631 5,805 1,349 45 78 26 1974 ...... , . . . . 576 2,706 247 41 84 22 1975 ...... 541 2,538 228 42 93 24 1976 ...... 545 3,081 207 42 124 30 1977 ...... 290 2,990 242 61 158 52 1978 ...... 285 3,460 429 88 259 111 1979 ...... 273 3,750 429 94 291 131 1980 ...... 255 3,244 478 110 280 135 1981 ...... 209 3,660 90 177 91 1982 ...... 195 4,364 320 86 104 57 1983 ...... 177 4.555 346 74 179 257 1984 (Oct.-Mar.) . . 187 4;977 408 101 211 68

State programs Violations Other-than- Serious willful Repeat serious (average (average (average (average Average penalty Average penalty Fiscal year penalty) penalty) penalty) penalty) per inspection per violation 1976, ...... 420 3,615 115 33 39 15 1977, ...... 293 2,487 100 39 35 1978...... 210 3,174 197 68 51 1979...... 239 2,869 251 74 76 30 1980 ...... 248 3,592 253 63 75 33 1981 ...... 211 3,156 450 75 65 30 1982 ...... 201 3,146 230 101 54 26 1963 ...... 167 3,112 191 139 68 31 1984 (Oct.-Mar.) . . 176 4,055 213 138 73 34 SOURCEOfflceof Technology Assessment, based on data suppliedby OSHA, Table A-12.—Occupational Safety and Health: Coverage of Workers

Agency Type of workers covered Number of workers covered Basis for agency authority Comments Department of Labor Occupational Safety and Health Administration (OSHA) and State Programs approved by OSHA ...... All employees and working 75,031 ,OOOa (1979 estimate) Occupational Safety and Health In some cases, another Federal conditions except: Federal Act of 1970 agency is responsible for only employees, and those covered certain aspects of safety and by other governmental health, and the same workers agencies according to other may be covered by OSHA for statutes the remaining aspects (see e.g., NRC in this table) Mine Safety and Health Administration (MSHA)...... Coal, metal and nonmetal 467,095 (1 2 preliminary Federal Mine Safety and Health mining workers. All employees estimate)b Act of 1977 on mine property are covered Dapartment of Transportation: Bureau of Motor Carrier Safety (BMCS) (Federal Highway Administration) ...... Employees in, on, or about Approximately 4.5 millionc Interstate Commerce Act Does not include workers in repair motor vehicles engaged in garages, or workers on loading interstate commerce docks, who are all covered by OSHA Federal Aviation Administration (FAA) ...... All flight crews; ground crews Approximately 170,000d Federal Aviation Act of 1956 Coverage of ground crews is the and mechanics during some focus of a dispute between the activities FAA and OSHA Federal Railroad Administration (FRA) ...... All operating employees, i.e., 143,617 (1979 preliminary Federal Railroad Safety Act of employees on rolling stock estimate) e 1970 plus certain railroad yard employees U.S. Coast Guard...... Seamen on Coast Guard- About 100,000f (1963 estimate) The Marine Safety Laws OSHA has jurisdiction over inspected and certificated shipyard workers and vessels longshoremen Other Federal Agencies: Department of Energy (DOE). . . . Employees in Government- 116,323 g (1962 estimate) Atomic Energy Act of 1954, as DOE has adopted OSHA’s health owned contractor operator amended and safety regulations; DOE (GOCO) facilities, e.g., those does not cover employees involved in research in nuclear during initial construction of energy, weapons research and facilities production, production of enriched uranium. Nuclear Regulatory Commission (NRC) ...... Workers exposed to radiation 327,350 h (1979 estimate) Atomic Energy Act of 1954, as NRC covers only radiation hazards from materials amended hazards; OSHA is responsible licensed by the NRC, for all other safety and health including: 1) source material aspects. NRC licenses State (uranium and thorium); 2) plans in some States, similar to special nuclear material OSHA State Programs (material capable of being fissioned); 3) by-products of a) fission; and b) tailings from uranium ore processing . .

Table A-12.—Occupational Safety and Health: Coverage of Workers-Continued

Agency Type of workers covered Number of workers covered Basis for agency authority Comments Federal Government departments and independent agencies ...... Each covers its own federally- 6,271,736 i(fiscal year 1962) Occupational Safety and Health Agency programs must be employed workers. Act of 1970 “consistent with” occupational safety and health standards promulgated by OSHA Environmental Protection Agency ...... Mixers, loaders, and applicators Federal Insecticide, Fungicide, For mixers, loaders and of pesticides; farm field and Rodenticide Act applicators, protection from workers pesticide exposure is through labeling requirements. OSHA has jurisdiction for other health and safety aspects of these jobs. %fflce of Statistical Studies and Analysis, OSHA. Includes all private-sector employees covered directly by Federal OSHA and State Programs. bperwn~ communication, MSHA, Mar. 3, l=. cper~ond communication, BM~.

dpersoal communications: Air Transport Association and Regional Aifllne Association eyea~k of R&”/o@ Facts, June 1~. fperWnal communication, U.S. ~~t GIJWd. gDOE, Rep@ of Emp/Oymerrt &@ Labor Turnover, Sept. 30, 19E2. hNRC ‘+occupation~ Radiation Exposure, Twelfth AnnUa/ Repofi, ~gT9 (l~z). Ius, ~epa~menr of L~r, Fadera/ Compliance ActMty Ftepofl, Jan. 4, 1~ SOURCE: Office of Technology Assessment. —

App. A—Supplemental Information on OSHA and NIOSH● 377

Table A-13. -Analyses of OSHA, NIOSH, and ACGIH Protective Levels

OSHA OSHA NIOSH NIOSH ACGIH ACGIH TWA ceiling TWA (1) ceiling TWA ceiling Substance (notes) (36) mg/m 3 mg/m3 mg/m3 mg/m 3 mg/m3 mg/m 3 Acetylene (10) ...... 2,662 none none 2,662 NONE NONE Acrylamide (35) ...... 0.3 none 0.3 0.3 0.6 Acrylonitrile (11) ...... 4.3 21.7 none 4.5 none Aldrin/Dieldrin (12,27,35) ...... 0.25 none 0.15 0.25 0.75 Alkanes: (14) Pentane ...... 2,945 none 350 1,800 2,250 Hexane ...... 1,800 none 350 180 none Heptane ...... 2,000 none 350 1,600 2,000 Octane ...... 2,350 none 350 1,450 1,800 Allyl chloride ...... 3 none 3.1 3 6 Ammonia ...... none none 27 Antimony ...... none 0.5 none Arsenic, inorganic compounds ...... 0.01 none none 0.2 none Asbestos (9)...... 2 10 0.1 2 none Asphalt (petroleum) fumes ...... NONE NONE none 10 Benzene (2,16) ...... 32 80 none 75 Benzoyl peroxide ...... 5 none 5 5 none Benzyl chloride ...... 5 none none 5 none Beryllium (2,16) ...... 0.002 0.005 0.0005 0.002 none Boron trifluoride (13,15) ...... none 3 NONE none 3 Cadmium, fume (2,23)...... 0.1 0.3 0.04 0.05 0.2 dust (2,23) ...... 0.2 0.6 0.04 0.05 0.2 Carbaryl ...... 5 none 5 5 10 Carbon black (18) ...... 3.5 none 3.5 3.5 Carbon dioxide (17)...... 9,000 none 18,000 9,000 27,000 Carbon disulfide (2) ...... 62 93 3 30 none Carbon monoxide ...... 55 none 40 55 Carbon tetrachloride (2,16) ...... 63 157 none 30 125 Chlorine ...... none 3 none 3 9 Chloroform (15,16)...... none 240 none 50 225 Chloroprene (35) ...... 90 none none 45 none Chromium (VI), water soluble(3) ...... 0.5 none 0.025 0.05 none Chromium (VI), insoluble (3) ...... 1 none 0.001 0.05 none Coal tar products (5)...... 0.2 none 0.1 0.2 none Cotton dust(6)...... 0.2 none 0.2 0.2 none Cresol ...... 22 none 10 22 none Cyanide (17,35) ...... 5 none none 5 none DDT (26,37,35) ...... none 0.5 1 3 l,2-dibromo-3-chloropropane (DBCP) (16,34) . . . . . 0,0096 none none none none Diisocyanates: Toluene-2,4-diisocyanate (15) ...... none 0.14 0.035 0.04 0.15 Diphenylmethane diisocyanate (13,15)...... none 0.2 0.05 none 0.2 Isophorene diisocyanate ...... NONE NONE 0.045 0.09 none Dinitro-ortho-cresol (35) ...... 0.2 none 0.2 0.2 0.6 Dioxane (35)...... none none 90 Epichlorohydrin (35) ...... 20 none 2 10 20 Ethylene dibromide (2) ...... 154 230 none none none Ethylene dichloride (2) ...... 202 405 40 60 Ethylene oxide (27,37) ...... 90 none 2 none Fibrous glass, (dust) (29) ...... 15 none 5 10 none Fluorides, inorganic(2) ...... 2.5 none 2.5 2.5 none Formaldehyde (2,13,16)...... 3.7 6 none 1.5 3 Furfuryl alcohol ...... , ...... none 40 60 Glycidyl ethers: Allylglycidyl ether (15) ...... none 45 none 22 44 n-Butyl glycidyl ether ...... 270 none none 135 none Di-2,3-epoxypropyl ether (diglycidyl ether) (DIE) ...... none 2,8 none 0.5 none Isopropyl glycidyl ether ...... 240 none none 240 Phenylglycidyl ether (PAGE ) ...... 60 none none 6 none 378 ● Preventing Illness and Injury In the Workplace

Table A-13.--Analysis of OSHA, NIOSH, and ACGIH Protective Levels

OSHA OSHA NIOSH NIOSH ACGIH ACGIH TWA ceiling TWA (1) ceiling TWA ceiling Substance (notes) (36) mg/m3 mg/m3 mg/m3 mg/m3 mg/m3 mg/m3 Hydrszines: (16) Hydrazine (16,35) ...... 1.3 none none 0.04 0.1 none 1,1-Dlmethyl hydrazine (16,35), ...... 1 none none 0.15 2 Phenyl hydrazine (16,35) ...... 22 none none 0.6 45 Methyl hydrazine (13,15,16,35)...... none 0.35 none 0.08 none 0.35 Hydrogen fluoride (2) ...... 2.45 none 2.5 5 2.5 5 Hydrogen sulfide (2,17) ...... none 16 none 15 14 21 Hydroquinone ...... 2 none none 2 4 isopropyl alcohol...... none 984 1,968 92 1,225 Ketones: Acetone ...... 2,400 none 590 none 1,780 2,375 Methyl ethyl ketone (MEK) ...... 590 none 590 none 590 885 Methyl n-propyl ketone ...... 700 none 530 none 700 875 Methyl n-butyl ketone...... 410 none 4 none 20 none Methyl n-amyl ketone ...... 465 none 465 none 235 465 Methyl isobutyi ketone...... 410 none 200 none 205 300 Methyl isoamyl ketone (20) ...... NONE NONE 230 none 240 none Diisobutyl ketone ...... 290 none 140 none 150 none Cyciohexanone ...... none 100 none 100 400 Mesityl oxide...... none 40 none 60 100 Diacetone alcohol...... none 240 none 240 360 lsophorone (13) ...... 140 none 23 none none 25 Lead, inorganic (33) ...... 0.05 none 0.1 none 0.15 0.45 Malathion ...... 15 none 15 none 10 none Mercury, inorganic (2,4,24)...... 0.1 none 0.05 none 0.1 none Methyl alcohol...... none 262 1,048 260 310 4,4-Methylene-bis-2-chloroaniline (MOCA) (8,27).. NONE NONE 0.003 none 0.22 none Methyl parathion ...... NONE NONE 0.2 none 0.2 0.6 Methylene chloride (2,19) ...... 1,736 3,476 261 1,740 350 1,740 Nickel carbonyl (27) ...... 0.007 none 0.007 none 0.35 none Nickel, inorganic and compounds ...... 1 none 0.015 none 0.1 0.3 Nitric acid ...... 5 none 5 none 5 10 Nitrides: Acetonitrile ...... 70 none 34 none 70 105 Tetramethyl succinonitrile...... 3 none none 6 3 9

Nitrogen, oxides NO2: (15) ...... none 9 none 1.8 6 10 (Nitric oxide) NO: ...... 30 none 30 none 30 45 Nitroglycerin (15,16) ...... none 2 none 0.1 0.5 Ethylene glycol dinitrate (15)...... none 1 none 0.1 0.3 Parathion ...... 0.1 none 0.05 none 0.1 0.3 Phenol ...... 19 none 20 19 38 Phosgene ...... 0.4 none 0.4 0.4 none Polychlorinated biphenyls: (35) Chlorodiphenyl (42%)...... 1 none 0.001 none 1 2 Chlorodiphenyl (54%)...... none 0.001 none 0.5 1 Refined petroleum solvents (7)...... 2,950 none 350 1,800 NONE NONE Silica (quartz, respirable dust) ...... 0.098 none 0.05 none 0.1 none Sodium hydroxide (13) ...... 2 none none 2 none 2 Sulfur dioxide ...... 3 none 1.3 none 5 10 Sulfuric acid...... 1 none 1 none 1 none 1,1,2,2-Tetrachloroethane (35) ...... none 6.87 none 7 35 Tetrachloroethylene (2)...... 1,358 339 678 335 1,340 Thiols: (21) Butyl mercaptan ...... 35 none 1.8 none 1.5 none Methyl mercaptan (15) ...... none 20 1 none 1 none Ethyl mercaptan (15)...... none 25 1.3 none 1 3 Tin, organic compounds ...... 0.1 none 0.1 none 0.1 0.2 o-Toluidine (35) ...... 22 none none 0.02 9 none Toluene (2,17)...... 753 1,129 375 750 375 1,1,1-Trichloroethane ...... 1,900 none none 1,190 1,900 2,450 Trichloroethylene (2) ...... 536 1,072 134 none 270 1,080 App. A–Supplemental Information on OSHA and NIOSH ● 379

Table A-13.-Analysis of OSHA, NIOSH, and ACGIH Protective Levels—Continued

OSHA OSHA NIOSH NIOSH ACGIH ACGIH TWA ceiling TWA (1) ceiling TWA ceiling Substance (notes) (36) mg/m3 mg/m 3 mg/m3 mg/m3 mg/m 3 mg/m3 Tungsten: (31) insoluble compounds ...... NONE NONE 5 none 5 10 soluble compounds ...... NONE NONE 1 none 1 3

Vanadium, as V2O 5 (dust) (15,32) ...... none 0.5 none 0.5 0.5 none (fume) (15,32) ...... none 0.1 none 0.05 0.05 none Ferrovanadium (32) ...... 1 none 1 none 1 3 Vinyl acetate ...... NON E NONE none 15 30 80 Vinyl halides: (22) Vinyl bromide ...... NONE NONE none 4 20 none Vinyl chloride ...... 2.5 13 none 2.55 10 none Vinylidene chloride...... NONE NONE none 4 20 80 Xylene (17) ...... 435 none 434 868 435 655 Zinc oxide (fume) ...... 5 none 5 15 5 10 SOURCE Office of Technology Assessment

1 NIOSHTWA recommmdahons arebasedon upto alo-hourexposure unlesaother- OSHA standarda on August 2s3, 1976 Both N1OSH ● nd ACGIH recommend TWA wse indicated protective levels for MOCA. 2 Un&r OSHAregulatlons, the ceiling levekfor these substances are labelkd’’Acmpt- 9 Asbestos-OSHA, NIOSH, and ACGttf nuasure aafxatoa corrantration by the nurnkr able CeshngConcmtrahon “hatibon,forea ~oftktids them ~an’’Accqt- of fibem longer than s micrornetem per cubic mntimeter of air: e.g., 2 fibem/cc > able Mammum Peak above the acceptable cdmg concentration for an &hour shift, ” 5pm in length In the comparison table the numbers ● re listed under mg/ml for ● asy which I]sts a thwd concentration level and maximum duration. Details can be found comparison, but they have not been converted to rng/m~. m table Z-2 m the OSHA Standards (29 CFR1910,1~). OS-IA and NIOSH have one standard for asbestos, while ACGIH divides asbestos 3 Chrommm (V/)–There are several ways to separate chromium (VI) compounds mto mto four types with different recommendations for ● ach. However, the standard for different classifications The difficulty m comparing the recommendations and stand- cfvysotile (the moat common) is used in the table. The rest of the ACGIH list con. ards is that each organization uses different classifications Chromium (VI) can be tains amosite (0.5 fibers/cc > sw in length), crocidolite (0.2 fibers/cc > Sum in classified carcinogenic or not, soluble and nonsoluble, salts, and chromates. Chromic length), and other forms (2 fibers/cc > 5pm in length). ACGIH does not indicate aod IS chromium (VI) oxide and includes aqueous solutions thereof whether this is a TWA or ceiling TLV recommendation; in the comparison table it The OSHA standard for chromium (VI) separates chromium into “soluble chromic is plad under TWA. and chromous salts” and “metal and resoluble salts,” both havingdiffermt PELs. These 10 Acetylene-The OSHA PEL for ● cetylene is not in the OSHA health standards list values can be found m OSHA table Z-2 There )s a different PEL for chromic acid (29 CFR 1910. IIXM3) but the N1OSH summary booklet does list one; this is the protec- and chromates tive Iwel used in the comparison table. There is an added note un&r the OSHA PEL, The 1975 N1OSH criteria document for chromium (VI) revises the 1973 recommen- “1O percent of lower exposure limit, ” with no further explanation. The NIOSH rec- dations for Chromic acid Chromic acid (or chromic acid anhydri&) is an oxide of ommendation for acetylenesays, ‘Occupational exposure to● irborne ● mtylene shall chrom]um (VI} and IS clas.wfwd as a noncarcusogenic chromium (Vi) Un&r N1OSH be controlled so that no employees will b exposed to amtylene at ● concentration recommendations, there are two recommended standards for chromium (VI) One in excess of 2,5oo ppm. This is not the same ● s an “acceptable cAing comtration. ” standard addresses Occupational exposure to a group of noncarcmogenic, but other- However, to allow some comparison among standarda, the recommended TLV is in wise hazardous, chromium materials, while the other standard covers occupahonal the ceiling column. ACGIH classifies amtylene as a simple asphyxiant; a TLV is not exposure to other chrom]um materials that are associated with an increased Incidence recomrnmded because the limiting factor is the availabk oxygen. ACGLH accutrtpaniea of lung cancer However, there N no practical means of distinguishing between these the desaiption of simple asphyxiants with warnings, including the additional fact that two groups on the basis of chemical analysis of airborne materials Until the airborne several simple asphyxiants presen t an explosive haxard. chromium in a partscuhr workplace k demonstrated to be of the noncarcincgenic type, 11 Acryfonitrilc The OSHA PEL for ● crylonitrile is in section 191O.1W of the OSHA all alrbome chromium IS consmfa-ed to be carcinogenic standards. In addition to tk TWA and ctiling PELa, the standard states that “tk emp- ACGIH recommends two standards for chromium (VI) by separating the compounds loyer shall aaaure that no employee is exposed to akin contact or eye contact with into “water soluble” and “certain water insoluble” compounds which are labelled car- liquid ● crylonitrile. ” ACGIH classifies acrylonitrile as a human carcinogen with an cinogenic However, the TLV IS the same for both ty-pes aaaigned TLv, This companscm table deals with this ambiguity by lwting two standards for 12 Aldrirr/~”eldrfn-NIOSH recommends as an exposure level tk ‘lowest reliably de- chrom]um (W) water soluble (non-carcinogenic), and resoluble, metal and salts (car- tectable lwel; 0.15 mg/mD TWA by NIOSH validated method. ” All three organiza- cmogenlcl Under the OSHA category, only the chromium values m the table Z-I tions indicate that skin contact u to be avoided. are used, chromic acid m table Z-2 IS gnored 13 The TLVS under ACGIH for the following chemicals ● re ● bsolute ceili~ limits, the 4 Mercury Under the OSHA standards, mercury IS listed m table Z-2 The PEL for mer- concentration of which should not exceed tk ceiling limit even inatantaneossaJy: boron cury is 1 mg /10 ml This IS equal to O 1 mg/m3 trifluorick, diphenylmethane diiaocyanate (MDI), methyl hy&azine, iaophorone, 5 Coal Tar Products-Protectwe Levels for coal tar products, or coal tar pitch volatiles, sodium hydroxide, and fonnal&hyde. This fact should be taken into conskleration are mwleading and difficult to compare because the methods of measuring ● xposure when doing a direct comparison among recommendations and standards. levels differ among OSHA, NIOSH, and ACGIH The OSHA PEL addresses the 14 Alkunes-Tbe NIOSH recommendation, unlike OSHA’S or ACGIH’S, have a protective benzene soluble frachon (anthracme, BaP, phenanthrene, acndme, chrysene and Iwel for mixtures of alkanea: “noetnployee shall b expoaed to individual W akattes pyrene), as does the ACGIH recommended TLV The NIOSH recommendation per- or mixtures of these alkanes at ceiling concentrations greater than l,WO rng/ml ● s tains to the cyclohexane extractable fract]on The comparison table contains all the determined over a sampling time of 15 minutes.” protectwe levels but direct comparison IS not apphcable in this case. 15 Un&r OSHA standards, the following subatanw have only ceiling PEls. Exposure 6 Cotton Dus—OSHA standards for cotton dust are set out in 19101043 of the stand- to these chemicals “shall at no time exceed the ceilingvafue given fo~ that materials, ” ards. There are three different standards for three major processes, yam manufactur- The chemicals regulated in this way are boron hifhsori&, chlorine, chloroform, mg, dashing and weaving, and all other operations. The PEL for yam manufacturing methylene biphenyl isocyanate (MDI), toluene-2,4-diisocy anate, allyglycidyl ● ther IS the most stringent, and this IS the protechve level that the compariscm table !Ists (AGE), diglycidyl ● ther (EXE), methyl hydruine, nitroglycerin, ethylene glycol NIOSH has only one standard for cotton dust (described as hnt-free cotton dust). dinitrate, nitrogen dioxide, ● thyl mercaptan, methyl mercaptan, and vanadium (dust ACGIH also recommends lust one standard for %nt-free dust According to all three and fume). standards (OSHA, NIOSH, and ACGIH), measurement of the ● xposure level is by 16. Long maximum exposwe-lhe NIOSH recommendation for benmw and carbon tetra- the vertical elutnator cotton dust sampler. chloride lists the maximum time Iimtt for exposure to conmntrationa at or above the 7 Refrned petroleum soluents-OSHA l]sts these substances simply as “petroleum dls- ceiling limit as 60 minutes (not the usual 15 minutes). Other maximum time limits hllates (naphtha) “ The N1OSH recommendation controls exposure to petroleum ether, for exposure to conswntrations at or above the N1OSH ceiling limit are as follows: rubber solvent, vammh makers’ and painters’ naphtha, mineral spirits, Stoddard beryUiurn (130 minutes), fo m’td&hy& (30 minutes), hydrazines (120 minutes), chloro- solvmts and kerosene The asurrphon is that the recommendation for these substanms form (60 minutes), dhxrmchloropropane (30 minutes), and nitroglycerin (Xl minutes), are equivalent and are measured the ~me way 17. Short Mu.mmum /Gposure-Un&r the NIOSH recommendation, tk maximum time 8 4 4 -Methylent--b[s (2-chloroarnl ljne)-The OSHA standard for limit is 5 mmutes for worker exposure to concentrations of ammonia at or above the 4,4 -methylme-bw-2-chloroandme (MOCA), section 1910 1c05, was deleted from the acceptable ctiling limit, Other maxiumum time limitg for exposutw to concmrtrationa 380 . Preventing Illness and Injury In the Workplace

● ▼ or above & aclaptabk MosH dung limit am ae follow,: Cubon diaide (lo rnln- Cymnide, exp8ure to be mnddemd Xiditive and en virmmentd limit to be calculated.” uta), cyanide (10 mfmltd, h- euifide (10 ndnuta), tofuene (10 minutes), xykne MOSH mcommdo TLVS for ten nitrlk compounds; TWA values for dx compmud (10 lnlnu=). and cviling vdtee for three compounds. CEHA and ACGIH set TLW for onfy two 18. Carbon bkck-NIOSH be two recommended TWAe for cubon bkck. When W- nitdfe compouA. The cornpubon table incfudes d * nltrik compound8 for Whidl bon M ie in the ~ of pCdyC’ydiC h~N, the recommended TWA & N1OSH eets cdhng TLVs. 0.1 ~lm’. In dl @her ~, the ~ TWA for carbon black k 3.5 29. FffmYtlsgks9-The OsHA Stan&d for fibrous gbiofietedin the,umrnuy of MOSH mg/mJ and Me Ie the value Iietvd in the compdum tik. mmmmmdations, but not in the OSHA ~. In the OSHA ~, fibrow 19. hldhy&Iw ch/odde-Under the NIOSH rvcomrnertdation for methyfene chlorfde the d- i8 daldfkd u “n~ dwt” A ~ by w~abh fraction. NIOSH @UIng (m ppm) h to h Iouwed fn the Preaenm of cubon monoxkfe. cbifks fibroue gfue as “totaf ffbrous glass.” ACGIH simply cksdfke it ae “ffbroue W Ketcmes-OSHA does not eet a smndard for Methyl kourtyl ketone. gkee du9t.” 21. Thiol and memptmt are 8yrtonyms. OSHA u8eJ the IUme merc+tan and he stand- m. sflica-Smndar& and mmmmedatioln for dlica am Compfvx; U9udly Utfuzing fWlrul- ards for ordy three of these compounds. NIOSH kbek them t!liOb utd recommend$ IU for difkrent pcrcatagee of sifica or quartz in mpirabk dust. The rnmsure is usu- u~ for 16 thktk. fn addition, NfOSH mak ● note that ‘“mixtures of thiok ally M/m’. In the comparison table, the “wont caee” $tandud or 100% eilica is Lued are to be controlled by Cdcukthll of qutvaknt cOrWmratioN.” The CompU’ilm a8 the Conpubon among the three ~tions. The NOSH recommendation for table Ay Contairu h TLVS for the k thAo18 tit - revered by afl th organi- dim is ● eingfe TLV. OSHA and ACGIH TLVS for dlica are calculated by ● fonmda, Mkru. and the9tar&rd chan@esaOmrd@ to the percent quutz in mspimbkdust. The 198344 22. Vinyl hdides-in the summary of MOSH Reco mmendBtkm Vtnyl haude9 are ACGIH recommendstioru no lorqIw requhv that a formuk be used to meuure ex- Ckedfkd in one Catagory; the note under the expo$ure limit 9tate9 ““m promu@ed IXWWV, ~~ ACGIH hM eh@fied the table to chow numerical TLVs. In theowA for Vtnyl Chforide In 29 CPR 1910.1017 with eventual Sod of zero expoowe.’” h the emnduds, dfica formuk for the MM* am covered in table z-3. h“nt NIOSH/OSHA Currvnt lrIteflQmce ~ulletin 26 d Sept. 21,1978, the vinyl hd- 31. Tungsten-Them ● re no OSHA standorde for tungstm. MOSH divides tungsten into ldeexpoeW8tMdud8 epecify vxpomwe limits for four vinyl ftdidm mononwre inad- four ca~ries. Both NIOSH and ACGIH hWC mcommmkthu for eolubk and in- dition to vinyl chloride: vinyl bromide, Vinylidene dtloride, Vinyudene fluoride, and eolubk tungsten A thae are included in the compuieon table. NIOSH also rvcom- vinyl fluodde. (Tfte9e Ue Usted in tk com@eon table.) The MCXH mommemk- menrb TLVS for %hwt of cemented hmgoten cubicle that b 2 perrsnt cobaft (0.1 mg/m~ tkn indudee the Su@dotl thu om@iond expomuv to vinyl brodde d vinyfidene TWA) and for “dust d canented ~ carbide thm of 0.3 percmt nkkd (ls#lg c#cgbrablr3uLWlX!dd010atom#e&*t uesupatcdcardWnm (nkkef)/ml TWA). ancn there are no ”etduds 32. Vandhmr-nw OSHA CtMdud8 w~ VmdUm hto Vtillm Pentoxide dust for k four vinyf M*, either dr@y or u 8 ~. Wwever, the summary d and fume (I/lo,), and fmwanadhm dust which is listed eepamtdy in theosHA table, M02H rvcomrnenddcnu Iiets the OSHA 9mndard for vinyl hdidee u ● group ae N102H mommede the eune standard for vanadium compounde (without further 1 ppm TWA and S ppm dfIns p~iVQ kvef. Theee are the prokcttve kvek uoed -dt ~ - v--- -urn carbide. -llw NfosH mmmmded in the compar&orl table. TLV for metallic vanadium is u9ed in the compubn tabk under ferrovarwdium. ACGIH doe, * bve ● recommended expouue fimit for vinyl halides ae ● group ACGIH, Uke OSHA, dawlfiee vanadtum or vanadium pentoxkk ae respirable duet but doee have mcommdatiotu for two member, of the group; vinyl brom* and and fume, and ferrovalwdium. VInyfidetw chfodde. ~ ACGIH ~ for Vinyl brodde Cu14e8anufded 33. &ad-The OSHA St4n&rd for fead fue ● WA TLV. In addition, there is ● formula Warning that it be Wpected ~ - for -. to detmnhe the ‘1’LV for ~ over 8 hour, long. NIOSH he Onfy a recom- 23. Cadmium-The OSHA @arIAI& eepsrste cadmium into dust 4 fumes, & fiw mended c+kng knit. ACGIH de,crtbee kuf exposuree ae hO~ fume and dust Wfemrlt PEbforeuh. ne8ummmry of Mo6H raunmMAtioem af90eepu8teIcad- and mmmmmds TWA and ceU@ TLVS. mium into duu and fuma. However, ~~ mcomnwn&tioIu to ACGIH 34. Z,24”bromo-3-c)doropropsrw-The OSHA etandd for DBCP can be found in sec- TLVS b mum &ffkuh. ACGIH lists “cubnium, - md ufts, - Cd” and “c&$nium tion 1910.1o44 of the OSHA etuKkd9. Eye contact and skin contact with DBCP are Oxide, fume, as cd” and “pdwtion.” Cadmium oxide fume he ● c4ins limit of prohibited and to be ● olded. fn addition, OSHA notes that it has k known to O.@ ~lma under ACGIH. TIM mmpukon table uees the protecdvv kvek for cad- cause stedlity in humuu and is ● potaltklcuuu risk. MOSH mcommed a C’dung mfum dust and safts for compmbn to both OSHA and NIOSH protective Ievefs. _vc til of O.1 W: $OptaXhUn of 30 tnhum. ACGIH dote not have 24. Mwcury-MOSH snd OSHA have only one daseifkation for mercury, which is “in- ● mommeddetandud -.” ACGIH dividee mercury into three cstagorim, euh with dlffemnt TL.VS; 3s. h theosHAstMduds the fouowtng”dwmkd $ are fhted indbting that ekin Cwltxt dkyl compodo, d forms excqt alkyl Vspor, Md al+ and inqanic compoW& b tO b WddNi: SCVhkk, ddfidtddlin, ChfOrOdi#Wlyt (both 42 PerCWIt and In the Cunpukon Me, the vxpoaum limit for Uyl and ine compounds i9 Wed ~ l=-), dtl-Pr-# CYddet Dm, Lldtmethylhydruine, dinitroatho-crewl, to canpuv with MOSH and OSHA protcck feds. di oxuw, qkhkwohydh, hydradne, malathion, ftwthyl hydrdne, Ntbgltr 25. Nkkel–’The OSHA standu& label nickel ae “metal and solubk compotmtk, u Ni.’” (both), Phd hydrubte, parathion, phenol, 1,1,2,2, -tetrddoroethme, d N1OSH recommendations &cribee nkkef 8e “inor’gmic and cQm- 36. The foflowing PW of _ can be found in the MOSH ~ of recommm- qvm 9epuue9 nickel into two ~, “trtetd” Md %Olubfe @mPOWUb# &tionc, but they are not included in the compdmm table: decompdtion prO&## ao Nf.” The latter Cbdfkation b d in the ~ tsbk of fluorocubn, Peeti4e rnanufmwfng d formulation, ultraviolet radiation, and 24. DDT-The MOSH ruummdation for DDT, as deuribd in the sumtnuy d rec- w-~ X ~ Y-m, ethykne thk$~, chry=ne, and benddene+ad ~tkm, ~ w MI-: “Loweet reliably detectable fevd; 0.S ~/ml IWA 4= by MOSH Vdb&td method. * contact to be Wddd.” 37. In 1W3 OHSA held hearins on ● propoeed rwtdon to Its ethykne oxide standard, 27. The foUowfrtg eubotartca am fabdkd “’epecid Mad twkw” by N1OSH: At thoae hearings MOSH mmmmmdd tit (XHA should eet ● i%L. knver than 0.1 */*l*f ~ *, chJ3Wne# DOT, ethyferte oxide, ethykrte ppm ae an O-hour lWA and S ppm ss ● cdffng limit for ● period of 10 minutes. In thioures, 4,4’-methykne-bb (240routllitw), d nfckel Cubonyl. June 1904, OSHA bud ● find etandd of 1 ppm or 2 mg/m$ for an &hour TWA, 28. Nftdes-llte eummuy of MOSH mammen4tione ttotcs, ● t the bottom of the Ilet but did not issue wmil~ Iimft, That rvvieed recommendation and rwked etamkrcf of nitrue momawmhtkmm, that “Whal preunt ae rnIxtutw or with other oouKeo of have not been Incorporated in the table or the comparbon. Appendix B. —Working Papers

For this assessment, OTA commissioned a number of reports on various topics concerning occupational health and safety. These contract reports, as well as two papers written by OTA staff, are being made available through the National Technical Information Service as Working Papers. The Working Papers have been grouped into three parts. Part A includes analyses of available data, historical discussions of several topics, descriptions and critiques of the techniques of economic analysis, and a discussion of nonregulatory incentives. Part B covers descriptions of the personal protective equipment industry, theories of injury causation and control, information exchange, and worker training and education, Part C consists of five Case Studies of different occupational hazards and diseases.

Working Papers #10—Springborn Management Services, Inc., Report on Workplace Protective Equipment and Cloth- Part A: ing, contract report, August 1982. #1—Kronebusch, K., Data on Occupationa/ Injuries #11—Purswell, J. L., and Stephens, R., Health and Safety Control Technologies in the Workplace: and Illnesses, working paper, December 1984. Accident Causation and Injury Control, contract #2—Mendeloff, J., An Analysis of OSHA Health In- report, July 1983. spection Data, contract report, April 1983. #3—Corn, J. C., Historical Perspective: The Evo- #12—Priest, W. C., Computer-Teleconferencing as a lution of the Definition of Certain Work-Related Mechanism to Improve Information Transferor Illnesses, contract report, July 1983. Workplace Safety and Health, contract report, May 1983. #4—Kent, M., A History of Occupational Safety and Health in the United States, contract report, #13—INFORM, Worker Training and Education Pro- grams in Occupational Health and Safety, con- April 1983. tract report, November 1983. #5—MacLean, D., A History of the Consideration of Economic Jmpacts in the Determination of the Feasibility of Standards Under the OSH Act, Part C: contract report, April 1983. #14—Arndt, R., and Chapman, L., Potential Office #6--Lave, J. R., and Lave, L. B., Decision Frame- Hazards and Controls, contract report, Septem- works to Enhance Occupational Health and Safe- ber 1984. ty Regulation, contract report, February 1983. #15—Bleecker, M., Carpal Tunnel Syndrome: A Case #7—MacLean, D., and Sagoff, M., A Critique of Study, contract report, May 1984. Cost-Benefit Analysis as a Technique for Deter- #16—Goble, R., Hattis, D., Ballew, M., and Thurs- mining Health Standards Under the Occupation- ton, D., Implementation of the Occupational al Safety and Health Act, contract report, April Lead Exposure Standard, contract report, No- 1983. vember 1983. #8—Harter, P. J,, Non-Regulatory Legal Incentives #17—Hickey, J. L. S., Rice, C. H., and Boehlecke, B. for the Adoption of Occupational Safety and A., Technologies for Controlling Worker Expo- Health Control Technologies, contract report, sure to Silica, contract report, August 1983. April 1983. #18–Ruttenberg, R., Compliance With the OSHA Cotton Dust Rule: The Role of Productivity Im- Part B: proving Technology, contract report, March #9—Gough, M., The Personal Protective Equipment 1983. and Clothing Industry, working paper, April 1984.

381 Appendix C.—Acknowledgments and Health Program Advisory Committee

OTA would like to thank the members of the advisory panel who commented on drafts of this report and the contractors who provided material for this assessment. In addition, OTA acknowledges the following individuals and groups for their assistance in reviewing drafts or furnishing information and materials:

Lori Abrams E. I. du Pont de Nemours & Co. Public Citizen Wilmington, DE Theme Auchter Joseph Durst Occupational Safety and Health Administration United Brotherhood of Carpenters and Joiners of America Peter Barth University of Connecticut Paul Dwyer House Committee on Education and Labor Cynthia Bascetta General Accounting Office James Foster Occupational Safety and Health Administration David Bell Occupational Safety and Health Administration Frank Frodyma Occupational Safety and Health Administration Alfred Blackman Potomac, MD Sara Garrison-Leo E. I. du Pent de Nemours & Co. Leslie Boden Harvard University Anthony Goldin Occupational Safety and Health Administration Mary Jane Belle Congressional Research Service Morris Gordy Liberty Mutual Insurance Co. Nancy Bollinger National Institute for Occupational Safety Bengt Gustafsson and Health Swedish Embassy Queta Bond Walter Haag National Research Council National Institute for Occupational Safety and Health Patricia Breslin Occupational Safety and Health Administration Daniel Ham Occupational Safety and Health Administration Thomas Brown Department of Labor Carmen Harleston Occupational Safety and Health Administration Sandra Christensen Congressional Budget Office Daniel Hoeschen Occupational Safety and Health Administration Patrick Coleman National Institute for Occupational Safety Thomas Holzman and Health Department of Labor, Office of the Soliciter Geraldine Cox John Inzana Chemical Manufacturers Association Bureau of Labor Statistics Thomas Crane Daniel Jacoby Office of Representative Barney Frank Department of Labor, Office of the Soliciter Raymond Donnelly William Johnson Occupational Safety and Health Administration Syracuse University Joseph DuBois John Katalinas Occupational Safety and Health Administration Occupational Safety and Health Administration App. C—Acknowledgments and Health Program Advisory Committee 383

Ruth Knight Randy Rabinowitz Occupational Safety and Health Administration Zwerdling, Schlossberg, Leibig, & Kahn Akio Konoshima Knut Ringen Occupational Safety and Health Administration National Cancer Institute David LeGrande Anthony Robbins Communication Workers of America House Committee on Energy and Commerce J. William Lloyd Frank Rosenthal Occupational Safety and Health Administration The Johns Hopkins University Jeremiah Lynch John Russell Exxon Chemical Co. Timonium, MD Mark MacCarthy Herbert Schaffer House Committee on Energy and Commerce Bureau of Labor Statistics Judson MacLaury Lyle Schauer Department of Labor, Historical Office Bureau of Labor Statistics L. M. Magner Scott Schneider E. I. du Pent de Nemours & Co. United Brotherhood of Carpenters and Joiners of America Andre Maisonpierre Alliance of American Insurers Jack Snyder Merck & Co. J. A. Martin United Technologies Kate Sommers National Research Council J. Donald Millar National Institute for Occupational Safety Dorothy Strunk and Health House Committee on Education and Labor Minnesota Mining & Manufacturing (3M) John Tritsch St. Paul, MN American Textile Manufacturers Institute Benjamin Mintz Ralph Vernon American University Texas A&M University Franklin Mirer Frank Wilcher United Automobile Workers Safety Equipment Institute National Institute for Occupational Safety Donald P. Wilmes and Health Minnesota Mining & Manufacturing (3M) Atlanta, GA, Cincinnati, OH, and Sidney Wolfe Morgantown, WV Public Citizen Occupational Safety and Health Administration Washington, DC John Page E. I. du Pent de Nemours & Co. 384 ● Preventing Illness and Injury In the Workplace

HEALTH PROGRAM ADVISORY COMMITTEE

Sidney S. Lee, Committee Chair President, Milbank Memorial Fund New York, NY Stuart H. Altman* Alexander Leaf Dean Professor of Medicine Florence Heller School Harvard Medical School Brandeis University Massachusetts General Hospital Waltham, MA Boston, MA H. David Banta Margaret Mahoney* Deputy Director President Pan American Health Organization The Commonwealth Fund Washington, DC New York, NY Carroll L. Estes** Frederick Mosteller Chair Professor and Chair Department of Social and Behavioral Sciences Department of Health Policy and Management School of Nursing School of Public Health University of California, San Francisco Harvard University San Francisco, CA Boston, MA Rashi Fein Norton Nelson Professor Professor Department of Social Medicine and Health Policy Department of Environmental Medicine Harvard Medical School New York University Medical School Boston, MA New York, NY Harvey V. Fineberg Robert Oseasohn Dean Associate Dean School of Public Health University of Texas, San Antonio Harvard University San Antonio, TX Boston, MA Nora Piore Melvin A. Glasser*** Senior Advisor Director The Commonwealth Fund Health Security Action Council New York, NY Committee for National Health Insurance Mitchell Rabkin** Washington, DC President Patricia King Beth Israel Hospital Professor Boston, MA Georgetown Law Center Dorothy P. Rice Washington, DC Regents Lecturer Joyce C. Lashof Department of Social and Behavioral Sciences Dean School of Nursing School of Public Health University of California, San Francisco University of California, Berkeley San Francisco, CA Berkeley, CA

● Until April 19S3. ● *Until March 19S4. “Until August 19S3. ● **Until October 19S3. ● *Until April 19S3. App. C—Acknowledgments and Health Program Advisory Committee 385

Richard K. Riegelman Frederick C. Robbins Associate Professor President George Washington University Institute of Medicine School of Medicine Washington, DC Washington, DC Rosemary Stevens Walter L. Robb Professor Vice President and General Manager Department of History and Sociology of Science Medical Systems Operations University of Pennsylvania General Electric Co. Philadelphia, PA Milwaukee, WI Appendix D. —Glossary of Acronyms and Terms

Glossary of Acronyms NIEHS —National Institute of Environmental Health Sciences AAIH —American Academy of Industrial NIH –National Institutes of Health (DHHS) Hygiene NIOSH —National Institute for Occupational ABET —American Board of Engineering and Safety and Health (CDC) Technology NOES —National Occupational Exposure ABIH —American Board of Industrial Hygiene Survey (NIOSH) ACGIH —American Conference of Governmental NOHS —National Occupational Hazard Survey Industrial Hygienists (NIOSH) AEPIC —Architectural and Engineering NRC –National Research Council (NAS) Performance Impact Center NSC —National Safety Council AIHA —American Industrial Hygiene NSMS —National Safety Management Society Association OMB —Office of Management and Budget ANSI —American National Standards Institute OSHA –Occupational Safety and Health AOMA —American Occupational Medical Administration (DOL) Association OSH Act —Occupational Safety and Health Act ASSE —American Society of Safety Engineers (Public Law 91-596) ASTM —American Society for Testing and OTA —Office of Technology Assessment Materials (U.S. Congress) ATMI —American Textile Manufacturers PEL –Permissible Exposure Limit (OSHA) Institute PHS –Public Health Service (DHHS) AUPOHS —American University Programs for SEER –Surveillance, Epidemiology, and End Occupational Safety and-Health Results program (NCI) BLS –Bureau of Labor Statistics (DOL) SIC —Standard Industrial Classification CBO –Congressional Budget Office (U.S. TLV –Threshold Limit Value (ACGIH) Congress) CDC –Centers for Disease Control (PHS) CFR —Code of Federal Regulations COSH –Committee for Occupational Safety Glossary of Terms and Health CRS –Congressional Research Service (U.S. Acute: Used to describe a disease or injury that is mani- Congress) fest soon after exposure to a hazard. CSHO —Compliance Safety and Health Officer Add-on controls: Measures for injury and illness pre- (OSHA) vention which are put into place after the workplace DHEW –U.S. Department of Health, is built and equipment installed [see chs. 5, 6, and Education, and Welfare 16]. DHHS –U.S. Department of Health and Administrative controls: Methods of reducing worker Human Services exposures to occupational hazards through admin- DOL —U.S. Department of Labor istrative arrangements. For example, rotating a EPA —U.S. Environmental Protection Agency worker from areas of high exposure to areas of low ERC —Educational Resource Center exposure reduces that worker’s average exposure GAO –General Accounting Office (U.S. level. Also includes scheduling of jobs or processes Congress) that generate hazards at times when few workers MSHA –Mine Safety and Health are present. See engineering controls, personal pro- Administration (DOL) tective equipment, and work practice controls [see NAS –National Academy of Sciences also chs. 5, 6, and 9]. NCHS —National Center for Health Statistics Asbestosis: A restrictive chronic disease of the respi- (DHHS) ratory system resulting from exposure to asbestos NCI —National Cancer Institute (NIH) dust. NFPA —National Fire Protection Association Asthma: Constriction of the bronchial tubes, in the NHIS —National Health Interview Survey upper regions of the lung, in response to irritation, (NCHS) allergy, or other stimuli. App. D—Glossary of Acronyms and Terms ● 387

Bioassay: The use of animals to test chemicals or phys- necessary for standards. Upon completion these ical agents for harmful effects. documents are transmitted to the Occupational Bronchitis: Inflammation of the bronchial tubes in the Safety and Health Administration as recommended upper respiratory system. standards [see ch. 12]. Byssinosis: An obstructive chronic disease of the res- Dermatitis: Inflammation or irritation of the skin. piratory system resulting from exposure to cotton Dibromochloropropane (DBCP): A chemical used as dust [see ch. 5]. a pesticide. In the mid-1970s, a group of male work- Cancer: The unrestrained growth of tissue. ers discovered that their exposure to DBCP had ren- Carcinogen: A substance or physical agent that causes dered them sterile. An OSHA regulation limiting cancer. exposures to DBCP was issued in 1978. Carpal Tunnel Syndrome: An affliction among work- Dose: The amount of energy or substance absorbed ers doing hand work caused by compression of the in a unit volume or an organ or individual. Dose median nerve in the carpal tunnel, the passage in rate is the dose delivered per unit of time. the wrist through which blood vessels and nerves Educational Resource Centers (ERC): Fifteen academ- pass to the hand from the forearm [see ch. 7]. ic centers established by the National Institute for Chronic: Used to describe a disease or injury that is Occupational Safety and Health to provide multi- manifest long after exposure to a hazard. Also used disciplinary training for industrial hygienists, safety to describe persistent disease. specialists, occupational health nurses, and physi- Collective bargaining: Negotiation between employers cians. These centers also provide continuing edu- and unions concerning wages, hours, and working cation and technical assistance for their regions. conditions [see ch. 15]. Emergency Temporary Standard (ETS): A standard is- Complaint inspections: OSHA inspections that re- sued under section 6(c) of the Occupational Safety spond to worker or union complaints about work- and Health Act. Such a standard may be issued place hazards. See fatality/catastrophe, follow-up, when OSHA determines that workers are exposed and programmed inspections [see also ch. 12]. to a “grave danger” from an occupational hazard Compliance Safety and Health Officer (CSHO): The and that an emergency standard is necessary to pro- formal title for OSHA inspectors. tect them from that danger. Consultation: As used in this assessment, an OSHA- Emphysema: A condition of the lungs characterized funded program that provides employers with a by an increase beyond the normal in the size of air confidential evaluation of the health and safety spaces in the furtherest reaches of the lung either hazards in their workplaces and recommendations from dialation of the alveolii (the tiny sacs in the concerning hazard abatement [see ch. 12]. the lung where oxygen from the air and waste car- Contaminant: An undesirable chemical constituent of bon dioxide in the blood are exchanged) or from a system or environment. the destruction of their walls. Control Technology Assessments: The name given to Engineering controls: Methods of controlling worker a series of National Institute for Occupational Safe- exposure by modifying the source or reducing the ty and Health reports assessing measures for pre- amount of contaminants released into the work- venting work-related injury and illness. place. Engineering controls include process design Cost-benefit analysis: An analytical technique that and modification, equipment design, enclosure and compares the costs of a project or technological ap- isolation, and ventilation. See administrative con- plication to the resultant benefits, with both costs trols, personal protective equipment, and work and benefits expressed by the same measure. This practice controls [see also chs. 5, 6, and 9]. measure is nearly always monetary [see ch. 14]. Epidemiology: The study of the distribution of diseases Cost-effectiveness analysis: An analytical technique and their precursors in human populations. that compares the costs of a project or of alterna- Ergonomics: The study of how humans and machines tive projects to the resultant benefits, with costs and interact. In the occupational setting, one goal of er- benefits/effectiveness expressed by different meas- gonomics is to design the workplace to match work- ures. Costs are usually expressed in dollars, but ben- er capabilities. efits/effectiveness are ordinarily expressed in terms Ethylene dibromide (EDB): A chemical used as a such as “lives saved,“ “disability avoided,” “quali- fumigant and as a gasoline additive. It causes can- ty-adjusted life years saved, ” or any other relevant cer in mice. objectives [see ch. 14]. Experience rating: A system for setting worker com- Criteria Document: A series of National Institute for pensation insurance premiums that is based on the Occupational Safety and Health reports which as- employer’s record or experience concerning injuries sess available literature to develop the background and illnesses. 388 ● Preventing Illness and Injury In the Workplace

Exposure: The length of time and dose of chemical or Medical Removal Protection (MRP): A program speci- physical agent to which a worker is subjected. fied by the OSHA lead standard. It requiresremov- Fatality/catastrophe inspection: An OSHA inspection al of workers from lead-contaminated environments to investigate occupational fatalities or incidents when their blood lead levels exceed specified levels that result in the hospitalization of five or more em- [see ch. 5]. ployees. See complaint, follow-up, and pro- Medical treatment case: As defined by OSHA and grammed inspections [see also ch. 12]. BLS, a work-related injury or illness that requires Follow-up inspection: An OSHA inspection conducted medical treatment beyond first aid [see ch. 2]. to verify employer abatement of a violation uncov- Merit rating: See experience rating. ered in a previous OSHA inspection. See complaint, Mesothelioma: A malignant tumor of the membrane fatality/catastrophe, and programed inspections that lines the internal organs of the body. [see also ch. 12]. Monomer: A chemical substance that can undergo pol- General Duty Clause: Section 5(a)(1) of the Occupa- ymerization. See polymer. tional Safety and Health Act. This section provides Mutagen: A substance that causes mutations—changes that “each employer shall furnish . . . employment in the genetic material of. cells. and a place of employment which are free from rec- New Directions Program: An OSHA program to pro- ognized hazards that are causing or are likely to vide grants to employee, employer, educations, and cause death or serious physical harm to his employ- nonprofit organizations for the purpose of provid- ees.” OSHA has used this clause for workplace con- ing workplace health and safety training, education- ditions that present serious occupational hazards al materials, and services. that are not covered by OSHA’s more detailed Permissible Exposure Limit (PEL): The maximum air- health and safety standards. borne concentration of a toxic substance permitted Health and Safety Committees: Groups made up of by OSHA standards [see ch. 13]. both management and labor within a plant that Personal protective equipment: Equipment and cloth- meet to discuss and take mutual action to resolve ing designed to control hazards. It includes hard health and safety problems. hats, safety shoes, protective eyewear, protective Health Hazard Evaluation (HHE): A hazard identifi- clothing and gloves, hearing protectors, and vari- cation service provided by NIOSH. After receiv- ous types of respirators, such as dust and gas masks. ing a request from employees or an employer, a See administrative, engineering, and work practice team of NIOSH researchers evaluate a workplace controls [see also ch. 8]. to determine the toxicity of substances or processes. Polymer: A chemical substance formed by the join- Hierarchy of controls: The preference for using engi- ing together of many simple molecules or mono- neering controls to reduce or eliminate hazards. mers. For instance, many vinyl chloride monomers This preference, long a tenet of professional health are chemically joined to form polyvinyl chloride and safety practice, has been followed by OSHA. (PVC). See monomer. For example, to reduce exposures to air contami- Positive pressure mask: Respirators in which air pres- nants, OSHA requires that employers use engineer- sure inside the facepiece exceeds the outside air ing controls except when those controls are not fea- pressure. sible, not capable of reducing exposures to the Programed inspection: Programed or general schedule required levels, or while they are being designed and inspections are those OSHA inspections that are installed. scheduled using the injury experience or compliance Labeling standard: OSHA’s Hazard Communication history of an industry. See also complaint, fatal- standard which requires that certain information be ity/catastrophe, and follow-up inspections [see provided to workers about the identity of work- ch. 12]. place chemicals and their hazards [see ch. 15]. Protection factor: The ratio of measured concentra- Loss-control service: Service provided by insurers to tions of an airborne contaminant inside and out- client companies. Loss-control specialists visit work- side the facepiece of a respirator, A measure of the sites and offer advice on the prevention of proper- effectiveness of the respirator. ty loss and work-related injuries and illnesses [see “Records review” inspections: After arriving at a work- ch. 15]. place, an OSHA inspector examines the employer’s Lost-workday case: As defined by OSHA and BLS, injury and employment records. The inspector cal- a work-related injury or illness that results in an em- culates the lost-workday injury rate for the employ- ployee missing time from work or that restricts the er. If that rate is below the national average for employee’s work activity [see ch. 2]. manufacturing, the inspection will usually be ter- App. D—Glossary of Acronyms and Terms ● 389 —

minated. For this assessment, these inspections have the American Conference of Governmental Indus- been termed “records review” inspections. Current- trial Hygienist [see ch. 13]. ly, this policy applies only to programed safety in- Tort liability: A legal basis for compensation when spections [see ch. 12]. property has been damaged or a person has been Repetitive motion disorders: Diseases caused by injured. For occupational injuries and illnesses, most repetitive movement of part of the body. See Car- lawsuits between employees and employers are pal Tunnel Syndrome. barred by State workers’ compensation laws, al- Retro-fit controls: See add-on controls, though suits against “third parties, ” such as manu- Right-to-know laws: State and local laws requiring facturers of machinery or producers of asbestos companies to identify the chemical names and haz- products, are generally permitted [see ch. 15]. ards of their products to workers and the com- Toxicology: As used in this report, the testing of sub- munit y. stances for toxic effects in animals. Rulemaking: The administrative process by which Vinyl chloride monomer (VCM): The basic building OSHA and other regulatory agencies set standards. block chemical for polyvinyl chloride plastic. See Silicosis: A restrictive chronic disease of the respira- monomer. tory system resulting from exposure to airborne sil- Voluntary Protection Programs: OSHA programs de- ica dust, signed to recognize the achievements of employers Standards issued after rulemaking: Health and safety and to provide additional opportunities for OSHA- standards issued by OSHA using the procedures es- employer consultation and cooperation. The three tablished by section 6(b) of the Occupational Safety programs are called “Star,” “Try,” and “Praise.” and Health Act. These procedures require that Voluntary standards: Protective limits developed by OSHA provide notice of intended changes and an companies, trade associations, and professional or- opportunity for public comment. ganizations, but which do not have the force of law. Startup standards: The initial group of standards Walsh-Healey Public Contracts Act: This Federal law, adopted by OSHA under section 6(a) of the Occu- enacted in 1936, directed the Department of Labor pational Safety and Health Act. These consisted of to issue requirements for safe work by Federal Gov- established Federal standards and consensus ernment contractors and to “blacklist” contractors standards. who did not comply with these requirements. State Program: Under section 18 of the Occupational Work practice controls: Methods of controlling haz- Safety and Health Act, States may set and enforce ards that involve only changes in job procedures workplace health and safety standards. In this and housekeeping. See administrative controls, en- assessment, these programs are termed “State Pro- gineering controls, and personal protective equip- grams” [see ch. 12]. ment [see also chs. 5, 6, and 9]. Teratogen: A chemical or physical agent that causes Worker’s compensation: State-required insurance pro- physical defects in offspring. grams which pay for medical costs and replace a Threshold Limit Value (TLV): Maximum airborne con- portion of employees’ wages lost due to work-re- centrations of toxic substances set as guidelines by lated injury and illness. References References

1. AFL-CIO v. Brerzrum, s30 F.2d 109 (3d Cir., 14. American Public Health Association, “Compen- Dec. 31, 1975), sation for and Prevention of Occupational Dis- 2. Amchan, A. J., “Callous Disregard” for Em- ease (Position Paper), ” APHA Policy Statement ployee Safety: The Exclusivity of the Workers’ 8329, adopted Nov. 16, 1983, published in Compensation Remedy Against Employers,” La- American Journal of Public Health, 74(3):292- bor Law ]oumal, pp. 683-696, November 1983. 295, March 1984. 3. American Bar Association, Commission on Law 15. American Standards Association, American and the Economy, Federal Regulation: Roads to Standard Safety Code for the Protection of Heads, lle~orrn (Washington, DC: American Bar Asso- Eyes, and Respiratory Organs, Z2-1938, Decem- ciation, 1979). ber 1938. 4. American Board of Engineering and Technol- 16. American Standards Association, American ogy, Cn”ten”a for Accrediting Programs in Engi- Standard Safety Code for Fiead, Eye, and Res- neering in the United States (New York: Amer- piratory Protection, Z2.1-1959, November 1959. ican Board of Engineering and Technology, Oct. 17. American Textile Manufacturers Institute, Znc. 18, 1980). v. Donovan, 452 U.S. 490; (June 17, 1981). 5. American Conference of Governmental Indus- 18. Ames, B. N., “Dietary Carcinogens and Anticar- trial Hygienists, Threshold Limit Values of Air- cinogens,” Science 221:1256- 1264, 1983. bome Contaminants and Jntended Changes, 19. Ames, B, N,, letter, Science 224:668 et seq., adopted by ACGIH for 1970 (Cincinnati, OH: 1984. ACGIH, 1970), 20. Anderson, L,, “Management of a Local Exhaust 6. American Conference of Governmental Indus- Ventilation Program, ” (slide presentation), Cer- trial Hygienists, Industrial Ventilation; A Man- tified Industrial Hygienist, IBM, Inc., Lexington, ual o) Recommended Practice, 18th ed, (Cin- KY, personal communication, June 14, 1983. cinatti, OH: ACGIH, 1984). 21. Armstrong, T. J., “An Ergonomics Guide to 7. American Conference of Governmental Indus- Carpal Tunnel Syndrome,” typescript, no date. trial Hygienists, Threshold Limit Values for 22. Armstrong, T. J., Foulke, J. A., Joseph, B. J., Chemical Substances and Physical Agents in the et al., “Investigation of Cumulative Trauma Dis- Work Environment with Intended Changes, orders in a Poultry Processing Plant, ” Ameri- adopted by ACGIH for 1982 (Cincinnati, OH: can Industrial Hygiene Association Journal 43: ACGIH, 1982). 103-116, 1982. 8. American Conference of Governmental Indus- 23. Armstrong, T. J. and Langolf, G. D., “Ergonom- trial Hygienists (ACGIH), Threshold Limit ics and Occupational Safety and Health, ” in En- Values for Chemical Substances and Physical vironmental and Occupational Medicine, W. N. Agents in the Work Environment with Intended Rem, A. D. Renzetti, and J. S. Lett (eds.) (Bos- Changes for 1983-84 (Cincinnati, OH: ACGIH, ton: Little Brown, 1983). 1983). 23a, Arndt, A. K., “Skin Disorders, ” in Occupational 9 American Industrial Hygiene Association and Health, B. S. Levy and D. H. Wegman (eds. ) American Conference of Governmental Indus- (Boston: Little, Brown & Co., 1983). trial Hygienists, Respiratory Protective Devices 24. Arndt, R., “Occupational Injuries and Worker Manual (Ann Arbor, MI: Braun and Brumfield, Capabilities,“ in Symposium on Occupational 1963). Safety Research and Education: A dialogue be- 10. American Zron & Steel Znstitute v. OSHA, 577 tween two communities, DHHS Publication No. F.2d 825 (3d Cir., Mar. 28, 1978). (NIOSH) 82-103 (Cincinnati, OH: National Insti- 11. American National Standards Institute, Amer- tute for Occupational Safety and Health, 1982). ican National Standard Practices for Respiratory 25. Arndt, R. and Chapman, L., Potential Office Protection, Z88.2-1969, August 1969. Hazards and Controls (Working Paper # 14 of 12. American National Standards Institute, Amer- Preventing Illness and Injury in the Workplace), ican National Standard Practices for Respiratory contract report prepared for the Office of Tech- Protection, Z88.2-1980, May 1980. nology Assessment, U.S. Congress, Washing-

13( American Petroleum Znstitute v, OSHA, 581 ton, DC, September 1984. F.2d 493 (5th Cir., Oct. 5, 1978). 26. Arthur Andersen and Co., Cost of Government

393 394 Preventing Illness and Injury in the Workplace

Regulation Study for the Business Roundtable, Labor, Washington, DC, NTIS Report No. Executive Summary, March 1979. ASPER/PUR77/1914 /A, Apr. 3, 1978. 27. Arthur D. Little, Inc., “Assessment of the Multi- 37. Bacow, L. S., Bargaining for Job Safety and Region Targeting Alternative Project, ” contract Health (Cambridge, MA: MIT Press, 1980). report prepared for the Office of Policy Analy- 38, Badaracco, J. L., Jr., A Study of Adversarial and sis, Integration, and Evaluation, Occupational Cooperative Relationships between Business and Safety and Health Administration, U.S. Depart- Government in Four Countries, prepared for the ment of Labor, Dec. 30, 1981. Office of Technology, Strategy, and Evaluation, 28. Asher, R., “Radicalism and Reform: State In- U.S. Department of Commerce, Feb. 3, 1981. surance of Workmen’s Compensation in Minne- 39. Bailey, R. W., Human Performance Engineer- sota, 1910-1933, ” Labor History 14(1):19-41, ing (Englewood Cliffs, NJ: Prentice-Hall, 1982). winter 1973. 39a. Baker, E. J., Jr., “Necrologic and Behavioral 29. Ashford, N. A., et al., National Support for Disorders,” in Environmental and Occupational Science and Technology: An Examination of Medicine, W. N. Rem, A, D. Renzetti, J. S. Lee, Foreign Experience (Cambridge, MA: Massa- et al., (Boston: Little, Brown & Co., 1983). chusetts Institute of Technology, Center for Pol- 40. Baker, S., Samkoff, J., Fisher, R., and Van icy Alternatives, Aug. 18, 1975). Buren, C., “Fatal Occupational Injuries,” ]our- 30! Ashford, N. A., Crisis in the Workplace: Occu- nal of the Amen”can Medical Association 248 pational Disease and Injury (Cambridge, MA: (6):692-7, Aug. 13, 1982. MIT Press, 1976). 41. Baker, S., “Medical Data and Injuries, ” Amen--

314 Ashford, N. A., Heaton, G, R., Jr., and Priest, can Journal of Public Health 73(7):733-34, July W. C., “Environmental, Health, and Safety Reg- 1983. ulation and Technological Innovation,” in Tech- 42. Baram, M. S. and Miyares, J. R., The Legal nological Change for a Dynamic Economy, C. Framework for Determining Unreasonable Risk T. Hill and J. M, Utterback (eds, ) (New York: From Carcinogenic Chernica2s, report prepared Pergamon Press, 1979). for the Office of Technology Assessment, May 32. Ashford, N., Hill, C., Mendez, W., et al., Bene- 16, 1980. fits of Environmental, Health, and Safety Reg- 43. Baram, M. S., “The Use of Cost-Benefit Analy- ulation, study prepared by the Center for Pol- sis in Regulatory Decision- Making is Proving icy Alternatives, Massachusetts Institute of Harmful to Public Health,” in Management of Technology, for the U.S. Senate, Committee on Assessed Risk for Carcinogens, W. J. Nicholson Governmental Affairs, committee print, Mar. (cd.) (New York: The New York Academy of 25, 1980. Sciences, 1981), 33. Ashford, N. A., Hattis, D,, Heaton, G., et al., 44. Baram, M. S., “The Right to Know and the Evaluating Chemical Regulations: Trade-Off Duty to Disclose Hazard Information, ” Ameri- Analysis and Impact Assessment for Environ- can Journal of Public Health, 74(4):385-390, mental Decision-Making, contract report to the April 1984. Council on Environmental Quality and the Envi- 45 Bartel, A. P,, and Thomas, L. G., “OSHA En- ronmental Protection Agency by the Center for forcement, Industrial Compliance, and Work- Policy Alternatives, Massachusetts Institute of place Injuries,” typescript, May 1984. Technology, Cambridge, MA, July 1980. 46. Barth, P. S., with Hunt, H. A., Workers’ Com- 34. Auchter, T., Assistant Secretary of Labor for pensation and Work-Related Illnesses and Dis- Occupational Safety and Health, OSHA com- eases (Cambridge, MA: MIT Press, 1980). ments on draft OTA report, letters, Feb. 3 and 47. Bell, D., Office of Regulatory Analysis, OSHA, 23, 1984. persomd communication, 1982. 35. Ayalp, A., and Myroniuk, D., “Evaluation of 48. Berger, E. H., Letter to Mr. J. Thomas Hall, Oc- Occupational Exposure to Free Silica in Alberta cupational Safety and Health Administration, Foundries, ” Amen-can Industr-z”al Hygiene Asso- Division of Consumer Affairs, Mar. 4, 1982. ciation ]ournal 43(11):825-831, 1982. 49 Berger, E. H., “Using the NRR to Estimate the 36, Bacow, L. S., “Regulating Occupational Haz- Real-World Performance of Hearing Protec- ards Through Collective Bargaining, ” report tors,” Sound and Vibration, January 1983. prepared for the Assistant Secretary for Policy, 50 Berger, E. H., Kerivan, J. E,, and Mintz, F., Evaluation, and Research, U.S. Department of “Inter-Laboratory Variability in the Measure- References 395

ment of Hearing Protector Attenuation, ”Sound 61 Bloomfield, B., “Engineering,” in Zndusttial Hy- and Vibration Magazine, 1982. giene Highlights vol. I., L. V. Cralley, L. J. 51. Berger, J. L,, and Riskin, S. D., “Economic and Cralley, and G. D. Clayton (eds.) (Pittsburgh, Technological Feasibility in Regulating Toxic PA: International Hygiene Foundation ofAmeri- Substances under the Occupational Safety and ca, Inc., 1968).

Health Act,” EcoZogy Law Quarterly 7(2):285- 62. Bluestone, B., and Harrisonr B., The Deindus- 358, 1978. tridization of America (New York: Basic Books, 52. Berkowitz, M., and Burton, J., ‘The Income 1982). Maintenance Objective in Workers’ Compensa- 63. Boden, L. I., Coal Mine Accidents and Gover- tion, ” Industn”al and Labor Relations Review, nment Enforcement of Safety Regulations, un- 24(1):14-31, 1970, cited in Darling-Hammond, published Ph.D. dissertation, Massachusetts In- L. and Kniesner, T. The Law and Economics of stitute of Technology, 1977. Workers’ Compensation (Santa Monica, CA: 64. Boden, L., and Wegman, D., “Increasing OSHA’S Rand Corporation, The Institute for Civil Jus- Clout: Sixty Million New Inspectors, ” Working tice, 1980). Papers 6:43-49, May-June 1978, 53. Berkowitz, M. “Aspects of the Economics of 65, Boden, L. I., “Cost-Benefit Analysis: Caveat Workmen’s Compensation, ” report to the Na- Emptor,” American Journal of Pubfic Health tional Workshop on Rehabilitation and Work- 69(12):1210-1211, December 1979. men’s Compensation, National Institutes of Re- 66, Boden, L. I., Hall, J. A., Levenstein, C., Pun- habilitation and Health Services, Washington, nett, L., ‘The Impact of Health and Safety Com- DC, 197I, cited in National Commission on mittees: A Study Based on Survey, Interview, State Workmen’s Compensation Laws, Com- and Occupational Safety and Health Adminis- pendium on Workmen’s Compensation (Wash- tration Data, ” Journal of Occupation Medicine ington, DC: U.S. Government Printing Office, 26(11): 829-34, November 1984. 1973). 67, Bollinger, N., ‘Testing and Certification Branch 54. Berman, D. M., “Why Work Kills: A Brief His- Activities,” presented to Respiratory Subcom- tory of Occupational Safety and Health in the mittee of the Mine Health Advisory Commit- U.S.,” lntemational Jouma! of Health Services tee, Jan. 4 and 5, 1982. 7(1):63-87, 1977. 68< Bollinger, N., National Institute of Occupational 55. Berman, D. M., Death on the ]ob: Occupational Safety and Health, Morgantown, WV, personal Health and Safety Struggles in the United States communication, Mar. 17, 1983. (New York: Monthly Review Press, 1978). 69. Borel v. Fibreboard Paper Products Corp., 493 56. Bingham, E., Asst. Sec. of Labor for Occupa- F. 2d 1076 (5th Cir, 1973) cited in ZVon-Regula- tional Safety and Health, memorandum on tory Legal Incentives for the Adoption of Occu- “Procedures for Implementing Economic Impact pational Safety and Health Control Technol- Analysis, ” to the Secretary of Labor, Washing- ogies, P. Harter, contract report prepared for ton, DC, May 11, 1977. the Office of Technology Assessment, U.S. Con- 57. Bird, F. E., Jr., Safety in The industrial gress, Washington, DC, April 1983. Environment–its Evaluation & Control (Wash- 70. Boucher, W., et al., Federal Incentives for in- ington, DC: National Institute for Occupational novation (Denver, CO: University of Denver, Safety and Health, 1973). Denver Research Institute, January 1976) cited 58. Bird, F., Management Guide to Loss Control in Ashford, N. A., Heaton, G. R., Jr., and (Atlanta, GA: Institute Press, 1974). Priest, W. C., “Environmental, Health, and 59. Blair, A., Abrasive Blasting Respiratory Protec- Safety Regulation and Technological Innova- tive Practices Survey, Report No. NIOSH-TR- tion, ” in Technologr”cal Change for a Dynamic 048-73 (Cincinnati, OH: National Institute for Economy, C. T. Hill and J. M. Utterback (eds.) Occupational Safety and Health, August 1973). (New York: Pergamon Press, 1979). 60. Bleecker, M., Carpal Tunnel Syndrome: A Case 71. Brandt, A. D., Industrial Health Erqp”neering Study (Working Paper #ls of Preventing Illness (New York: John Wiley & Sons, 1947). and Injury in the Workplace), contract report 72. Breyer, S., Regulation and Its Reform (Cam- prepared for the Office of Technology Assess- bridge, MA: Harvard University Press, 1982). ment, U.S. Congress, Washington, DC, May 73. Brief, R., Corn M., Firenze R., et al., Evalua- 1984. tion of the NZOSH Certification Program Divi- 396 ● Preventing Illness and Injury in the Workplace

sion of the Safety Research Testing and Cer- Division of Labor Statistics and Research, Cali- tification Branch, DHEW (NIOSH) Publication fornia Sawmills and Planing Mills Industry: No. 80-113 (Washington, DC: National Institute Analysis of Work Injuries and Illnesses, Re- for Occupational Safety and Health, November search Bulletin # 3, December 1978. 1979). 87. California Department of Industrial Relations, 74. Brief, R., Lipton, S., Amamani, S., et. al., ’’De- Division of Labor Statistics and Research, Cal- velopment of Exposure Control Strategy for ifomia Fabn”cated Structural Metal Products Xn- Process Equipment,” Anna/s of the American dustry: Analysis of Work Injuries and Illnesses, Conference of Governmental Industrial Hygien- Research Bulletin #4, December 1978. ists, vol. 5 (Cincinnati, OH: ACGIH, 1983). 88. California Department of Industrial Relations, 75. Brown, M. S., “Setting Occupational Health Division of Labor Statistics and Research, Ca2- Standards: The Vinyl Chloride Case,” in Con- ifomia Meat Products Industry: Analysis of troversy: Politics of Technical Decisions, D. Work Injuries and Illnesses, Research Bulletin Nelkin (cd. ) (Zd cd.) (Beverly Hills, CA: Sage #s, December 1979. Publications, 1984). 89. California Department of Industrial Relations, 76. Brownstein, R., “Asbestos Litigation a Legal Division of Labor Statistics and Research, Cal- Nightmare that Congress is Being Asked to ifornia Rooji”ng and Sheet Metal Work: Analy- End,” National Journal, pp. 1942-47, Sept. 24, sis of Work lnjt.m”es and Zlbzesses, Research 1983. Bulletin #6, October 1982. 77. Bureau of National Affairs, Basic Patterws in 90< Campbell, D. L., Zndu#rial Face Shield Perfonn- Union Contracts (Washington, DC: Bureau of unce Test (Morgantown, WV: National Institute National Affairs, various years). for Occupational Safety and Health, 1976). 78. Bureau of National Affairs, Safety Policies and 91. Campbell, D, L., Report of Tests of Welding Fil- the Impact of OSHA (Washington, DC: Bureau ter Plates (Morgantown, WV: National Institute of National Affairs, Inc., 1977). for Occupational Safety and Health, 1976). 79. Bureau of National Affairs, The lob Safety and 92. Campbell, D. L., and Collins, R. L., Tests of Health Act of 2970 (Washington, DC: Bureau Flexible Fitting Safety Goggles (Morgantown, of National Affairs, 1971). WV: National Institute for Occupational Safety 80. Burgess, W. A., Recognition of Health Hazards and Health, 1976). in Industry: A Review of Materials and Process 93. Campbell, D. L., Collins, R. L., and Wolfe, R. (New York: John Wiley and Sons, Inc., 1981), S., Tests of Eyecup Goggles (Morgantown, WV: 81. Burton, J. F., “Compensation for Permanent National Institute for Occupational Safety and Partial Disabilities, ” in Safety and the Work Health, 1977). Force, J. D. Worrall (cd.) (Ithaca, NY: ILR Press, 94. Campbell, D. L., and Collins, R. L., Tests of 1983). Glass Piano Safety Spectacles (Morgantown, 82 Business Roundtable, Improving Construction WV: National Institute for Occupational Safety Safety Performance, Report A-3 of the Con- and Health, 1977). struction Industry Cost Effectiveness Project 95. Cannon, L. J., Bemacki, E. J., and Walter, (New York, NY: The Business Roundtable, Jan- S. D., “Personal and Occupational Factors As- uary 1982). sociated With Carpal Tunnel Syndrome,” Jour- 83a Butler, R. J., “Wage and Injury Rate Response nal of Occupational Medicine 23:255-258, 1981. to Shifting Levels of Workers’ Compensation ‘“ 96. Caplan, K., “Philosophy and Management of in Safety and the Work Force, J. D. Worrall, Engineering Controls,” in Patty’s Industrial ffy- (cd.) (Ithaca, NY: ILR Press, 1983). gz”ene and Toxicology, vol. III, L. J. Cralley and 84. California Department of Industrial Relations, L. V. Cralley (eds.) (New York: John Wiley and Division of Labor Statistics and Research, Iron Sons, Inc., 1979). and Steel Foundn”es, Research Bulletin # 1, June 97. CapIan, K., letter to OTA, Mar. 15, 1983. 1977. 98. Carter, J., “Improving Government Regula- 85. California Department of Industrial Relations, tions,” Executive Order No. 12044, Federal Reg- Division of Labor Statistics and Research, Ca/i- ister 43:12661, Mar. 23, 1978. fomia Logging Industry: Analysis of Work In- 99. Carter, J., “Improving Government Regula- juries and illnesses, Research Bulletin # 2, No- tions, ” Executive Order No. 12221, Federal Reg- vember 1978. ister 45:44249, July 1, 1980, 86, California Department of Industrial Relations, 100, Castrovinci, J. L,, “Prelude to Welfare Capital- References ● 397

ism: The Role of Business in the Enactment of 111, Chemical Week, “Worker Protection Will Be- Workmen’s Compensation Legislation in Illi- come More Personal, ” pp. 31-32, Apr. 22, 1981. nois, 1905 -12,” in Compassion and Responsibil- 112, Clifton, R, “Asbestos,” Bureau of Mines Miner- ity, F. R. Breul and S. J. Diner (eds. ) (Chicago: als Handbook (Washington, DC: U.S. Depart- University of Chicago Press, 1980). ment of the Interior, 1977 and 1982). 101 Center for Policy Alternatives, Environmental/ 113, Collins, R. L., and Wolfe, R, S., Tests of Pkzstic Safety Regulation and Technological Change in Piano Safety Spectacles (Morgantown, W: Na- the U.S. Chemical Industry, cited in Ashford, tional Institute for Occupational Safety and N. A., Heaton, G. R., Jr., and Priest, W. C., Health, 1977). “Environmental, Health, and Safety Regulation 114. Connerton, M., Accident Control Through Reg- and Technological Innovation, ” in Technologi- ulation: The 1969 Coal Mine Health and Safety cal Change for a ~ynamic Economy, C. T. Hill Act Experience, unpublished Ph.D. dissertation, and J. M, Utterback (eds. ) (New York: Perga- Harvard University, June 1978. mon Press, 1979). 115. Connerton, M., and MacCarthy, M., Cost- 102 Chaff in, D. B., Director for the Center for Oc- Benefit Analysis and Regulation: Expressway to cupation Health and Safety Engineering, “CDC- Reform or Blind Alley? (Washington, DC: Na- NIOSH University Grants,” testimony presented tional Policy Exchange, 1982). before the Subcommittee on Labor, Health and 116. Cook, W. I., and Groce, D. W., Report on Tests Human Services, Education and Related Agen- of Class B Industrial Helmets (Morgantown, cies, Senate Committee on Appropriations, U.S. WV: National Institute for Occupational Safety Congress, Washington, DC, Apr. 27, 1983. and Health, 1975). 103 Charles River Associates, Economic and Envi- 117. Cook, W. I, A Report on the Performance of ronmental Analysis of the Current OSHA Lead Firefighters’ Helmets (Morgantown, WV: Na- Standard, contract report prepared for the Oc- tional Institute for Occupational Safety and cupational Safety and Health Administration, Health, 1976). U.S. Department of Labor, CRA Project No. 118, Cook, W. 1., A Report on the Performance of 5356.60, NTIS Report No. OSHA/IS-83/001, Mens Safety-Toe Footwear (Morgantown, WV: September 1982. National Institute for Occupational Safety and 104, Chelius, J. R., “An Empirical Analysis of Safet y Health, 1976). Regulation, ” in Supplemental Studies, National 119. Cook, W. I., and Fletcher, W. G., A Report on Commission on State Workmen’s Compensa- the Performance of Linemen’s Rubber Insulating tion Laws (Washington, DC: U.S. Government Gloves (Morgantown, WV: National Institute Printing Office, 1973). for Occupational Safety and Health, 1977). 105. Chelius, J. R., “Liability for Industrial Ac- 120. Cook, W. I., and Love, J. R., A Report on the cidents: A Comparison of Negligence and Strict Performance of Miners’ Safety Caps (Morgan- Liability Systems, ” Journal of Legal Studies town, WV: National Institute for Occupational 5:293-309, June 1976. Safety and Health, 1977). 106. Chelius, J. R., Workplace Safety and Health: 121. Cooke, W., and Gautschi, F., “OSHA, Plant The Role of Workers’ Compensation (Washing- Safety Programs, and Injury Reduction, ” lndus- ton, DC: American Enterprise Institute, 1977). tria] Relations 20(3):245-57, fall 1981. 107. Chelius, J., ‘The Influence of Workers’ Compen- 122. Corn, J. K., and Corn, M., “The Myth and the sation on Safety Incentives, ” Industrial and La- Reality, “ in Economic Effects of Government- bor Relations Review 35(2):235-42, January Mandated Costs, R. F. Lanzillotti (cd. ) 1982. (Gainesville, Florida: University Presses of 108. Chelius, J. R., “The Incentive to Prevent In- Florida, 1977). juries, ” in Safety and the Work Force, J. D. 123 Corn, J. K,, “Historical Aspects of Industrial Worrall (cd. ) (Ithaca, NY: ILR Press, 1983). Hygiene: Changing attitudes toward occupa- 109. Chelius, J, R. and Smith, R. S., “Experience Rat- tional health, ” Amen”can Industrial Hygiene ing and Injury Prevention, ” in Safety and the Association Journal 39(9):695-699, 1978. Work Force, J. D, Worrall (cd. ) (Ithaca, NY: ILR 124. Corn, J. K., Historical Perspective; The Evolu- Press, 1983). tion of the Definition of Certain Work-Related 110. Chemical and En~”neering News, “OSHA Memo Illnesses, (Working Paper #3 of Preventing 111- to Delay Rule Making Creates Stir, ” 61(39):7, ness and Injury in the Workplace), contract re- Sept. 26, 1983. Dort DreDared for the Office of Technolozv As- 398 . preventing Illness and Injury in the Workplace

sessment, U.S. Congress, Washington, DC, July 138. Dickens, W. T., “Differences Between Risk Pre- 1983. miums in Union and Nonunion Wages and the 125. Corn, M., for the Board of Directors of the Case for Occupational Safety Regulation, ” American Conference of Governmental Indus- American Economic Review 74(2):320323, May trial Hygienists, comments submitted to OSHA 1984. Docket No. 160, Exhibit 2-32. 139. Diebold, J., “Manage the System More Effi- 126. Corn, M., personal communication, letter, Jan- ciently, ” interview with T. C. Hayes, New York uary 1984. Times, p. F-n, Sept. 2, 1984. 127. Cornell, N., Nell, R., & Weingast, B., “Safety 140. Dinman, B., testimony before Subcommittee on Regulation, ” in Setting National Pn”ori”ties: The Administrative Law and Government Relations, Next Ten Years, H. Owen and C. L. Schultze House Committee on the Judiciary, U.S. Con- (eds. ) (Washington, DC: Brookings Institution, gress, Regulatory Reform Act, Hearings, June- 1976). July 1983. 128. Cothrin, S. G., ASARCO, Inc., Department of 141. DiPietro, A., “An Analysis of the OSHA Inspec- Environmental Sciences, Salt Lake City, UT., tion Program in Manufacturing Industries, 1972- personal communication, April 1983. 73, ” Technical Analysis Paper, prepared for the 128a. Craft, B. F., “Occupational and Environmental Asst. Sec. for Policy, Evaluation, and Research, Health Standards,” in Occupational Health, B. U.S. Department of Labor, August 1976. S. Levy and D. H. Wegman (eds. ) (Boston: Lit- 142. Dirks-Mason, S., and Ruttenberg, R., “The Ef- tle, Brown & Co., 1983). fects of the OSHA Vinyl Chloride Standard on 129. Cralley, L. J., and Cralley, L. V. (eds. ) Patty’s the Vinyl Chloride Industry,” typescript, August Industrial Hygiene and Toxicology, vol. III 1979. (New York: John Wiley and Sons, Inc., 1979). 143. Discher, D., Kleinman, G., and Foster, F. J., 130. Croyle, J. L., “Industrial Accident Liability Pol- Pilot Study for Development of an Occupational icy of the Early Twentieth Century, ” The Jour- Disease Surveillance Method, NIOSH Publica- nal of Legal Studies 7(2):279-297, June 1978. tion No. 75-162, May 1975. 131. Darling-Hammond, L., and Kniesner, T. J., The 144. Dittoe, J., ‘The Treatment of Sexual Impairment Law and Economics of Workers’ Compensation Injuries under Workers’ Compensation Laws, ” (Santa Monica, CA: Rand Corporation, The In- Hastings Law Journal, 30:1207-36, March 1979. stitute for Civil Justice, 1980). 145. Dixon, S. W., and Nelson, T. J., E.I. DuPont 132. Davis, D. L., Bridboard, K., and Schneiderrnan, de Nemours & Co., Inc., Procedures for Deter- M., “Estimating Cancer Causes: Problems in mining Workplace Performance of Respirators, Methodology, Production and Trends, ” in Parts 1 and U, presented at the American Indus- Quantification of Occupational Cancer: Ban- trial Hygiene Conference, Philadelphia, PA, Lwry Report 9, R. Peto and M. Schneiderman May 23, 1983. (eds, ) (Cold Spring Harbor, NY: Cold Spring 146. Dixon, S. W., and Nelson, T. J., E.I. du Pent Harbor Laboratory, 1981). de Nemours & Company, personal communica- 133. Davis, J. L., “Regulatory Reform and Congres- tion, letter, Jan. 5, 1984. sional Control of Regulation, ” New England 147. Doll, R., “Mortality From Lung Cancer in As- Law Review 17(4):1199-1235, 1982. bestos Workers,” British ]oumal of Industrial 134. Davis, J. W., “The Use of Sand Substitution to Medicine 12:81, 1955. Solve the Free Silica Problem in Foundry Atmos- 148. Doll, R,, and Pete, R., “The Causes of Cancer: pheres, ” Amen”can Industrial Hygiene Associa- Quantitative Estimates of Avoidable Risks of tion Journal 40(7):609-618, 1979. Cancer in the United States Today,” Journal of 135. DeMuth, C. C., “Constraining Regulatory the National Cancer Institute, 66:1191-1308, Costs—Part I: The White House Review Pro- 1981. grams, ” Regulation, 4(1):13-26, January- 149. Doniger, D. D., The Law and Policy of Toxic February 1980. Substances Control (Washington, DC: Re- 136. DeMuth, C. C., “A Strategy for Regulatory sources for the Future, 1978). Reform,” Regulation 8(2):25- 30, March-April 150. Dorsey, S., “Employment Hazards and Fringe 1984. Benefits: Further Tests for Compensating Differ- 137. Denisen, E. F., Accounting for Slower Economic entials, ” in Safety and the Work Force, J. D. Growth (Washington, DC: The Brookings In- Worrall (cd.) (Ithaca, NY: ILR Press, 1983). stitution, 1979). 151. Dorsey, S., and Walzer, N., “Workers’ Com- References ● 399

pensation, Job Hazards, and Wages, ” industrial 166. Etzioni, A., “Social Progress vs. Economic Prog- and Labor Relations Review 36(4):642-654, July, ress, ” Social Po2icy, pp. 4-9, March-April 1980. 1983. 167. Etzioni, A., An Immodest Agenda: Rebuilding 152. Douglas, D. D., Hesch, P. R., and Lowry, P. America Before the Twenty-First Century (New L., Supplied-Air Hood Report, Informal Report, York: McGraw Hill, 1983). LA-NUREG-6612-MS (Los Alamos, NM: Los 168. Executive Office of the President, Economic Re- Alamos Scientific Laboratory, December 1976). port of the President (Washington, DC: U.S. 153. Dubos, R., The Mirage of Health (New York: Government Printing Office, 1983). Harper and Brothers Publishers, 1959). 169. Executive Office of the President, Economic Re- 155. E. I. du Pent de Nemours & Co., Applied Tech- port of the President (Washington, DC: U.S. nology Division, Fabrics and Finishes Depart- Government Printing Office, 1984). ment, Safety Training Obsewation Program, 2d 170. Executive Office of the President, Council of ed. (Wilmington, DE: E. I. du Pent de Nemours Economic Advisors, memorandum on “Eco- & Co., Inc., 1980). nomic Impact Analyses, ” to Economic Policy 156. E. I. du Pent de Nemours & Co., comments on Group, Washington, DC, Mar. 10, 1977. draft report, letter to OTA from J. G. Page, Jan. 171. Executive Office of the President, Office of Man- 25, 1984. agement and Budget, Information Collection 157. Eads, G., “Harnessing Regulation: The Evolv- Budget of the United States Government: Fiscal ing Role of White House Oversight, ” Regula- Year 1984 (Washington, DC: U.S. Government tion, pp. 19-26, May-June 1981. Printing Office, 1984). 158. Eads, G., and Fix, M., “Regulatory Policy, ” in 172. Fawcett, H. H., “Respiratory Hazards and Pro- The Reagan Experiment, J. L. Palmer and I. V. tection, ” in Safety and Accident Prevention in Sawhill (eds. ) (Washington, DC: Urban Insti- Chemical Operations, 2d cd., H. H. Fawcett and tute, 1982), W. S. Wood (eds. ) (New York: John Wiley and 159. Eastman Kodak Company, Health, Safety, and Sons, Inc., 1982). Human Factors Laboratory, Human Factors 173. First, M. W., “Engineering Control of Occupa- Section, Ergonomic Design for People at Work tional Health Hazards,” 1983 Cummings Memo- (Belmont, CA: Lifetime Learning Publications, rial Lecture, American industrial Hygiene Asso- 1983). ciation ]ouma! 44(9):621-626, 1983. 160. Economist, December 1980, quoted in Com- 174. Fischbein, A., “Envrionmental and Occupa- pliance with the OSHA Cotton Dust Rule: The tional Lead Exposure, ” in Environmental and Role of Productivity Zmproving Technology, R. Occupational Medicine, W. N. Rem, A. D. Ruttenberg, contract report prepared for the Of- Renzetti, J. S. Lee, et al., eds. (Boston: Little, fice of Technology Assessment, U.S. Congress, Brown & Co., 1983). Washington, DC, March 1983. 175. Fischhoff, B., “Cost-Benefit Analysis: An Un- 161. Enterline, P. E., memorandum on “Comments certain Guide to Public Policy” in Management on ‘Estimates of the Fraction of Cancer in the of Assessed Risk for Carcinogens, W. J. United States Related to Occupational Factors’ “ Nicholson (cd.) (New York: New York Acad- (Washington DC: Asbestos Information Asso- emy of Sciences, 1981). ciation, 1978). 176, Ford, G. R., “Inflation Impact Statements,” Ex- 162. Enterline, P. E., “Cancer Caused by Asbestos, ” ecutive Order No. 11821, Federal Register in Quantification of Occupational Cancer: Ban- 39:41502, Nov. 29, 1974. bury Report 9, R. Peto and M. Schneiderman 177. Ford, G. R., “Economic Impact Statements, ” Ex- (eds. ) (Cold Spring Harbor, NY: Cold Spring ecutive Order No. 11949, Federal Register Harbor Laboratory, 1981). 42:1017, Dec. 31, 1976. 163, Environmental Health Letter, “EPA Seeks to 178. Fortune, “Look Who’s Covered in Red Tape,” Establish More Consistent Civil Penalty Policy,” interview with Jewell Westerman, May 4, 1981, p. 2, Apr. 1, 1984. pp. 357-62. 164 Epstein, S. S., and Swartz, J. B., letter, Science 178a. Franklin, D., “From Bust to Dust: Protective 224:660-666, 1984. Masks on Trial, ” Science News 125(25):394-5, 165 Erdreich, J., reported in “Higher Noise Doses June 23, 1984. Will Be Calculated with 3dB Trading Ratio, Er- 179, Freedman, A., lndu.st~ Response to Health Risk dreich Reports, ” Occupational Safety and (New York: The Conference Board, 1981). Health Reporter, 13(51):1346-7, May 24, 1984. 180, Froines, J., paper presented at First International 400 ● Preventing Illness and Injury in the Workplace

Respirator Research workshop, sponsored by 193. Greek, “Vinyl Chloride May Face Shortage by NIOSH, Sept. 9-11, 1980, quoted in “Working 1977, ” Chemical and Engineerz”ng News, p. 9, Paper on Problems Associated with Respiratory Aug. 11, 1975, quoted in The Law and Policy Protection Programs,” M. Nicas, typescript of Toxic Substances Control, D. D. Doniger (Washington, DC: Workers’ Institute for Safety (Washington, DC: Resources for the Future, and Health, May 11, 1981). 1978). 181, Gillen, M., letter to Dr. Anthony Robbins, Di- 194 Green, M., and Waitzman, N., “Cost, Benefit, rector, National Institute for Occupational and Class, ” Working Papers, pp. 39-51, May- Safety and Health, Sept. 4, 1980. June 1980. 182, Gleason, J. M., and Barnum, D. T., “Effective- 195. Green, M., and Waitzman, N,, Business War on ness of OSHA Sanctions in Influencing Employ- the Law: An Analysis of the Benefi”ts of Federal er Behavior: Single and Multi-Period Decision Health/Safety Enforcement, 2d ed. (Washing- Models,” Accident Analysis & Prevention 10:35- ton, DC: Corporate Accountability Research 49, 1978. Group, 1981). 183. Ginnold, R. E., “A View of the Costs and Bene- 196. Hack, A., Bradley, 0. D., and Trujillo, A., Res- fits of the Job Safety and Health Law,” Monthly pirator Studies for the Nuclear Regulatory Com- Labor Review, pp. 24-26, August 1980. mission, Oct. 1, 1976 Sept. 30, 1977, Protection 184. Goble, R., Hattis, D., Ballew, M., and Thur- Factors for Supplied-Air Respirators (Los ston, D., lrnplementation of the Occupational Alamos, NM: Los Alamos Scientific Labora- Lead Exposure Standard (Working Paper # 16 tory, January 1978). of Preventing Illness and Injury in the Work- 197, Haddon, W., Jr., Suchman, E. A., and Klein, place), contract report prepared for the Office D., (eds. ) Accident Research Methods and Ap- of Technology Assessment, U.S. Congress, proaches (New York: Harper and Row, 1964). Washington, DC, November 1983. 198, Haddon, W., Jr., and Baker, S. P., “Injury Con- 185 Goldsmith, F., and Kerr, L., Occupational trol, ” in Preventive and Community Medicine, Safety and Health: The Prevention and Control D. W. Clark and B. MacMahon (eds. ) (Boston: of Work-related Hazards (New York: Human Little, Brown & Co., 1981). Sciences Press, 1982). 199, Hamilton, A., Exploring the Dangerous Trades 186, Gordon, J., Akman, A., and Brooks, M., in- (Boston: Little, Brown & Co., 1943).

dustrial Safety Statistics: A Re-examination 200< Harris, H. E., Coal Mine Dust Respiratory Pro- (New York: Praeger, 1971). tection Devices, Final Report, Contract H-CPE 187, Gordon, J. E., “The Epidemiology of Acci- 70-127 (Cincinnati, OH: National Institute for dents,” in Accident Research Methods and Ap- Occupational Safety and Health, October 1973). proaches, W. Haddon, Jr., E. A. Suchman, and 201, Harris, H. E., Burgess, W. A., and DeSieghardt, D. Klein (eds. ) (New York: Harper and Row, W. C., “Effectiveness of Dust Respirators in 1964). Underground Coal Mines, ” Society of Mining 188. Gorham, M., “Bum Rap for OSHA?” Federal Engineers Transactions, pp. 153-61, 1973. Reserve Bank of San Francisco Weekly Letter, 202. Harris, H. E., DeSieghardt, W. C., Burgess, W, pp. 1-3, Jan. 19, 1979. A., et al., “Respirator Usage and Effectiveness 189. Gottleib, M., and Coleman, P., “Inspection Im- in Bituminous Coal Mining Operations, ” Amer- pact on Injury and Illness Totals, ” Wisconsin ican Industrial Hygiene Association Journal Department of Industry, Labor, and Human Re- 35:159-64, 1974. lations, typescript, no date. 203. Harter, P., Non-Regulatory Legal Incentives for 190. Gough, M., “Sources and Interpretations of the Adoption of Occupational Safety and Health Asbestos Exposure Data, ” Journal of Toxicolo- Control Technologies (Working Paper # 8 of gy–clinical Toxicology 21:211-236, 1984. Preventing Illness and Injury in the Workplace), 191. Graham, J., and Shakow, D,, “Risk and Re- contract report prepared for the Office of Tech- ward: Hazard Pay for Workers, ” Environment nology Assessment, U.S. Congress, Washing- 23(8):14-20,44, October 1981. ton, DC, April 1983. 192. Graham, J., Shakow, D. M., and Cyr, C., “Risk 204. Harvard Law Review, Notes, “Compensating Compensation—in Theory and in Practice, ” Victims of Occupational Disease,” 93(5):916-37, Environment 25(1):14-20,39-40, January- March 1980. Febrary 1983. 205. Hastings Center, Institute of Societv, Ethics and References ● 401

the Life Sciences, “Values, Ethics, and CBA in 217, Hughes, J. T., “Byssinosis Compensation in Health Care,” in The Implications of Cost- North Carolina, ” in Legal and Ethical Dilemmas Eflectiveness Analysis of Medical Technology, in Occupational Health, J. S. Lee and W. N. U.S. Congress, Office of Technology Assess- Rom (eds. ) (Ann Arbor, MI: Ann Arbor Science, ment (Washington, DC: U.S. Government 1982), Printing Office, August 1980). 218. Hunter, D., The Diseases of Occupations, 6th 206. Hattis, D., Mitchell, C., McC1eary-Jones, J., et ed. (London: Hodder and Stoughton, 1978). al., Control of Occupational Exposures to For- 219. Hyatt, E. C., Respirator Protection Factors, In- maldehyde: A Case Study of Methodology for formal Report LA-6084-MS (Los Alamos, NM: Assessing the Health and Economic Impacts of Los Alamos Scientific Laboratory, January OSHA Health Standards, report prepared for 1976). the U.S. Department of Labor (Cambridge, MA: 220. ICF, Inc., Development of a Methodology for Massachusetts Institute of Technology, Center Production of Industry Specific Research Plan- for Policy Alternatives, April 1981). ning Information, report prepared for the U.S. 207. Heinrich, H. W., Industrial Accident Preven- Environmental Protection Agency and the Na- tion: A Scientific Approach, 4th ed. (New York: tional Institute for Occupational Safety and McGraw-Hill Book Co., 1959). Health, Washington, DC, April 1982.

208. Held, B. J., “Personal Protection, ” in Patty’s in- 222< industrial Safety and Hygiene News, “Certify- dustrial Hygiene and Toxicology, vol. III, L. J. ing Safety Products, ” pp. 1-3, July 1983. Cralley and L. V. Cralley (eds. ) (New York: 223< industrial Union Department, AFL-C1O v. John Wiley and Sons, Inc., 1979). Hodgson, 499 F,2d 467, 474 (D. C. Cir., Apr. 209. Henigson, S., “The Oil Drilling Blacklists, ” 15, 1974). Dallas Times Herald, Nov. 30, 1981. 224, industrial Union Department, AFL-C1O V. 210. Hickey, J. L. S., Rice, C, H., and Boehlecke, American Petroleum Institute, 448 U.S. 607 (Su- B. A., Technologies for Controlling Worker preme Court, July 2, 1980). Exposure to Silica (Working Paper # 17 of Pre- 225. [nfante, P. F., “Vinyl Chloride: A Case History venting Illness and Injury in the Workplace), of Regulatory Action in Relation to Scientifc contract report prepared for the Office of Tech- Knowledge of Cancer Causing Effects, ” paper nology Assessment, U.S. Congress, Washing- presented at the Canadian Cancer Society Inter- ton, DC, August 1983. national Conference on Carcinogens in the 211. Hickey, J. L. S., Wright, M. D., Warren, J. L., Workplace, Vancouver, BC, May 1983. et al., Research Triangle Institute, Abrasive 226. INFORM, At Work in Copper (New York: Blasting, Technological Feasibility Assessment INFORM, 1979). and Economic Impact Analysis of the Proposed 227. INFORM, Worker Training and Education Pro- Federal Standard, contract No. J-9-F-6-0225, re- grams in Occupational Health and Safety port prepared for the Occupational Safety and (Working Paper # 13 of Preventing Illness and Health Administration, U.S. Department of La- Injury in the Workplace), contract report pre- bor, Washington, DC, September 1978. pared for the Office of Technology Assessment, 212. Higginson, J., Bahar, J. C., Clemmesen, J., et U.S. Congress, Washington, DC, November al., “Proportion of Cancers Due to Occupa- 1983. tions,” Preventive Medicine 9:180, 1980. 228 Interagency Task Force on Workplace Safety 213. Hilaski, H., “Understanding Statistics on Oc- and Health, Making Prevention Pay (Washing- cupational Illness, ” Monthly Labor Review, pp. ton, DC, 1978). 25-9, March 1981. 229. International Labour Office, Encyclopedia of 214. Hilaski, H., and Wang, C. L., “How Valid are Occupational Safety and Health, vols. I & II Estimates of Occupational Illness?” Monthly La- (New York: McGraw-Hill Book Co., 1976). bor Review, pp. 27-35, August 1982. 230, International Labour Office, Encyclopedia O) 215. Hoerger, F. D., “Indicators of Exposure Trends, ” Occupational Safety and Health, vols. I & II, in Quantification of Occupational Cancer: Ban- 3d ed. (Geneva: International Labour Office, bury Report 9, R. Peto and M. Schneiderman 1983) (eds. ) (Cold Spring Harbor, NY: Cold Spring 231 Iverstine, J. C., Kinard, J. L., and Slaughter, Harbor Laboratory, 1981). W. S,, impact of Environmental Protection Reg- 216. Hogan, M. D., and Heel, D. G., “Estimated ulations on Research and Development in the Cancer Risk Associated with Occupational As- Industrial Chemical Industry, report prepared bestos Exposure, ” Risk Analysis 1:67-76, 1980. for the National Science Foundation (May 1978), 402 . Preventing Illness and Injury in the Workplace

cited in Ashford, N. A., Heaton, G. R., Jr., and 245. Kelman, S., Regulating America, Regulating Priest, W. C., “Environmental, Health, and Sweden (Cambridge, MA: MIT Press, 1981). Safety Regulation and Technological Innova- 246. Kelman, S., “Cost-Benefit Analysis: An Ethical tion, ” in Technological Change for a Dynamic Critique,” Regulation, pp. 33-4o, January- Economy, C. T. Hill and J. M. Utterback (eds.) February 1981. (New York: Pergamon Press, 1979). 247 Kendall, D., “Etiology, Diagnosis and Treat- 232, Javits, J., Williams, H., and Taft, R., letter to ment of Paresthesia in the Hand, ” British Med- Casper Weinberger, Secretary of Health, Edu- ical ]oumal 2:1633-1640, 1960. cation, and Welfare, Aug. 3, 1973. 248 Kent, M, B., personal communication, 1983. 233. John, G., Small Business Administration, per- 249. Kent, M. B., A History of Occupational Safety sonal communication, Apr. 7, 1983. and Health in the United States (Working Pa- 234. Johnson, W. G., “Compensation for Occupa- per #4 of Preventing Illness and Injury in the tional Illness,” in Management of Assessed Risk Workplace), contract report prepared for the for Carcinogens, W. Nicholson (cd.) vol. 363 Office of Technology Assessment, U.S. Con- (New York: Annals of the New York Academy gress, Washington, DC, April 1983. of Sciences, 1981). 249a. Kent, M. B., personal communication, 1984. 235, Johnson, W. G., Management Oversight and 250. Keyserling, W. M., “Occupational Safety and Risk Tree (Washington, DC: U.S. Government Ergonomics,” in Occupational Health, B. S. Printing Office, 1973). Levy and D. H. Wegrnan (eds.) (Boston, MA: 236, Johnson, W. G., and Heler, E., ‘The Costs of Little, Brown & Co., 1983), Asbestos-Associated Disease and Death,” Mil- 251. Kneese, A. V., and Schultze, C, L., Pollution, bank Memorial Fund Quarterly, Health and Prices, and Public Policy (Washington, DC: Society, 61(2):177-194, Spring 1983. Brookings Institution, 197s). 237. Jones, C. A., ModeZs of Regulatory Enforcement 252 Kochan, T., and Dyer, L,, “A Model of Organi- and Compliance, with an Application to the Oc- zational Change in the Context of Union-Man- cupational Safety and Health Administration agement Relations, ” Journal of Applied Asbestos Standard, unpublished Ph.D. disser- Behavioral Science, 12(1):59-78, February 1976. tation, Harvard University, Department of Eco- 253 Kochan, T., Dyer, L., and Lipsky, D., The Ef- nomics, July 1982. fectiveness of Union-Management Safety and 238( ]oumal of Occupational Medicine, “Medical In- Health Committees (Kalamazoo, MI: W.E. Up- formation ~ystems Roundtable,” 24(10):781- john Institute for Employment Research, 1977). 866, October 1982, Supplement. 254. Kossoris, M. D., “Industrial Injuries and the 239, Kantowitz, B., and Sorkin, R., Human Factors: Business Cycle,” Monthly Labor Review, pp. Understanding People-System Relationships 579-94, March 1938. (New York: John Wiley and Sons, Inc., 1983). 255. Kostem, M. H. “Counting the Costs: The Busi- 240. Karlson, T., and Baker, S., “Fatal Occupational ness Roundtable Study, ” Regulation, pp. 17-25, Injuries Associated with Motor Vehicles,” Pro- July-August 1979. ceedings of the zzd Conference of the Ameri- 256 Krause, F. E., ‘The Prevention of PVC Reactor can Association for Automotive Medicine 1:229- Fouling,” in Symposium Proceedings—Control 41, 1978. Technology in the Plastics and Resins Industry 241. Kay, 1832, quoted in Historical Perspective: The DHHS Publication No. (NIOSH) 81-107 (Cin- Evolution of the Definition of Certain Work- cinnati, OH: NIOSH, January 1981). Related Illnesses, J. K. Corn, contract report pre- 257. Kronebusch, K,, “Occupational Injury Data: pared for the Office of Technology Assessment, Are We Collecting What We Need for Identifica- U.S. Congress, Washington, DC, July 1983. tion, Prevention, and Evaluation?” in Prion”ties 242. Kasperson, R. E., ‘Worker Participation in Pro- in Health Statistics, pp. 124-129, Proceedings of tection: The Swedish Alternative,” Environ- the 19th National Meeting of the Public Health ment, 25(4):13-20,40-43, May 1983. Conference on Records and Statistics, August 243 Keller, B., ‘Workers Are Often a Hazard to 1983, DHHS Publication No. (PHS) 81-1214 Themselves,” New York Times, sec. 4, p. 8E, (Hyattsville, MD: National Center for Health July 8, 1984. Statistics, December 1983). 244 Kelley, W. D., American Conference of Gov- 258. LaDou, J., ‘The Not-So-Clean Business of Mak- ernmental Industrial Hygienists, letter to OTA, ing Chips,” Technology Review, 87(4):22-36, Oct. 17, 1983. May-June 1984. References ● 403

259. Landefeld, J. S., and Seskin, E, P,, “The Eco- 269, Levy, B. S., and Wegman, D. H., (eds. ) Oc- nomic Value of Life: Linking Theory to Prac- cupational Health: Recognizing and Preventing tice, ” American ]ournal of Public Health Work-Related Disease (Boston: Little, Brown & 72(6):555-66, June 1982. Co., 1983). 259a. Landrigan, P., Cohen, M., Dowdle, W., et al., 269a, Levy, B. S., “The Teaching of Occupational “Medical Surveillance of Biotechnology Work- Health in United States Medical Schools: Five- ers: Report of the CDCINIOSH Ad-Hoc Working Year Follow-up of an Initial Survey,” American Group on Medical Surveillance for Industrial lournal of Public Health 75(1):79-80, January Application of Biotechnology.” Recombinant 1985. DNA Technical Bulletin 5(3):133-38, September 270, Lewis-Beck, M. S., and Alford, J. R., “Can Gov- 1982. ernment Regulate Safety? The Coal Mine Exam- 260. Larson, A,, Workmens’ Compensation Law, ple,” American Political Science Review 74:745- vol. 1 (New York: Mathew Bender, 1973) cited 756, 1980. in Barth, P. S., with Hunt, H. A., Workers’ 271 Loss, J., Architecture and Engineering Perform- Compensation and Work-Related Illnesses and ance Impact Center, University of Maryland, Diseases (Cambridge, MA: MIT Press, 1980). personal communication, fall 1983. 261. Larson, L. W,, “Analysis of Current Laws Re- 272< Lowry, P. L,, Hesch, P. R., and Revoir, W. H., flecting Worker Benefits for Occupational Dis- “Performance of Single-Use Respirators, ” Amer- eases, ” report prepared for the Assistant Secre- ican Industrial Hygiene Association ]ournal tary for Policy, Evaluation and Research, U.S. 38:462-467, 1977. Department of Labor, Washington, DC, NTIS 273< Lublin, J, S. “Tougher Limits For Asbestos Report No. ASPER/PUR-78/4385 /A, May Issued by U. S.” The Wall Street Journal, Nov. 1979. 3, 1983. 262. Lave, J. R., and Lave, L. B., Decision 274. Lubove, R., “Workmen’s Compensation and the Frameworks to Enhance Occupational Health Prerogatives of Voluntarism, ” Labor History and Safety Regulation (Working Paper #6 of 8(3):254-79, fall 1967. Preventing Illness and Injury in the Workplace), 275. Luopajarvi, T., Kuorinka, 1,, Virolainen, M., contract report prepared for the Office of Tech- et al., “Prevalence of Tenosynovitis and Other nology Assessment, U.S. Congress, Washing- Injuries of the Upper Extremities in Repetitive ton, DC, February 1983. Work, ” Scandinavian Journal of Work, Envi- 263. Lave, L. B., testimony, in U.S. Congress, House ronment and Health, vol. 5, Supplement 3, Committee on Interstate and Foreign Com- 1979. merce, Use of Cost-Benefit Analysis by Regula- 276. MacAvoy, P. (cd. ) OSHA Safety Regulation: tory Agencies, joint hearings before the Sub- Report of the Presidential Task Force (Washing- committee on Oversight and Investigations and ton, DC: American Enterprise Institute, 1977). the Subcommittee on Consumer Protection and 277. MacCarthy, M., “A Review of Some Normative Finance, July 30, Oct. 10 and 24, 1979 (Wash- and Conceptual Issues in Occupational Safety ington, DC: U.S. Government Printing Office, and Health, ” Boston College Environmental Af- 1980). fairs Law Review 9(4):773-814, fall 1981. 264. Lave, L. B., The Strategy of Social Regulation 277a. Machinery and Allied Products Institute, Oc- (Washington, DC: The Brookings Institution, cupational Safety and Health: Transcript of a 1981 ). MAPI Seminar, Feb. 21-22, 1973, quoted in 265, Leamon, T. B,, “The Introduction of Ergonom- Kent, M. B., A History of Occupational Safety ics: A Problem of Industrial Practice, ” Applied and Health in the United States, contract report Ergonomics 11(3):161-64, 1980. prepared for the Office of Technology Assess- 266< Leamon, T. B., ‘The Organization of Industrial ment, U.S. Congress, Washington, DC, April Ergonomics—A Human :Machine Model,” Ap- 1983. plied Ergonomics 11(4):223-26, 1980. 278. MacLaury, J., cd., Protecting People at Work: 267. Lempert, B. L., and Edwards, R. G., “Field In- A Reader in Occupational Safety and Health vestigations of Noise Reduction Afforded by (Washington, DC: U.S. Department of Labor, Insert-Type Hearing Protectors, ” American In- 1980). dustrial Hygiene Association Journal (in press). 279. MacLaury, J., “The Job Safety Law of 1970: Its 268. Levenstein, C., University of Connecticut, per- Passage was Perilous, ” Monthly Labor Review sonal communication, 1983. 104(3):18-24, March 1981. 404 ● Preventing Illness and Injury in the Workplace

280. MacLean, D., A History of the Consideration “Mesothelioma as an Index of Asbestos Impact,” of Economic Impacts in the Determination of the in Quantiji”cation of Occupational Cancer: Ban- Feasibility of Standards Under the OSH Act bury Report 9, R. Peto and M. Schneiderman (Working Paper #s of Preventing Illness and In- (eds. ) (Cold Spring Harbor, NY: Cold Spring jury in the Workplace), contract report prepared Harbor Laboratory, 1981). for the Office of Technology Assessment, U.S. 295. McFarland, R. A., The Epidemiology of Indus- Congress, Washington, DC, April 1983. trial Accidents (Cambridge, MA: Harvard 281 MacLean, D, and Sagoff, M., A Critique of School of Public Health, 1965). Cost-Benefit Analysis as a Technique for Deter- 296. McGarity, T. O., “Substantive and Procedural mining Health Standards under the Occupation- Discretion in Administrative Resolution of al Safety and Hea2th Act (Working Paper #7 of Science Policy Questions, ” Georgetown Law Preventing Illness and Injury in the Workplace), ]ournal 67:729-810, February 1979. contract report prepared for the Office of Tech- 297. McGarity, T. O., “Media Quality, Technology, nology Assessment, U.S. Congress, Washing- and Cost-Benefit Balancing Strategies for Health ton, DC., April 1983. and Environmental Regulation, ” Law and Con- 282. MacMahon, B. and Pugh, T. F., Epidemiology: temporary Problems 46:159-233, summer 1983. Principles and Methods (Boston: Little, Brown, 297a. McGarity, T. O., personal communication, June 1970). 8, 1984. 283. Magaziner, I. C., and Reich, R. B., Minding 298. McGraw-Hill Publications Company, Econom- America’s Business (New York: Harcourt Brace ics Department, “Annual McGraw- Hill Survey Jovanovich, 1982). of Investment in Employee Safety and Health, ” 284. Main, J., “When Accidents Don’t Happen,”For- mimeo, various years. tune 106(5):62-68, Sept. 6, 1982, 299. McKenzie, F., Storment, J., Van Hook, P., et 285. Maisonpierre, A., statement of the Alliance of al., “A Program for Control of Repetitive American Insurers before the Senate Labor and Trauma Disorders in a Telephone Factory, ” Human Resources Committee, Aug. 26, 1980. typescript, no date. 286. Maisonpierre, A., Alliance of American In- 300. Mendeloff, J., Regulating Safety: An Economic surers, letters to OTA, Sept. 16, 1983, and Jan. and Political Analysis of Occupational Safety 13, 1984. and Health Policy (Cambridge, MA: MIT Press, 287. Marlow, M. L., “The Economics of Enforce- 1979). ment: The Case of OSHA, ” Journal of Econom- 301. Mendeloff, J., An Analysis of OSHA Health in- ics and Business 34:165-171, 1982. spection Data (Working Paper # 2 of Prevent- 288. Mazor, J., “How Accurate Are Employers’ Ill- ing Illness and Injury in the Workplace), con- ness and Injury Reports?” Monthly Labor Re- tract report prepared for Office of Technology view, pp. 26-31, September 1976. Assessment, U.S. Congress, Washington, DC, 289, McCaffrey, D. P,, ‘Work-Related Amputations April 1983. by Type and Prevalence” Monthly Labor Re- 302. Mendeloff, J., ‘The Role of OSHA Violations view, pp. 35-41, March 1981. in Serious Workplace Accidents,” Journal of Oc- 290, McCaffrey, D. P., OSHA and the Politics of cupational Medicine 26(5):353-60, May 1984. Health Regulation (New York: Plenum Press, 303. Millar, J. D,, CDC White Paper–The NIOSH 1982). Move Memorandum (Atlanta, GA: Centers for 291. McCaffrey, D. P., “An Assessment of OSHA’S Disease Control, Mar. 5, 1982). Recent Effects on Injury Rates, ” Journal of 304. Miller, J. C., III, and Yandle, B., Benefit-Cost Human Resources, January 1983. Analyses of Social Regulation (Washington, 292. McCormick, E. J., and Sanders, M., Human DC: American Enterprise Institute, 1979). Factors in Engineering and Design (New York: 305. Miller, W., “MSA’S Comments Relevant to the McGraw-Hill Book Company, 1982). Advantages and Disadvantages of In-House 293. McDaniel, P. R,, and Kaplinsky, A. S., ‘The Versus External Testing of Respirators, ” Pre- Use of the Federal Income Tax System to Com- sented to Respirator Subcommittee of the Mine bat Air and Water Pollution,” Boston College Health Advisory Committee, Jan. 4 and 5, 1982. Industrial and Commercial Law Review 12(3): 306. Mine Health Research Committee, Respirator 351-86, February 1971. Research Subcommittee, Respirator Certifica- 294. McDonald, J. C., and McDonald, A. D., tion, typescript, Jan. 27, 1982. .

References ● 405

307. Mintz, B. W,, OSHA: History, Law, and Pol- 323. National Safety Council, “This is the National icy (Washington, DC: Bureau of National Af- Safety Council, ” 1980. fairs, 1984). 324. National Safety Council, Accident Facts (Chi- 308. Mishan, E. J., Cost-Benefit Analysis (New York: cago: National Safety Council, 1981). Praeger Publishers, 1971). 325. National Safety Council, “Documentation of 309. Moran, J., 1980, quoted in “Comments Sub- National Safety Council Statistics Department mitted to OSHA on the Issue of the Revised Estimating Procedures for Motor Vehicle,

Respirator Use Policy, ” J. M. Paul] and F. Workr Home, and Public Deaths and Death Rosenthal, OSHA Docket H-160, Exhibit 2-70, Rates,” typescript, February 1982. typescript, 1983. 326. Nelson, T. J., E.I. DuPont de Nemours & Co., 310. Moran, J., personal communication, December Inc., Wilmington, DE., personal communica- 1983. tion, August 1983. 311. Monitor, “Error-Free Performance Starts in the 327. Neumann, G. R., and Nelson, J. P., “Safety Reg- Classroom, ” pp. 18-19, May 1982. ulation and Firm Size: Effects of the Coal Mine 312. Monson, R, R., Occupational Epidemiology Health and Safety Act of 1969,” Journal of Law (Boca Raton, FL: CRC Press, Inc., 1980). and Economics 25:183-199, October 1982. 313. Moskowitz, D. B., “Unions Shut Out in Safety 328. IVew York Times, “N. L.R.B. Says Safety Pro- Cases,” Washington Post, Jan. 2, 1984. test by Lone Worker Was Illegal, ” p. A-26, Jan. 314, Myers, W. R. and Peach, M. J., III., “Perform- 12, 1984. ance measurements on a powered air-purifying 329. Nicas, M., ‘Working Paper on Problems Asso- respirator made during actual field use in a silica ciated with Respiratory Protection Programs, ” bagging operation, ” typescript (Morgantown, typescript (Washington, DC: Workers’ Institute WV: National Institute for Occupational Safety for Safety and Health, May 11, 1981). and Health, 1982). 330. Nichols, A. L., and Zeckhauser, R., “Govern- 315. Myers, W., data presented in Report of Scien- ment Comes to the Workplace: An Assessment tific Peer Group Review of Respirator Quanti- of OSHA, ” The Public Interest 49:39-69, fall tative Fit Test Methods, D. L. Swift, C. E. Bill- 1977.

ings, and F. Shanty (eds. ) typescript (Baltimore: 331< Nicholson, W. J., Perkel, G., Selikoff, I. J., et Johns Hopkins University, Sept. 22, 1983). al., “Cancer From Occupational Asbestos Ex- 316, National Commission on State Workmen’s posure Projections 1980-2000, ” in Quantifica- Compensation Laws, Compendium on Work- tion of Occupational Cancer: Banbury Report men’s Compensation (Washington, DC: U.S. 9, R. Peto and M. Schneiderman (eds. ) (Cold Government Printing Office, 1973). Spring Harbor, NY: Cold Spring Harbor Lab- 317. National Commission on State Workmen’s oratory, 1981). Compensation Laws, Z?eport (Washington, DC: 332< Northrup, H. R., Rowan, R. L., Perry, C. R., U.S. Government Printing Office, 1973). et al., The Impact of OSHA (Philadelphia, PA: 318. National Commission on State Workmen’s University of Pennsylvania, The Wharton Compensation Laws, Supplemental Studies School, Industrial Research Unit, 1978). (Washington, DC: U.S. Government Printing 333 Nothstein, G., The Law of Occupational Safety Office, 1973). and Health (New York: MacMillan, The Free 319. National Research Council, Assembly of Engi- Press, 1981).

neering, Marine Board, Committee on Assess- 334< Occupational Hazards, “ISEA 50th Anniver- ment of Safety of OCS Activities, Safety and sary, ” pp. 29-31, May, 1983. Offshore Oil (Washington, DC: National Acad- 335. Occupational Health & Safety Letter, “Study of emy Press, 1981). in-plant health committees yields mixed results, ” 320, National Research Council, Committee on In- p. 4, Dec. 8, 1982. door Pollutants, Indoor Pollutants (Washing- 336. Occupational Health & Safety Letter, “Health ton, DC: National Acadamy Press, 1983). and Safety Being Promoted in Business Schools, ” 321, National Research Council, Video Displays, p. 8, May 22, 1983. Work, and Vision (Washington, DC: National 337. Occupational Health & Safety Letter, “Positive Academy Press, 1983). results reported on 6 Cal-OSHA self-inspection 322. National Safety Council, Accident Prevention projects, ” p. 6, Dec. 8, 1983. Manual for Industrial Operations: Engineering 338 Occupational Health & Safety Letter, “Gulf and Technology, 8th cd., F. E. McElroy (cd. ) Coast Company Gets Experimental Variance (Chicago, IL: National Safety Council, 1980). From Lead Standard,” 14(7):4-5, Apr. 8, 1984. 406 . Preventing Illness and Injury in the Workplace

338a, Occupational Health & Safety Letter, “Occupa- ployee Assessment, Advisory Council Remain,” tional Medical Practice Shifting to Hospitals, ” pp. 1309-1310, May 10, 1984. 14(13):1-2, July 8, 1984. 352. Occupational Safety & Health Reporter, 339. Occupational Safety & Health Reporter, “ACGIH Makes Revisions, Additions to Thresh- 5(24):829, Nov. 13, 1975, cited in Technologies old Limit Value Listings, ” 13(52):1355, May 31, for Controlling Worker Exposure to Silica, J. L. 1984

S. Hickey, C. H. Rice, and B. A. Boehlecke, 3534 Occupational Safety & Health Reporter, contract report prepared for the Office of Tech- “Goundwork Must Be Laid if Committees Are nology Assessment, U.S. Congress, Washing- to Be Effective, Speaker Maintains, ” pp. 8-9, ton, DC, August 1983. June 7, 1984. 340. Occupational Safety & Health Reporter, 354. Occupational Safety & Health Reporter, ‘Trans- “Ergonomics can Boost Productivity, Reduce In- fer of 4,4’-MDA Control to OSHA Approved juries, NIOSH Consultant Says, ’ 12(29):584-5, by EPA Official; Report Considered,” pp. 7-8, &C. 16, 1982. June 7, 1984. 341. Occupational Safety & Health Reporter, “Little 355. Occupational Safety & Health Reporter, Impact on Safety Seen From Mere Existence of “Revised Rules Expanding Program, Providing Committee,” pp. 519-20, Nov. 25, 1982. Exemptions Issued by OSHA,” 14(3):35-36, June 342. Occupational Safety & Health Repotier, “Meas- 21, 1984. ure Before Legislature Would Permit Worker to 358. O’Malley, T., U.S. Department of Labor, per- Sue Employer for Work Accident,” p. 776, Feb. sonal communication, Sept. 25, 1983. 17, 1983. 359. Ohman, K. H. G., “Prevention of Silica Ex- 343. Occupational Safety & Health Reporter, posure and Elimination of Silicosis, ” Amen-can ‘limitations on OSHA Inspections under Fund- Zndusti”al Hygiene Association Journal 39(11): ing Bill Explained to Field, ” 12(45):961-2, Apr. 847-859, 1978. 14, 1983, 360. Oppold, J. A., “NIOSH Respirator Research 344. Occupational Safety & Health Reporter, “Millar Program, ” presented to Respirator Subcommit- Returns Certification Draft to Staff, Seeks Bal- tee of the Mine Health Advisory Committee, ance in Revision, ” 13(9):197, July 28, 1983. Aug. 13 and 14, 1981. 345. Occupational Safety & Health Reporter, “Areas 361. Page, J., and O’Brien, M., Bitter Wages; Ralph for New Federal Role Suggested by Bingham; Nader’s Study Group Report on Disease and in- ‘Right to Act’ Movement Urged, ” pp. 677-678, jury on the Job (New York: Grossman, 1973). Nov. 24, 1983. 362. Patty, F, A. (cd.), Industrial Hygiene and Tox- 346. Occupational Safety & Health Reporter, “Joint icology, 2d cd., vol. I (New York: John Wiley Review of Health Risks to Be Made by OSHA, and Sons, Inc., 1958). EPA, Based on Animal Test Data,” p. 816, Jan. 363. Paull, J. M., and Rosenthal, F., “Comments 5, 1984. Submitted to OSHA on the Issue of the Revised 347. Occupational Safety & Health Reporter, “Lower Respirator Use Policy,” OSHA Docket H-160, Injury Rates, Costs Seen as Result of Voluntary Exhibit 2-70, typescript, 1983. Compliance, CALfOSHA States,” p. 822, Jan. 364. Pen, S., “An Evaluation of a Hearing Conser- 5, 1984. vation Program —A Five-year Longitudinal 348. Occupational Safety 6 Health Reporter, Study, ” American Industrial Hygiene Associa- “OSHA’S Grain Elevator Proposal Suggests tion Journal, pp. 82-91, 1972. Three Methods for Controlling Dust Levels, ” 365. Perry, C. R., “Vinyl Chloride Protection: Less 13:875-75, Jan. 12, 1984. Costly than Predicted,” Monthly Labor Review, 349. Occupational Safety & Health Reporter, “Lead pp. 22-24, August 1980. Company Given Six Month Variance From Lead 366. Peterson, C., “OSHA May Drop Standard- Standard to Test New Respirator,” p. 1180, Apr. Setting Efforts, ” Washington Post, p. A2, Sept. 5, 1984. 21, 1983, . 350! Occupational Safety & Health Reporter, “Bench- 367. Peterson, C., “Labor Reduces Exposure Limit mark Language Was Seen as ‘Lost Issue, ’ for For Asbestos,” Washington Post, Nov. 3, 1983. Fiscal 1985 Request, Tyson Tells Panel,” 13(46): 368. Petersen, D., Safety Management—A Human 1256, Apr. 19, 1984. Approach (Englewood, NJ: Doran, 1975). 351. Occupational Safety & Health Reporter, ‘Wash- 369. Peterson, J. E., “Principles for Controlling the ington Governor Vetoes Major Part of Law; Em- Occupational Environment, ” in The Industrial .—

References ● 407

Environment–Its Evaluation and Control, U.S. 380. Purswell, J. L., and Stephens, R., Heakh and Department of Health, Education, and Welfare, Safety Control Technologies in the Workplace: National Institute for Occupational Safety and Accident Causation and Injury Control (Work- Health (Washington, DC: U.S. Government ing Paper # 11 of Preventing Illness and Injury Printing Office, 1973). in the Workplace), contract report prepared for 370. Pete, J., Henderson, B. E., and Pike, M. C., the Office of Technology Assessment, U.S. Con- ‘Trends in Mesothelioma Incidence in the United gress, Washington, DC, July 1983. States and the Forecast Epidemic due to Asbes- 381. Rattner, S., “Did Industry Cry Wolf?” New tos Exposure During World War II, ” in Quan- York Times, p. Cl, Dec. 28, 1975. tification of Occupational Cancer: Banbury Re- 382. Reagan, R., “Federal Regulation, ” Executive Or- port 9, R. Peto and M. Schneiderman (eds. ) der No. 12291, Federal Register 46(33):13193, (Cold Spring Harbor, NY: Cold Spring Harbor Feb. 17, 1981. Laboratory, 1981). 383< Reich, R. B., “Why the U.S. Needs an Indus- 371. Pete, R., and Schneiderman, M., eds., Quanti- trial Policy, ” Haruard Business Reivew, pp. 74- fication of Occupational Cancer: Banbury Re- 81, January-February 1982. port 9 (Cold Spring Harbor, NY: Cold Spring 384 Reich, R. B., The Next Ametican Frontier (NY: Harbor Laboratory, 1981). Times Books, 1983). 372. Phillips, J. L., E.I. DuPont de Nemours & Co., 385 Reich, R. B., ‘The Next American Frontier,” The Inc., Wilmington, DE., personal communica- Atlantic 251(4):96-108, April 1983. tion, March 1983, 386 Revoir, W, H., Comparison of Results of 373. Policy Research Incorporated, Protecting Work- Qualitative Respirator Facepiece Fitting Test Uti- ers’ Health: Federal, State and Local Compli- lizing lsoamyl Acetate Vapor as Test Agent with ance-Assistance Programs, contract No. HHS- Results of Quantitative Respirator Facepiece Fit- 100-79-0172, report submitted to. the Office of ting Tests Utilizing Polydisperse Dioctyl Phthal- the Assistant Secretary for Planning and Evalua- ate Aerosol as Test Agent, Report No. R-115 tion, Department of Health and Human Serv- (Cranston, RI: Norton Company, Safety Prod- ices, January 1981. ucts Division, Sept. 17, 1979). 374. Posner, R., “A Theory of Negligence, ” ]ournal 387. Richman, J., “Hearing Conservation Screening of Legal Studies 1(1):29-96, January 1972. Programs in Effective Prevention, ” letter, ]our- 375. Powell, C. H., for the Board of Directors of the nal Occupational Medicine 25(8):571, August American Industrial Hygiene Association, let- 1983. ter to T. Auchter, Assistant Secretary of Labor 388. Ries, P. W., “Episodes of Persons Injured: for Occupational Safety and Health, Nov. 14, United States, 1975, ” Advance Data, Number 1983. 18, DHEW (PHS) Publication No. 78-1250 (Hy- 376, President’s Report on Occupational Safety and attsville, MD: National Center for Health Sta- Health for 1971, May 1972, tistics, Mar. 7, 1978). 377, Priest, W. C., Computer-Teleconferencing as a 389, Riko, K,, and Alberti, P. W., “Hearing Protec- Mechanism to Improve Information Transfer for tors: A Review of Recent Observations, ” Jour- Workplace Safety and Health (Working Paper nal of Occupational Medicine 25(7):523-526, #12 of Preventing Illness and Injury in the Work- July 1983. place), contract report prepared for the Office 390. Ringen, K., and Smith, W. J., “Occupational of Technology Assessment, U.S. Congress, Diseases and Equity Issues, ” Virginia Journa2 of Washington, DC, May 1983. Natural Resources Law 2:212-231, 1982. 378, Pritchard, J. A., A Guide to Industrial Respira- 391. Roberts, K., Manville Corporation, Denver, tory Protection, Publication No. 76-189 (Cin- CO,; personal communication, Feb. 6, 1984. cinnati: National Institute for Occupational 392. Rockette, H., “Mortality Among Coal Miners Safety and Health, June 1976). Covered by the UMWA Health and Retirement 379 Presser, W., Law of Torts, 4th ed. (St. Paul, Funds,” DHEW (NIOSH) Publication No. 77- MN: West Publishing Co., 1971) cited in Baram, 155 (Cincinnati, OH: 1977). M. and Miyares, J. R., The Legal Framework 393. Rodenhouse, L, “Statement of the Industrial for Determining Unreasonable Risk From Car- Safety Equipment Association,” Presented to the cinogenic Chemicals, report prepared for the Of- Respirator Research Subcommittee, Mine Health fice of Technology Assessment, U.S. Congress, Research Advisory Committee, Jan. 4 and 5, Mav 16.1980. 1982. 408 . Preventing illness and Injury in the Workplace

394. Rodgers, S. J., and Jayaraman, N., “Control of contract report prepared for the Office of Tech- Respirable Coal Dust in Conventional Coal nology Assessment, U.S. Congress, Washing- Mining Operations,” paper #39 presented at the ton, DC, 1982. American Industrial Hygiene Conference, Phil- 408, Rothstein, M. A., Occupational Safety and adelphia, PA, May 23, 1983. Health Law, 2d ed. (St. Paul, MN: West Pub- 395. Rohatyn, F., statement before U.S. House of lishing Co., 1983). Representatives, Committee on Banking, Fi- 409, Russell, L, B., “Pricing Industrial Accidents, ” in nance, and Urban Affairs, Subcommittee on Suppkmenta2 Studies, National Commission on General Oversight and Renegotiation, Hearing State Workmen’s Compensation Laws (Wash- on Establishment of a National Development ington, DC: U.S. Government Printing Office, Bank and Related Matters, Sept. 15& 28, 1982, 1973), serial # 97-89. 410. Russo, W. J., and Rutledge, G. L., “Plant and 396. Rem, W, N., Renzetti, A. D., Lee, J. S., et al., Equipment Expenditures by Business for Pollu- (eds. ) Environmental and Occupational Medi- tion Abatement, 1983 and Planned 1984, ” Sur- cine (Boston: Little, Brown & Co., 1983). vey of Current Business, pp. 31-34, June 1984. 397. Root, N., and McCaffrey, D,, “Providing More 411, Ruttenberg, R., ‘Why Social Regulatory Policy Information on Work Injury and Illness, ” Requires New Definitions and Techniques for Monthly Labor Review, pp. 16-21, April 1978. Assessing Costs and Benefits, ” Labor Studies 398. Root, N., and Hoefer, M., “The First Work- Journal 6(1):114-23, spring 1981. Injury Data Available From New BLS Study” 412. Ruttenberg, R., “Regulation is the Mother of In- Month2y Labor Review, pp. 76-80, January vention, ” Working Papers, pp. 42-47, May-June 1979. 1981. 399. Root, N., and Daley, J. R., “Are Women Safer 413. Ruttenberg, R., Compliance with the OSHA Workers?” Monthly Labor Review, pp. 3-10, Cotton Dust Rule: The Role of Productivity Im- September 1980. proving Technology (Working Paper #18 of Pre- 400. Root, N., “Injuries at Work Are Fewer Among venting Wness and Injury in the Workplace), Older Employees, ” Monthly Labor Review, pp. contract report prepared for the Office of Tech- 30-34, March 1981. nology Assessment, U.S. Congress, Washing- 401. Rosenberg, M., “Beyond the Limits of Executive ton, DC, March 1983. Power: Presidential Control of Agency Rule- 414. Ryer, R. H., “Air Pollution Exposures to Five making under Executive Order 12,291, ” Michi- Target Health Hazards, ” American Industrial gan Law Review 80:193-247, December 1981. Hygiene Association Journal 39(11):928-931, 402. Rosenthal, F. S., and Paull, J. M., Respirator 1978. program effectiveness analyzed by OSHA com- 415. Safety Sciences, Feasibility of Securing pliance statistics, typescript, 1984. Research-Defining Accident Statistics, NIOSH 403. Rosenthal, F. S., data presented in Report of Technical Report, NIOSH Publication No. 78- Scientific Peer Group Review of Respirator 180, September 1978. Quantitative Fit Test Methods, D. L. Swift, C. 416, Salvendy, G., cd., Handbook of Industrial Engi- E. Billings, and F. Shanty, eds., typescript (Balti- neering (New York: John Wiley and Sons, Inc., more: Johns Hopkins University, Sept. 22, 1982). 1983). 417. Samimi, B., Weill, H., and Ziskind, M., “Res- 404. Rosenthal, F., Johns Hopkins University, School pirable Silica Dust Exposure of Sandblasters and of Hygiene and Public Health, Baltimore, MD., Associated Workers in Steel Fabrication Yards, ” personal communication, June 1983. Archives of Environmental Health, pp. 61-66, 405. Rothfleisch, S., and Sherman, D., “Carpal Tun- 1974. nel Syndrome: Biochemical Aspects of Occupa- 418. Samimi, B., Neilson, A., Weill, H., et al,, “The tional Occurrence and Implications Regarding Efficiency of Protective Hoods Used by Sand- Surgical Management, ” Orthop. Rev., 7:107- blasters to Reduce Silica Dust Exposure, ” Amer- 109, 1978. ican Industrial Hygiene Association Journal, pp. 406. Rothstein, M. A., “OSHA After Ten Years: A 140-148, 1975. Review and Some Proposed Reforms, ”Vander- 419. Sanderson, J. T., “Engineering and the Feds,” bilt Law Review 34:71-139, 1981. Engineering Education, pp. 422-5, March 1983. 407. Rothstein, M., Legal issues Raised by Biochem- 420. Sanderson, W. T., Abrons, H., Petersen, M., ical and Cytogenetic Testing in the Workplace, et al., “Results of Industrial Hygiene Surveys of References . 409

Portland Cement Plants,” paper #34 presented NTIS No. BP-256 238, Report No. ASPER/ at the American Industrial Hygiene Conference, PUR-75/0S40/A, February 197s. Philadelphia, PA, May 23, 1983. 433. Settle R. F., quoted in Business War on the Law: 421. Schauer, L., and Ryder, T., “New Approach to An Analysis of the Benefits of Federal Health/ Occupational Safety and Health Statistics, ” Safety Enforcement, M. Green and N. Waitz- Monthly Labor Review, pp. 14-19, March 1972. man, 2d ed. (Washington, DC: Corporate Ac- 422< Schneider, Scott, “OSHA’S Effectiveness: A Re- countability Research Group, 1981). joinder to McCaffrey, ” typescript, 1983. 434. Shabecoff, P., “Safety Agency to Forgo Cost- 423, Schulte, H. F., “Personal Protective Devices, ” Benefit Analysis, New York Times, July 13, in The lndustn”al Environment—its Evaluation 1981. and Control, U.S. Department of Health, Edu- 435. Shaw, F. M,, “Some Design Criteria for Reduc- cation, and Welfare, National Institute for Oc- ing Dust During the Cleaning and Finishing of cupational Safety and Health (Washington, DC: Iron Castings, ” in Proceedings of Symposium U.S. Government Printing Office, 1973). on Occupation Health Hazard Control Technol- 424. Schultz, G. P., Director, U.S. Office of Man- ogy in the Foundry and Secondary Non-ferrous agement and Budget, memorandum on “Agency Smelting Industries, U.S. Department of Health regulations, standards, and guidelines pertain- and Human Services, National Institute for Oc- ing to environmental quality, consumer protec- cupational Safety and Health, DHHS (NIOSH) tion, and occupational and public health and Publication No. 81-114, August 1981. safety, ” to the Heads of Departments and Agen- 436. Sheridan, P. J., “What Are Accidents Really cies, Oct. 5, 1971. Costing You?” Occupational Hazards, pp. 41- 425. Schultze, C. L., The Public Use of Private in- 43, March 1979. terest (Washington, DC: The Brookings Insti- 437. Shor, G., “An Analysis of the 1978 Survey of tution, 1977). Disability and Work, ” report prepared for the 426. %hultze, C. L., “Industrial Policy: A Dissent, ” Office of the Assistant Secretary for Policy, The 13rookings Review 2(1):3-12, fall 1983. Evaluation, and Research, U.S. Department of 427. Schwartz, J. B., and Epstein, S. S., “Problems Labor, Washington, DC, Dec. 31, 1981. in Assessing Risk From Occupational and Envi- 438. Shoub, E. P., “Statement presented to Respira- ronmental Exposure to Carcinogens, ” in Ban- tor Research Subcommittee of the Mine Health bury Report 9: Quantification of Occupational Research Advisory Committee, ” Jan. 27, 1982. Cancer, R. Peto and M. Schneiderman (eds. ) 439. Simon, M. E., “What We Did: The Business (New York: Cold Spring Harbor Laboratory, Roundtable Study,” Regulation, pp. 20-21, July- 1981). August 1979. 428. Schwartz, J. E., “The Hidden Truth About Reg- 440. Simon, P. J., Reagan in the Workplace: Unravel- ulation, ” Challenge, pp. 54-56, November- ing the Health and Safety Net (Washington, DC: December 1983. Center for Study of Responsive Law, 1983). 429. Scott, R., Muscle and Blood (New York; E.P. 441. Simonds, R. H., “OSHA Compliance: ‘Safety Dutton & Co., Inc., 1974). Is Good Business’, ” Personnel 50(4):30-38, July- 430. Selikoff, I. J., “Constraints in Estimating Oc- August 1973. cupational Cancer Mortality, ” in Quantification 442, Singleton, W. T., “Future Trends in Accident of Occupational Cancer: Banbury Report 9, Research in European Countries, ” in Proceed- R. Peto and M. Schneiderman (eds. ) (Cold ings of International Seminar on Occupational Spring Harbor, NY: Cold Spring Harbor Lab- Accident Research, Stockholm, Sweden, 1983. oratory, 1981). 443. Smith, M. J., Cohen, H. H., Cohen, A., and 431. Selifoff, 1, J., et al., Disability Compensation for Cleveland, R. J., “Characteristics of Successful Asbestos-Associated Disease in the United Safety Programs, ” ]ournal of Safety Research States, contract no. J-9-M-8-0165, report pre- 10(1):5-15, Spring 1978. pared for the Office of the Assistant Secretary 444. Smith, R. S., “Intemporal Changes in Work In- for Policy, Evaluation, and Research, U.S. De- jury Rates, ” IRRA Proceedings, 1972, pp. partment of Labor, Washington, DC, June 1981. 167-74. 432 Settle, R. F., Benefits and Costs of the Federal 445. Smith, R. S., “The Feasibility of an ‘Injury Tax’ Asbestos Standard, contract no, B-9-M-s-0s40, Approach to Occupational Safety, ” Law and report prepared for the Office of the Assistant Contemporary Problems 38(4):730-44, Summer- Secretary for Policy, Evaluation, and Research, Autumn 1974. U.S. Department of Labor, Washington, DC, 446. Smith, R. S., The Occupational Safety and 410 Preventing Illness and Injury in the Workplace

Health Act: Its Goals and Its Achievements (Stanford, CA: Stanford University, April (Washington, DC: American Enterprise Insti- 1982), p. 712.1 OOOB-E. tute, 1976). 458. Steel Tripartite Advisory Committee, “Report 447, Smith, R. S., ‘The Impact of OSHA Inspections to the President by the Steel Tripartite Advisory on Manufacturing Injury Rates, ” Journal of Committee on the United States Steel Industry,” Human Resources 14(2):145-70, Spring 1979. typescript, Sept. 25, 1980. 448. Smith, R. S., “Compensating Wage Differentials 459. Steiger, W., quoted in R. Zeckhauser and A. and Public Policy: A ReviewJ” Zrzdustn”al and Nichols, ‘The Occupational Safety and Health Labor Relations Review 32(3):339-52, April Administration—An Overview, ” in Study on 1979. Federal Regulation, Appendix to vol. VI, U.S. 449. Smith, R. S., “Protecting Workers’ Health and Congress, Senate Committee on Governmental Safety,” in Zrzstead of Regulation, R. W. Poole, Affairs (Washington, DC: U.S. Government Jr. (cd.) (Lexington, MA: Lexington Books, DC Printing Office, December 1978). Heath & Co., 1982. 460. Stender, J., “An OSHA Perspective and Pro- 450. Smith, T. J., Ferrell, W. C,, Varner, M. O., et spective, Labor Law Journal 26:73 (1975) quoted al. “Inhalation Exposure of Cadmium Workers: in Occupational Safety and Health Law, M. Effects of Respirator Usage, ” Arnen’can Zndus- Rothstein, 2d ed. (St. Paul, MN: West Publish- trz”al Hygiene Association Journal 41:624-629, ing Co., 1983). 1980. 461 Stevens, A. D., Director of the Division of 451. Smith, T. J., “Industrial Hygiene, ” in Occupa- Training and Manpower Development, Nation- tional Health: Recognizing and Preventing al Institute for Occupational Safety and Health, Work-Related Disease, B. S. Levy and D. H. U.S. Department of Health and Human Serv- Wegman (eds.) (Boston: Little, Brown & Co., ices, personal communication, 1983. 1983). 462. Stilkind, J., ‘Why Do Accidents Happen?,” Job 452. Snider, C., U.S. Department of Labor, Mine Safety and Health, pp. 4-8, April 1977. Safety and Health Administration, Publication 463, Stokey, E., and Zeckhauser, R., A Primer for Information Office, personal communication, Po2icy Analysis (New York: Norton, 1978). 1983. 464. Strand, S. H., and Johnson, W. G., “An Anal- 453. Snook, S. H., “Workloads,” presentation at ysis of the Safety Incentive Provided by Experi- symposium on “Low Back Pain and Industrial ence Rating under the Workers’ Compensation and Social Disablement, ” organized by the Back Program,” report prepared for the Assistant Sec- Pain Association, held at the Royal College of retary for Policy, Evaluation, and Research, Physicians, London, Oct. 7-8, 1982. U.S. Department of Labor, Washington, DC, 454. Snyder, H. “Vision and the VDU,” Health and March 1980. Ergonomic Considerations of Visual Display 465. Sun, M,, “Reagan Reforms Create Upheaval at Units: Symposium Proceedings (Akron, OH: NIOSH,” Science, 214:166- 168, Oct. 9, 1981. American Industrial Hygiene Association, 466. Surrey, S. S., “Tax Incentives as a Device for 1983). Implementing Government Policy: A Compari- 454a. Society of the Plastics Industry, Inc. v. OSHA, son with Direct Government Expenditures, ” 509 F.2d 1301 (2d Cir., Jan. 31, 1975). Haruard Law Review 83(4):705-38, February 455. Soule, R. D., “Industrial Hygiene Engineering 1970. Controls, ” in Patty’s Industrial Hygiene and 467. Swartzman, D., Liroff, R., and Croke, K., Cost- Toxicology, vol. 111, L. J. Cralley and L. V. Benefit Analysis and Environmental Regulations Cralley (eds.) (New York: John Wiley and Sons, (Washington, DC: The Conservation Founda- Inc., 1979). tion, 1982). 456. Springbom Management Services, Inc., Report 468. Swedish Work Environment Fund, Programme on Workplace Protective Equipment and Cloth- of Activities and Budget: 2981/82-1983/84 ing (Working Paper # 10 of Preventing Illness (Stockholm, Sweden, 1981). and Injury in the Workplace), contract report 470. Tabb, W. K., “Government Regulations: Two prepared for the Office of Technology Assess- Sides to the Story,” Challenge, pp. 40-48, ment, U.S. Congress, Washington, DC, August November-December 1980. 1982. 471. Taylor, quoted in Human Performance Engi- 457. Stanford Research Institute, “Asbestos—Salient neering, R. W. Bailey (Englewood Cliffs, NJ: Statistics” Chemical Economics Handbook Prentice-Hall, Inc., 1982). References 411

472. Teknekron, Inc., “A Report on the National 486. U.S. Congress, “Energy Security Act, ” Public Survey of Workers’ Compensation Insurers, ” in Law 96-294, sec. 131(e). Research Report of the Interdepartmental Work- 487. U.S. Congress, “Regulatory Flexibility Act, ” ers Compensation Task Force, Interdepart- Public Law 96-354, mental Workers’ Compensation Task Force, vol. 488. U.S. Congress, “Paperwork Reduction Act of 7 (Washington, DC: U.S. Government Printing 1980, ” Public Law 96-511. Office, June 1979) 489 U.S. Congress, “Omnibus Budget Reconciliation 473. Teknekron, Inc., “Workers’ Compensation: Act of 1981, ” Public Law 97-35, sec. 1905. Analysis of Insurer Operations, ” in Research Re- 489a U.S. Congress, “Departments of Labor, Health port of the Interdepartmental Workers’ Compen- and Human Services, and Education and Related sation Task Force, Interdepartmental Workers’ Agencies Appropriation Act of 1984,” Public Compensation Task Force, vol. 4 (Washington, Law 98-139. DC: U.S. Government Printing Office, June 490. U.S. Congress, Conference Report No. 91-1765, 1979) Dec. 16, 1970 (Washington, DC: U.S. Govern- 474 Thacker, S. B., Choi, K., and Brachman, P. S., ment Printing Office, 1970). “The Surveillance of Infectious Diseases, ” Jour- 491. U.S. Congress, H.R. 5735, reprinted in U.S. nal of the Amen-can Medical Association 249(9): Congress, House Committee on Education and 1181-1185, Mar. 4, 1983. Labor, Subcommittee on Labor Standards, Oc- 475. Thackrah, C. T., The Effects of the Principal cupational Health Hazards Act of 1982, Hear- Arts, Trades and Professions, and the Civic ings, Mar. 4, Apr, 21, 22, 1982 (Washington, States and Habits of Living on Health and Lon- DC: U.S. Government Printing Office, 1983), gevity (London: Longman’s, 1831). 492. U.S. Congress, Congressional Budget Office, 476. Tichauer, E. R., The Biomedical Basis of Small Issue Industrial Revenue Bonds (Washing- Ergonomics (New York: John Wiley and Sons, ton, DC: U.S. Government Printing Office, Inc., 1978). April 1981). 477 Todradze, C., “Underground Use of Diesel En- 493. U.S. Congress, Congressional Budget Office, gines, ” in Encyclopedia of Occupational Safety The Industrial Policy Debate (Washington, DC: and Health, Luigi Parmeggiani (cd. ) (Geneva, U.S. Government Printing Office, December Switzerland: International Labor Office, 1983). 1983). 478. Tolchin, S. J., “Presidential Power and the Pol- 494, U.S. Congress, General Accounting Office, itics of RARG, ” Regulation, pp. 44-49, July- Slow Progress Likely in the Development of August 1979. Standards for Toxic Substances and Harmful 479. Tolchin, S. J., and Tolchin, M., Dismantling Physical Agents Found in Workplaces, B-163375 America: The Rush to Deregulate (Boston: (Washington, DC: U.S. Government Printing Houghton Mifflin Co., 1983). office, Sept. 28, 1973). 480. Tomenko, M., Manager for Corporate Informa- 495, U.S. Congress, General Accounting Office, tion, Johns-Manville Corp., personal communi- Answers to Questions on the Issuance of An cation, Nov. 30, 1983. Emergency Temporary Standard for Certain 481. Toney, C. R., and Barnhart, W. L., Pe~orm- Chemicals Considered to be Carcinogens, ance Evaluation of Respiratory Protective Equip- MWD-75-33 (Washington, DC: U.S. Govern- ment Used in Paint Spraying Operations, Report ment Printing Office, Jan. 6, 1975). No. NIOSH 76-177 (Cincinnati: National Insti- 496. U.S. Congress, General Accounting Office, tute for Occupational Safety and Health, June Emergency Temporary Standards on Organo- 1976). phosphorous Pesticides, MWD-75-55 (Washing- 482. Trend, “NIOSH—Our National Institute for Oc- ton, DC: U.S. Government Printing Office, Feb. cupational Safety and Health, ” 7(1):3-9. 24, 1975), 483, Tyrer, F. H., and Lee, K., A Synopsis of Oc- 497 U.S. Congress, General Accounting Office, cupational Medicine (Bristol, Great Britain: John Worker Protection Must Be Insured When Wright and Sons, 1979). Employers Request Permission to Deviate From 484. U.S. Chamber of Commerce, Analysis of Work- Safety and Health Standards, IvIWD-76-19 ers Compensation Laws: 1983 (Washington, (Washington, DC: U.S. Government Printing DC: U.S. Chamber of Commerce, 1983). Office, Dec. 31, 1975). 485. U.S. Congress, “Occupational Safety and Health 498 U.S. Congress, General Accounting Office, Fed- Act of 1970,” Public Law 91-596. eral Efforts to Protect the Public From Cancer- 412 . Preventing Illness and Injury in the Workplace

Causing Chemicals Are Not Very Effective, 510. U.S. Congress, General Accounting Office, im- MWD-76-59 (Washington, DC: U.S. Govern- proved Quality, Adequate Resources, and Con- ment Printing Office, June 16, 1976). sistent Oversight Needed lf Regulatory Analy- 499. U.S. Congress, General Accounting Office, Z3et- sis Is To Help Control Costs of Regulations, ter Data orI Severity and Causes of Worker PAD-83-6 (Washington, DC, Nov. 2, 1982). Safety and Health Problems Should be Obtained 511 U.S. Congress, General Accounting Office, Zn- From Workplaces, HRD-76-118 (Washington, formal Settlement of OSHA Citations: Comm- DC: U.S. Government Printing Office, Aug. 12, ents on the Legal Basis and Other Selected 1976). Issues, HRD-85-11 (Washington, DC: October 500. U.S. Congress, General Accounting Office, 26, 1984). States’ Protection of Workers Needs Improve- 512. U.S. Congress, House Committee on Banking, ment, HRD-76-161 (Washington, DC: U.S. Finance, and Urban Affairs, Subcommittee on Government Printing Office, Sept. 9, 1976). Economic Stabilization, Catalog of Federal Loan 501 U.S. Congress, General Accounting Office, De- Guarantee Programs, Committee Print 97-10 lays in Setting Workplace Standards for Can- (Washington, DC: U.S. Government Printing cer Causing and Other Dangerous Substances, Office, January 1982). HRD-77-71 (Washington, DC: U.S. Govern- 513. U.S. Congress, House Committee on Banking, ment Printing Office, May 10, 1977). Finance, and Urban Affairs, Subcommittee on 502, U.S. Congress, General Accounting Office, Spo- Economic Stabilization, Industrial Policy, Hear- radic Workplace Inspections for Lethal and ings, 1983-84, Parts 1-6 (Washington, DC: U.S. Other Serious Health Hazards, HRD-77-143 Government Printing office, 1983, 1984). (Washington, DC: U.S. Government Printing 514. U.S. Congress, House Committee on Banking, Office, Apr. 5, 1978). Finance, and Urban Affairs, Subcommittee on 503. U.S. Congress, General Accounting Office, Economic Stabilization, An Industrial Policy for Health Hazard Evaluation Program Needs Im- America: Is It Needed?, Committee Print 98-2 provement, HRD-78-13 (Washington, DC: U.S. (Washington, DC: U.S. Government Printing Government Printing Office, May 18, 1978). Office, April 1983). 504. U.S. Congress, General Accounting Office, 515. U.S. Congress, House Committee on Banking, Workplace Inspection Program Weak in Detect- Finance, and Urban Affairs, Subcommittee on ing and Correcting Sen”ous Hazards, HRD-78- General Oversight and Renegotiation, Hearing 34 (Washington, DC: U.S. Government Print- on Establishment of a National Development ing Office, May 19, 1978). Bank and Related Matters, Sept. 15, 28, 1982, 505. U.S. Congress, General Accounting Office, La- Serial No. 97-89 (Washington, DC: U.S. Gov- bor Needs to Manage Its Workplace Consulta- ernment Printing Office, 1982). tion Program l?etter, HRD-78-15s (Washington, 516 U.S. Congress, House Committee on Education DC: U.S. Government Printing Office, Dec. 18, and Labor, Report No, 91-1291, July 9, 1970 1978). (Washington, DC: U.S. Government Printing 506. U.S. Congress, General Accounting Office, Office, 1970). Grain Dust Explosions-An Unsolved Problem, 517 U.S. Congress, House Committee on Education HRD-79-1 (Washington, DC: U.S. Government and Labor, Subcommittee on Labor Standards, Printing Office, Mar. 21, 1979). Occupational Health Hazards Compensation 507. U.S. Congress, General Accounting Office, Act of 1982, Hearings, Mar. 4, Apr. 21-22, 1982 How Effective are OSHA’S Complaint Proce- (Washington, DC: U.S. Government Printing dures?, HRD-79-48 (Washington, DC: U.S. Office, 1983). Government Printing Office, Apr. 9, 1979). 517a. U.S. Congress, House Committee on Education 508. U.S. Congress, General Accounting Office, and Labor, Subcommittee on Health and Safety, How Can Workplace Znjuries Be Prevented? The OSHA Oversight–Review of 116 llorrnant Answers May be in OSHA FiZes, HRD-79-43 Rulemaking Projects, Hearing, Oct. 4, 1983 (Washington, DC: U.S. Government Printing (Washington, DC: U.S. Government Printing Office, May 3, 1979). Office, 1984). 509. U.S. Congress, General Accounting Office, “Re- 518. U.S. Congress, House Committee on Energy view of Small Business Administration’s occupa- and Commerce, Subcommittee on Oversight tional safety and health loan program, ” HRD- and Investigations, Capital Formation and in- 79-82, letter to Ray Marshall and A. Vernon dustrial Policy (Washington, DC: U.S. Govern- Weaver from Gregory J. Ahart, June 6, 1979. ment Printing Office, July 1981). References ● 413

519 U.S. Congress, House Committee on Govern- sional Research Service, Estimating the Costs of ment Operations, OMB Interference with Federal Regulation, by J. Allen, Report No, 78- OSHA Rzdernaking, House Report No. 98-583, 205E, Sept. 26, 1978. Dec. 1, 1983 (Washington, DC: U.S. Govern- 527, U.S. Congress, Library of Congress, Congres- ment Printing Office, 1983). sional Research Service, A National Develop- 520. U.S. Congress, House Committee on Govern- ment Bank, by G. R. Meadows, and J. Mitrisin, ment Operations, Subcommittee on Manpower Report No. 79-89E, April 1979. and Housing, Office of Marzagernent and Budget 528. U.S. Congress, Library of Congress, Congres- Control of OSHA Rulemaking, Hearings, Mar. sional Research Service, lndustn”al Policy: An 11, 18, and 19, 1982 (Washington, DC: U.S. Ovemiew, by K. F. Winch, Report No. 80-221E, Government Printing Office, 1983), Dec. 4, 1980.

520a< U.S. Congress, House Committee on Govern- 529. U.S. Congress, Library of Congress, Congres- ment Operations, Subcommittee on Manpower sional Research Service, industrial Policy im- and Housing, OSHA Injury and Illness infor- plicit in Federal Business Credit Programs, N. mation System, Hearing, June 20, 1984 (Wash- A. Noto, Report No. 81-12E, Dec. 31, 1980. ington, DC: U.S. Government Printing Office, 530. U.S. Congress, Library of Congress, Congres- 1984) sional Research Service, OSHA Reform: An 520b. U.S. Congress, House Committee on Government Economic Analysis of the Occupational Safety Operations, Occupational Illness Data Collec- and Health Act, by M, J. Belle, Report No. 81- tion: Fragmented, Unreliable, and Seventy Years 162E, Aug. 6, 1981. behind Communicable Disease Surveillance, 531. U.S. Congress, Library of Congress, Congres- House Report No. 98-1144 (Washington, DC: sional Research Service, Deductions, Credits, U.S. Government Printing Office, 1984). Exemptions and Exclusions in the Federal In- 521. U.S. Congress, House Committee on Interstate come Tax System, by E. S. Burstein, Report No. and Foreign Commerce, Subcommittee on Over- 82-137E, Aug. 12, 1982. sight and Investigations, cost-Benefit Analysis: 532, U.S. Congress, Library of Congress, Congres- Wonder Tool or Mirage?, Committee Print No. sional Research Service, “Products Liability: A 96-lFC 62, December 1980 (Washington, DC: Legal Overview, ” by H. Cohen, Issue Brief No. U.S. Government Printing Office, 1980). IB77021, oct. 7, 1982. 522. U.S. Congress, House Committee on Interstate 533. U.S. Congress, Library of Congress, Congres- and Foreign Commerce, Subcommittee on Over- sional Research Service, “OSHA Reform: Dis- sight and Investigations and Subcommittee on mantlement or Overhaul?” by M. J. Belle, Issue Consumer Protection and Finance, Use of Cost- Brief No. IB81068, Sept. 30, 1982. Benefit Analysis by Regulatory Agencies, Joint 534. U.S. Congress, Library of Congress, Congres- Hearings, July 30, Oct. 10 and 24, 1979 (Wash- sional Research Service, “Industrial Policy in the ington, DC: U.S. Government Printing Office, 98th Congress,” by C. K. Brancato and K. F. 1980). Winch, Issue Brief No. IB83125, Nov. 3, 1983. 523. U.S. Congress, House Committee on the Judi- 535. U.S. Congress, Office of Technology Assess- ciary, Subcommittee on Administrative Law ment, Assessing the Effz”cacy and Safety of Med- and Governmental Relations, Regulatory ical Technologies (Washington, DC: U.S. Gov- Reform Act, Hearings, June 8,9,15, 16,22,23, ernment Printing Office, September 1978). 29, July 13, 14, 27, and 29, 1983 (Washington, 536. U.S. Congress, Office of Technology Assess- DC: U.S. Government Printing Office, 1984). ment, The Direct Use of Coal, OTA-E-86 524. U.S. Congress, House Committee on the Judi- (Washington, DC: U.S. Government Printing ciary, Subcommittee on Administrative Law Office, April 1979). and Governmental Relations, Regulatory 537. U.S. Congress, Office of Technology Assess- Reform Act—Supplement, Part 2 (Washington, ment, An Assessment of Oil Shale Technologies, DC: U.S. Government Printing Office, 1984). OTA-M-118 (Washington, DC: U.S. Govern- 525. U.S. Congress, Joint Economic Committee, in- ment Printing Office, June 1980). dustrial Policy, Economic Growth and the Com- 538. U.S. Congress, Office of Technology Assess- petitiveness of U.S. Industry, Hearings, June 24 ment, Technology and Steel Industry Competi- & 30, July 13, 14, & 20, 1983 (Washington, DC: tiveness, OTA-M-I22 (Washington, DC: U.S. U.S. Government Printing Office, 1983). Government Printing Office, June 1980).

526. U.S. Comzress.V, Librarv, of Corwress,“, Conzres-“ 539, U.S. Congress, Office of Technology Assess- 414 . Preventing Illness and Injury in the Workplace

ment, The Implications of Cost-Effectiveness 549, U.S. Congress, Senate Committee on Finance, Analysis of Medical Technology (Washington, Subcommittee on Savings, Pensions, and Invest- DC: U.S. Government Printing Office, August ment Policy, Promotion of High-Growth Indus- 1980). tries and U.S. Competitiveness, Hearings, Jan. 540. U.S. Congress, Office of Technology Assess- 19-20, 1983 (Washington, DC: U.S. Govern- ment, The Implications of Cost-Effectiveness ment Printing Office, 1983). Analysis of Medical Technology, Background 550. U.S. Congress, Senate Committee on Labor and Paper #I: Methodological Issues and Literature Human Resources, Occupational Safety and Review (Washington, DC: U.S. Government Health Improvements Act of 1980, Hearings, Printing Office, September 1980). Apr. 15, 17, and 25, 1980 (Washington, DC: 541. U.S. Congress, Office of Technology Assess- U.S. Government Printing Office, 1980). ment, Impacts of Applied Genetics: Micro- 551. U.S. Congress, Senate Committee on Labor and Organisms, Plants, and Animals, OTA-HR-132 Public Welfare, Report No. 91-1282, Oct. 6, (Washington, DC: U.S. Government Printing 1970 (Washington, DC: U.S. Government Print- Office, April 1981). ing Office, 1970). 542, U.S. Congress, Office of Technology Assess- 552. U.S. Department of Commerce, “Opportunities ment, Assessment of Technolo~”es for Determin- and Strategies for U.S. Textile Machinery Man- ing Cancer Risks From the Environment, OTA- ufacturers to Improve Their Competitive Posi- H-138 (Washington, DC: U.S. Government tions in Domestic and Foreign Textile Markets, Printing Office, June 1981). 1980-1985, ” September 1980. 543. U.S. Congress, Office of Technology Assess- 553. U.S. Department of Commerce, Bureau of the ment, U.S. Industrial Competitiveness: A Com- Census, County Business Patterns, 1981: United parison of Steel, Electronics, and Automobiles States (Washington, DC: U.S. Government (Washington, DC: U.S. Government Printing Printing Office, 1982). Office, July 1981) 554. U.S. Department of Commerce, Bureau of In- 544. U.S. Congress, Office of Technology Assess- dustrial Economics, U,S. Industrial Outlook: ment, Technology Transfer at the National in- Prospects For Over 300 industries, 25th ed. stitutes of Health, OTA-TM-H-1O (Washington, (Washington, DC: January 1984). DC: U.S. Government Printing Office, March 555. U.S. Department of Health, Education, and 1982). Welfare, “Estimates of the Fraction of Cancer 545. U.S. Congress, Office of Technology Assess- in the United States Related to Occupational ment, Strate~”es for Medica2 Technology Assess- Factors, ” typescript (Bethesda, MD: National ment (Washington, DC: U.S. Government Cancer Institute, National Institute of Environ- Printing Office, September 1982). mental Health Sciences, National Institute for 546. U.S. Congress, Office of Technology Assess- Occupational Safety and Health, Sept. 15, ment, Technologies and Management Strate~”es 1978), reprinted in Quantification of Occupa- for Hazardous Waste Control (Washington, tional Cancer: Banbury Report 9, R. Peto and DC: U.S. Government Printing Office, March M, Schneiderman (eds. ) (Cold Spring Harbor, 1983). NY: Cold Spring Harbor Laboratory, 1981). 547. U.S. Congress, Office of Technology Assess- 556. U.S. Department of Health and Human Serv- ment, The Role of Genetic Testing in the Pre- ices, Promoting Health/Preventing Disease— vention of Occupational Disease, GPO Stock Objectives for the Nation (Washington, DC: No. OTA-BA-194 (Washington, DC: U.S. Gov- Government Printing Office, 1980), ernment Printing Office, April 1983). 557. U.S. Department of Health and Human Serv- 547a. U.S. Congress, Office of Technology Assess- ces, Committee to Coordinate Environmental ment, The Information Content of Premanufac- md Related Programs (CCERP), The Occupa- ture Notices, Background Paper, OTA-BP-H-I7 tional Cancer Risk Subcommittee, On Occupa- (Washington, DC: U.S. Government Printing tional Cancer Estimation: Report of the Oc- Office, April 1983). wpational Cancer Risk Subcommittee of the 548. U.S. Congress, Office of Technology Assess- Department of Health and Human Services ment, Commercial Biotechnology: An Interna- ~ommittee to Coordinate Environmental and tional Analysis, GPO Stock No. OTA-BA-218 ?elated Programs, typescript (Cincinnati, OH: (Washington, DC: U.S. Government Printing WOSH, Nov. 4, 1984) Office, January 1984). 558. J.S. Department of Health and Human Serv- References 415

ices, Public Health Service, Centers for Disease 568. U.S. Department of Health, Education and Wel- Control, “Results of Blood Lead Determinations fare, Public Health Service, Centers for Disease Among Workers Potentially Exposed to Lead— Control, National Institute for Occupational United States,” Morbidity and Mortality Weekly Safety and Health, “Occupational Safety and Report, Apr. 29, 1983. Health Implications of Increased Coal Utiliza- 560, U.S. Department of Health, Education and Wel- tion,” (Rockville, MD, Dec. 1s, 1977). fare, Public Health Service, Centers for Disease 569. U.S. Department of Health, Education and Wel- Control, National Institute for Occupational fare, Public Health Service, Centers for Disease Safety and Health, Criteria for a Recommended Control, National Institute for Occupational Standard: Occupational Exposure to Crystalline Safety and Health, An Evaluation of Occupa- Silica, HEW (NIOSH) Publication No. 75-120, tional Health Hazard Control Technology for 1974. the Foundry Industry, DHEW (NIOSH) Publica- 561, U.S. Department of Health, Education and Wel- tion No, 79-114, 1978. fare, Public Health Service, Centers for Disease 570. U.S. Department of Health, Education, and Control, National Institute for Occupational Welfare, Public Health Service, Centers for Dis- Safety and Health, “Recommended Standard for ease Control, National Institute for Occupa- Occupational Exposure to Vinyl Chloride, ” ex- tional Safety and Health, A Nationwide Sw-uey cerpts from M. Key, Director, NIOSH, memo- of the Occupational Safety and Health Work randum to J. Stender, Assistant Secretary of La- Force, DHEW (NIOSH) Publication No. 78-164 bor, Mar. 11, 1974. (Washington, DC: U.S. Government Printing 562. U.S. Department of Health, Education and Wel- Office, July 1978). fare, Centers for Disease Control, National In- 571. U.S. Department of Health, Education, and stitute for Occupational Safety and Health, En- Welfare, Public Health Service, Centers for Dis- gineering Control and Work Practices Manual, ease Control, National Institute for Occupa- HEW Publication No. (NIOSH) 76-179, 1976. tional Safety and Health and U.S. Department 563, U.S. Department of Health, Education and Wel- of Labor, Occupational Safety and Health Ad- fare, Public Health Service, Centers for Disease ministration, Current Intelligence Bulletin 28: Control, National Institute for Occupational Vinyl Halides–Carcinogenicity, DHEW (NIOSH) Safety and Health, Recommended Industrial Publication No. 79-102, Sept. 21, 1978. Ventilation Guidelines, HEW Publication No. 572, U.S. Department of Health and Human Serv- (NIOSH) 76-172, 1976. ices, Public Health Service, Centers for Disease 564 U.S. Department of Health, Education and Wel- Control, National Institute for Occupational fare, Public Health Service, Centers for Disease Safety and Health, Division of Safety Research, Control, National Institute for Occupational “A Strategy for the Respiratory Protection Re- Safety and Health, National Occupational Haz- search Program, ” typescript, 1980. ard Survey, 1972 - 1974, vol, I, Survey Man- 573. U.S. Department of Health and Human Serv- ual, DHEW (NIOSH) Publication No. 74-127 ices, Public Health Service, Centers for Disease (Cincinnati, OH: NIOSH, 1974). Control, National Institute for Occupational 565 U.S. Department of Health, Education and Wel- Safety and Health, Update of the Literature: fare, Public Health Service, Centers for Disease Crystalline Silica, 1980. Control, National Institute for Occupational 574. U.S. Department of Health and Human Serv- Safety and Health, National Occupational Haz- ices, Public Health Service, Centers for Disease ard Suruey, 1972- 1974, vol. 11, Data Editing Control, National Institute for Occupational and Data Base Development, DHEW (NIOSH) Safety and Health, Manual of Safety and Health Publication No. 77-213 (Cincinnati, OH: Hazards in the School Science Laboratories NIOSH, 1977). (Cincinnati, OH: NIOSH, 1980). 566 U.S. Department of Health, Education and Wel- 574a U.S. Department of Health and Human Serv- fare, Public Health Service, Centers for Disease ices, Public Health Service, Centers for Disease Control, National Institute for Occupational Control, National Institute for Occupational Safety and Health, National Occupational Haz- Safety and Health, Summary of NIOSH Recom- ard Survey, 1972-2974, vol. 111, Survey Anal- mendations for Occupational Health Standards ysis and Supplemental Tables, DHEW (NIOSH) (Cincinnati, OH: NIOSH, November 1980), Publication No. 78-114 (Cincinnati, OH: 575 U.S. Department of Health and Human Serv- NIOSH, 1977). ices, Public Health Service, Centers for Disease 416 . Preventing Illness and Injury in the Workplace

Control, National Institute for Occupational 583. U.S. Department of Health and Human Serv- Safety and Health, Occupational Hazard Assess- ices, Public Health Service, Centers for Disease ment: Coal Liquefaction, Vol. Ii—Assessment, Control, National Institute for Occupational DHHS (NIOSH) Publication No. 81-132 (Wash- Safety and Health, “Leading Work-related Dis- ington, DC: U.S. Government Printing Office, eases and In juries,” Morbidity and Mortality 1981). Weekly Report, 32:24-26, 32, Jan. 21, 1983. 576. U.S. Department of Health and Human Serv- 584, U.S. Department of Health and Human Serv- ices, Public Health Service, Centers for Disease ices, Public Health Service, Centers for Disease Control, National Institute for Occupational Control, National Institute for Occupational Safety and Health, Symposium Proceedings: Safety and Health, NZOSH Program Plan by Control Technology in the Plastics and Resins Program Areas: Fiscal Year 1983, GPO stock Jndustry, DHHS (NIOSH) 81-107 (Washington, No. 661-089 (Washington, DC: U.S. Gover- DC: U.S. Government Printing Office, January nment Printing Office, February 1983). 1981), 585. U.S. Department of Health and Human Serv- 577. U.S. Department of Health and Human Serv- ices, Public Health Service, Centers for Disease ices, Public Health Service, Center for Disease Control, National Institute for Occupational Control, National Institute for Occupational Safety and Health, “Comments to OSHA for Safety and Health, Proceedings of Symposium Docket H-160: Health Standards: Methods of on Occupation Health Hazard Control Technol- Compliance,” typescript, June 1983. ogy in the Foundry and Secondary Non-ferrous 586. U.S. Department of Health and Human Serv- Smelting Industries, DHHS (NIOSH) Publica- ices, Public Health Service, Centers for Disease tion No. 81-114, August 1981. Control, National Institute for Occupational 579. U.S. Department of Health and Human Serv- Safety and Health, Registry of ‘Toxic Effects of ices, Public Health Service, Centers for Disease Chemical Substances: 1981-82 Edition, DHHS Control, National Institute for Occupational (NIOSH) Publication No. 83-107 (Cincinnati, Safety and Health, Mission Statement: National OH: NIOSH, June 1983). Institute for Occupational Safety and Health 587. U.S. Department of Health and Human Serv- (Atlanta, GA: Centers for Disease Control, ices, Public Health Service, Centers for Disease 1982). Control, National Institute for Occupational 580. U.S. Department of Health and Human Serv- Safety and Health, Engineering Control of Oc- ices, Public Health Service, Centers for Disease cupational Safety and Health Hazards: Recom- Control, National Institute for Occupational mendations for Improving Engz”neen”ng Practice, Safety and Health, “Respirator Information Education, and Research, Summary Report of Notice on MSA Powered Air Purifying Respi- the Engineering Control Technology Workshop rator, Mine Safety Appliance Co., Pittsburgh, Technical Panel (Cincinnati, OH: NIOSH, July PA, Model Numbers: 463354, 466607, 466608, 1983). Approval Number TC-21C-M6° (Atlanta: Na- 588. U.S. Department of Health and Human Serv- tional Institute for Occupational Safety and ices, Public Health Service, Centers for Disease Health, Nov. 15, 1982). Control, National Institute for Occupational 581 U.S. Department of Health and Human Serv- Safety and Health, NIOSH Report on Occupa- ices, Public Health Service, Centers for Disease tional Safety and Health for Fiscal Year 1982 Control, National Institute for Occupational Under Public Law 91-596, GPO stock No. 659- Safety and Health, Report on Occupational 096-1026 (Washington, DC: U.S. Government Safety and Health for FY 1982–Under Public Printing Office, September 1983). Law 91-596 (Washington, DC: Government 590. U.S. Department of Health and Human Serv- Printing Office, 1983). ices, Public Health Service, Centers for Disease 582. U.S. Department of Health and Human Serv- Control, National Institute for Occupational ices, Public Health Service, Centers for Disease Safety and Health, NZOSH Program Plan by Control, National Institute for Occupational Program Areas: Fiscal Years 1984-89, DHHS Safety and Health, Division of Physical Sciences (NIOSH) Publication No. 84-107, February and Engineering, Overview-The Division of 1984. Physical Sciences and Engineering, NIOSH in- 591. U.S. Department of Health and Human Serv- ternal document (Cincinnati, OH: 1983). ices, Public Health Service, National Center for References ● 417

Health Statistics, Current Estimates From the 605. U.S. Department of Labor, Bureau of Labor Sta- National Health Intemiew Survey, various tistics, Back lnjun”es Associated with Lifting, years. Bulletin No. 2144 (Washington, DC: U.S. Gov- 592. U.S. Department of Health, Education and Wel- ernment Printing Office, August 1982). fare, Public Health Sewice, National Center for 606. U.S. Department of Labor, Bureau of Labor Sta- Health Statistics, PersonsInjured and Disability tistics, Occupational injuries and Illnesses in the Days by Detailed Type and Class of Accident, United States by Industry, 1982, Bulletin No. United States, 1971-72, January 1976. 2164 (Washington, DC: U.S. Government Print- s93. U.S. Department of Health and Human Serv- ing Office, January 1983). ices, Public Health Service, National Center for 607. U.S. Department of Labor, Bureau of Labor Sta- Health Statistics, Vital Statistics of the U. S., tistics, “Occupational Injuries and Illnesses in 1977. 1982, ” typescript, press release, Nov. 4, 1983. 595. U.S. Department of Health and Human Serv- 608. U.S. Department of Labor, Bureau of Labor Sta- ices, Public Health Service, National Toxicology tistics, Occupational injuries and Illnesses in the Program, National Toxicology Program; Fiscal United States by Industry, 1982, Bulletin No. Year 1984 Annual Plan, NTP-84-023 (Research 2196 (Washington, DC: U.S. Government Print- Triangle Park, NC: National Toxicology Pro- ing Office, April 1984). gram, February 1984). 609. U.S. Department of Labor, Mine Safety and 596. U.S. Department of Labor, Assistant Secretary Health Administration, “Respiratory Protective for Policy, Evaluation, and Research (ASPER), Devices,” Code of Federal Regulations, vol. 30, An Interim Report to Congress on Occupational part 11. Diseases, Washington, DC, June 1980. 610. U.S. Department of Labor, Mine Safety and 597. U.S. Department of Labor, Bureau of Labor Sta- Health Administration, Mine Safety and Health tistics, “Comparison of 1974 NSC and BLS Advisory Committee, Respirator Research Sub- Death Estimates, ” typescript, n.d. committee, “Respirator Certification, ” type- 598. U.S. Department of Labor, Bureau of Labor Sta- script, Jan. 27, 1982. tistics, Accidents involving Eye injuries, Report 611. U.S. Department of Labor, Occupational Safety No. 597 (Washington, DC: Government Print- and Health Administration, “Air Contami- ing Office, 1980). nants, ” Code of Federal Regulations, vol. 29, s99. U.S. Department of Labor, Bureau of Labor Sta- sec. 1910.1000(e). tistics, Accidents Involving Face Injuries, Report 612. U.S. Department of Labor, Occupational Safety No. 604 (Washington, DC: Government Print- and Health Administration, “Occupational ing Office, 1980). Noise Exposure, ” Code of Federal Regulations, 600. U.S. Department of Labor, Bureau of Labor Sta- vol. 29, sec. 1910.95(b)(l). tistics, Accidents Involving Head Injuries, Re- 613. U.S. Department of Labor, Occupational Safety port No. 605 (Washington, DC: Government and Health Administration. “Respiratory Pro- Printing Office, 1980), tection,” Code of Federal Regulations, vol. 29, 601. U.S. Department of Labor, Bureau of Labor Sta- sec. 1910.134(a)(l). tistics, Accidents Involving Foot injuries, Report 614. U.S. Department of Labor, Occupational Safety No. 626 (Washington, DC: Government Print- and Health Administration, Division of Statis- ing Office, January 1981). tical Studies and Analysis, “Occupational Fatali- 602. U.S. Department of Labor, Bureau of Labor Sta- ties Among Workman’s Compensation Claims, tistics, Handbook of Labor Statistics, 1980. 1972, ” typescript, n.d. 603. U.S. Department of Labor, Bureau of Labor Sta- 615. U.S. Department of Labor, Occupational Safety tistics, Work-related Hand Injuries and Upper and Health Administration, “Emergency Tem- Extremity Amputations, Bulletin No. 2100 porary Standard for Exposure to Vinyl Chlo- (Washington, DC: Government Printing Office, ride, ” Federal Register 39:12342, Apr. 5, 1974. 1982). 616. U.S. Department of Labor, Occupational Safety 604. U.S. Department of Labor, Bureau of Labor Sta- and Health Administration, “Vinyl Chloride— tistics, Occupational Injuries and Illnesses in the Proposed Standard, ” Federal Register 39:16896, United States by Industry, 1980, Bulletin No. May 10, 1974. 2130 (Washington, DC: U.S. Government Print- 617. U.S. Department of Labor, Occupational Safety ing office, April 1982). and Health Administration, “Standard for Ex- 418 . Preventing Illness and Injury in the Workplace

posure to Vinyl Chloride, ” Federal Register 630. U.S. Department of Labor, Occupational Safety 39:35890, Oct. 4, 1974. and Health Administration, press release, Aug. 618. U.S. Department of Labor, Occupational Safety 25, 1982. and Health Administration, “Recordkeeping Re- 631. U.S. Department of Labor, Occupational Safety quirements under the Occupational Safety and and Health Administration, Regulatory Impact Health Act of 1970,” revised, 1978. and Regulatory Flexibility Assessment of the 619, U.S. Department of Labor, Occupational Safety Proposed Standard on Servicing Single-Piece and Health Administration, “A Comparison of Rim Wheels, September 1982. Occupational Injury Rates, ” typescript, Febru- 632. U.S. Department of Labor, Occupational Safety ary 1978. and Health Administration, press releases: Sept. 620, U.S. Department of Labor, Occupational Safety 24 and 30, 1982. and Health Administration, “Exposure to Coke 633. U.S. Department of Labor, Occupational Safety Oven Emissions,” Federal Register 41:46742- and Health Administration, “Scheduling System 46790, Oct. 22, 1976. for Programmed Inspections, ” Instruction No. 621, U.S. Department of Labor, Occupational Safety CPL 2.25C, Oct. 1, 1982. and Health Administration, Benefi”ts Analysis 634. U.S. Department of Labor, Occupational Safety for the Proposed Federal Standard Regulating and Health Administration, “Educational/Scien- Conveyors, 1980. tific Diving, ” Federal Register 47:s33s7, Nov. 622. U.S. Department of Labor, Occupational Safety 26, 1982. and Health Administration, Final Regulatory 635< U.S. Department of Labor, Occupational Safety Analysis of the Hearing Conservation Amend- and Health Administration, Field Operations ment, January 1981. Manual, revised 1983. 623. U.S. Department of Labor, Occupational Safety 636. U.S. Department of Labor, Occupational Safety and Health Administration, Occupational and Health Administration, OSHA: Meeting its Safety and Health Statistics of the Federal Gov- Mandate, 1983. ernment, CY 1979, Publication No. (OSHA) 637. U.S. Department of Labor, Occupational Safety 2066, January 1981. and Health Administration, “Response to Sub- 624. U.S. Department of Labor, Occupational Safety committee Questionnaire, ” in Regulatory and Health Administration, General Industry Reform Act—Supplement, Hearings Before the OSHA Safety and Health Standards, Publica- Subcommittee on Administrative Law and Gov- tion No. (OSHA) 2206, Revised, June 1981. ernmental Relations, House Committee on the 625, U.S. Department of Labor, Occupational Safety Judiciary, U.S. Congress, Serial No. 25, Part 2, and Health Administration, “Occupational Noise 1983. Exposure; Hearing Conservation Amendment; 638. U.S. Department of Labor, Occupational Safety Final Rule” Federal Register 46:4078-4179, Jan. and Health Administration, “Occupational Ex- 16, 1981. posure to Inorganic Arsenic, Supplemental 626. U.S. Department of Labor, Occupational Safety Statement of Reasons,” Federal Register 48:1865, and Health Administration, Regulatory Impact Jan. 14, 1983. and Regulatory Flexibility Analysis for the Lead 639. U.S. Department of Labor, Occupational Safety Standard/Revised Supplemental Statement of and Health Administration, “Health Standards; Reasons and Amendment of Final Rule, October Methods of Compliance, ” Federal Register 1981. 48:7473-7476, Feb. 14, 1983.

627. U.S. Department of Labor, Occupational Safety 640< U.S. Department of Labor, Occupational Safety and Health Administration, “scheduling System and Health Administration, “Occupational for Programmed Inspections, ” Instruction No. Noise Exposure; Hearing Conservation Amend- CPL 2.25B, Oct. 1, 1981. ment; Final Rule, ” Federal Register 48:9738- 628. U.S. Department of Labor, Occupational Safety 9758, Mar. 8, 1983. md Health Administration, Draft Regulatory 641, U.S. Department of Labor, Occupational Safety Impact Analysis and Regulatory Flexibility and Health Administration, “Occupational Ex- Analysis of the Hazard Communication Pro- posure to Cotton Dust, ” Federal Register posal, March 1982. 48:26962, June 10, 1983. 629 U.S. Department of Labor, Occupational Safety 642. U.S. Department of Labor, Occupational Safety and Health Administration, “Employer Abate- and Health Administration, “Occupational Ex- ment Assistance, ” Instruction No. CPL 2.53, posure to Ethylene Dibromide, ” Federal Regis- July 15, 1982. ter 48:45956- 46003, oct. 7, 1983. —

References ● 419

643. U.S. Department of Labor, Occupational Safety nancial Incentives (Santa Monica, CA: Rand, and Health Administration, “OSHA Head 1982). Thorne G. Auchter Comments on 1982 BLS 657. Victor, R. B., Cohen, L. R., and Phelps, C. E., Data on Job-Related Injuries and Illnesses, ” press Workers’ Compensation and Workplace Safety: release, Nov. 4, 1983. Some Lessons From Economic Theory (Santa 644. U.S. Department of Labor, Occupational Safety Monica, CA: Rand, 1982). and Health Administration, “OSHA Lowers 658. Viscusi, W. K., “The Impact of Occupational Targeting Figure for Safety Inspections, ” press Safety and Health Regulation, ” Bell Journal of release, Nov. 8, 1983. Economics 10(1):117-40, spring 1979. 64.5. U.S. Department of Labor, Occupational Safety 658a, Viscusi, W, K., Employment Hazards (Cam- and Health Administration, “OSHA Head Re- bridge, MA: Harvard University Press, 1980). ports Greatly Improved Worker Protection 659. Viscusi, W. K., “Presidential Oversight: Con- Before Congressional Subcommittee, ” press re- trolling the Regulators, ” Journal of Policy Anal- lease, Nov. 9, 1983. ysis and Management 2(2):157-73, 1983. 646. U.S. Department of Labor, Occupational Safety 660, Wallick, F., The American Worker: An En- and Health Administration, “Guidelines for dangered Species (New York: Ballantine Books, Noise Enforcement, ” Instruction No. CPL 2- 1972), republished with additional material 2.35A, Dec. 19, 1983. under the title Don’t Let Your Job Kill You 647. U.S. Department of Labor, Occupational Safety (Washington, DC: Progressive Press, 1984). and Health Administration, “Occupational Ex- 661. Wallingford, K., Tight Building Syndrome— posure to Asbestos, ” Federal Register 49:14116- Appraches to Indoor Air Quality in Non-Indus- 14145, Apr. 10, 1984. tn”al Workplaces, Presentation to Conference of 648. U.S. Department of Labor, Occupational Safety State Sanitary Engineers, 1983 Annual Meeting, and Health Administration, “Consultation Bloomington, MN, May 11, 1983. Agreements, “ Federal Register 49:25082-25100, 662, Walters, N. K., quoted in “Ned Walters hangs June 19, 1984. up his hard hat, ” Dupont News, p. 6, March 649. U.S. Department of Labor, Occupational Safety 1983. and Health Administration, Federal Register 663. Warner, K. E., and Luce, B. R., Cost-Benefit 49:25734-25809, June 22, 1984. and Cost-Effectiveness Analysis in Health Care 650. U.S. Department of Labor, Office of Informa- (Ann Arbor, MI: Health Administration Press, tion, “Secretary Donovan Comments on 1981 1982). BLS Data on Job-Related Injuries and Illnesses, ” 664, Warner, K. E., Technology and Handicapped press re]ease, Nov. 17, 1982. People, Background Paper #l: Mandatory 651. U.S. Department of Treasury, Office of Tax Passive Restraint Systems in Automobiles, re- Analysis, The Use of Tar Subsidies for the Cost port prepared for the Office of Technology As- of Compliance with Safety and Health Regula- sessment, U.S. Congress (Washington, DC: U.S. tions (Washington, DC, 1981) Government Printing Office, September 1982). 652. U.S. Environmental Protection Agency, Office 665. Weeks, J. L., and Fox, M., “Fatality Rates and of Research and Development, Office of Ex- Regulatory Policies in Bituminous Coal Mining, ploratory Research, Office of Strategic Assess- United States, 1959-81, ” American Journal of ment and Special Studies, Anticipating U.S. Public Health 73(11):1278-1280, November industrial Changes—implications For the Enui- 1983. ronrnent (Washington, DC: in press). 666. Weinstein, J. W., “Big Business and the Origins 653. U.S. Synthetic Fuels Corporation, “Environ- of Workmen’s Compensation, ” Labor History mental Monitoring Plan Guidelines, ” Feder~l 8(2):156-74, spring 1967. Register 48:46676, Oct. 13, 1983. 667. Weinstein, J. W., The Corporate Ideal in the 654. United Steelworkers of America u. Marshall, Liberal State; 1900-1928 (Boston: Beacon Press, 647 F.2d 1189 (D.C, Cir., Aug. 15, 1980). 1968). 655. Van Cott, H. P., and KinKade, R. G., (eds. ), 668. Whiting, B., Deputy Assistant Secretary for Oc- Human Engineen”ng Guide to Equipment Design cupational Safety and Health, testimony before (Washington, DC: U.S. Government Printing the Senate Committee on Labor and Human Re- Office, 1972). sources, U.S. Congress, Mar. 21, 1980. 656, Victor, R. B., Workers’ Compensation and 669. Wiedenbaum, M. L., and DeFina, R., The Cost Workplace Safety: The Nature of Employer Fi- of Federal Regulation of Economic Activity, Re- 420 ● Preventing Illness and Injury in the Workplace

prim No. 88 (Washington, DC: American Enter- 679 Wood, W. C., “Putting a Price on Radiation, ” prise Institute, May 1978). [oumal of Policy Analysis and Management 670. Wiedenbaum, M. L., ‘The Costs of Government 2(2):291-95, winter 1983. Regulation of Business,” a study prepared for 680. Worrall, J. D., “An Analysis of Some Aspects the Subcommittee on Economic Growth and of the U.S. Department of Labor’s Draft Interim Stabilization, Joint Economic Committee, U.S. Report to the Congress on Occupational Dis- Congress (Washington, DC: U.S. Government eases, ” April 1980, in “Statement of the Alliance Printing Office, Apr. 10, 1978). of American Insurers Before the Senate Labor 671. Wiener, F., ‘The Vinyl Chloride Threat and the and Human Resources Committee, ” A. Maison- Response of Government, Industry and Labor” pierre, Aug. 26, 1980. in Protecting People at Work, J. MacLaury (cd. ) 681. Worrall, J. D., “Compensation Costs, Injury (Washington, DC: U.S. Government Printing Rates, and the Labor Market,” in Safety and the Office, 1980). Work Force, J. D. Worrall (cd.) (Ithaca, NY: ILR 672. Wilcher, F. E. Jr., President, Safety Equipment Press, 1983). Institute, personal communication, Aug. 23, 682. Worrall, John D., (cd. ) Safety and the Work 1983. Force: Incentives and Disincentives in Workers’ 673. Wilmes, D. P. “Respirator Usage: A Rational Compensation (Ithaca, NY: ILR Press, 1983). Program, ” Hazardous Materials Management 683. Young, J. L., Percy, C. L., Asire, A. J., et al., loumal 2(2):18-26, January-February 1981. Cancer Incidence and Mortality in the United 674. Wilmes, D. P., “Respirator Usage: A Rational States, 1973-77, National Cancer Institute Program, Part II,” Hazardous Materials Man- Monograph No. 57 (Bethesda, MD: National agement Journal 2(3):28-32, March-April 1981. Cancer Institute, 1981). 674a. Wilmes, D. P., personal communication, De- 684. Zabetakis, M., Safety Manual No. 4: Accident cember 1983. Prevention, Mine Safety and Health Adminis- 675. Wingate, R., comments on draft report, letters, tration (Washington, DC: 1975). Nov. 4, 1983, Jan. 11 and 27, 1984. 685. Zeckhauser, R., and Nichols, A., ‘The Occupa- 676. Wischer, F. E., Jr., “Statement of the Safety tional Safety and Health Administration—An Equipment Institute, Presented to the Respira- Overview,” in Study on Federal Regulation, Ap- tor Research Subcommittee of the Mine Health pendix to Volume VI, U.S. Congress, Senate Research Advisory Committee,” Jan. 4-5, 1982. Committee on Governmental Affairs (Washing- 677. Witt, M., and Early, S., “The Worker as Safety ton, DC: U.S. Government Printing Office, De- Inspector, ” Working Papers, pp. 21-29, cember 1978). September-October 1980. 686. Zohar, D., Cohen, A., and Azar, N., “Pro- 678. Wolfe, S. M., and Abrams, L., 1983 Suney of moting Increased Use of Ear Protectors in Noise Fourteen Union Occupational Safety and Health through Informational Feedback, ” Human Fac- Programs (Washington, DC: Public Citizen, tors 22(1):69-79, 1980. Health Research Group, January 1984). Index Index

American Conference of Governmental Industrial Educational Resource Centers (ERC), 10, 195 Hygienists, 9, 86, 206, 257-261 Environmental Protection Agency (EPA), 9, 63, American Industrial Hygiene Association, 9 157, 280 American National Standards Institute (ANSI), 9, ergonomics and human factors, 123 71, 206, 300 applying ergonomics to VDT design, 135 American Occupational Medical Association, 205 classification of ergonomics, 124 assessment of OSHA and NIOSH activities, 257 prevention of musculoskeletal injuries, 127 comparison of recommendations and standards, back disorders, 132-135 257-262 carpal tunnel syndrome (CTS), 127-132 impacts of NIOSH, 269 impacts of OSHA, 262 Ford, President Gerald, 280

B. F. Goodrich, 81 General Accounting Office, 220, 238, 287, 331 Bureau of National Affairs, 314 governmental activities concerning worker health and safety, 219-254 Carter, President Jimmy, 280 current Federal/State framework, 219 Centers for Disease Control, 22o assistant Secretaries and Directors, 221 Congress: Federal spending, 221 House Committee on Energy and Commerce, 4 separation of research and regulation, 22o Senate Committee on Labor and Human National Institute for Occupational Safety and Resources, 4 Health, 242-253 Council on Wage and Price Stability, 280, 282, development of controls, 247 283 dissemination, 249 identification, 243 data on occupational injuries and illnesses, 29-38 priorities, 251 occupational illnesses, 37 Occupational Health and Safety Administration, occupational injuries, 30 224-242 sources of information, 29 enforcement, 232 decisionmaking for occupational safety and health, public education and service, 237 275-294 standard-setting, 224 general issues, 275 State programs, 239 OSHA decisionmaking on standards, 278 uses and limits of economic analysis, 289 health hazard identification, 41-64 Department of the Interior occupational disease, 42 Bureau of Mines, 145 cancer, 45 Department of Labor musculoskeletal disorders, 45 Bureau of Labor Statistics (BLS), 4, 5, 17, 29, necrologic disorders, 49 30, 36, 132, 167 reproductive disorders, 48 Mine Safety and Health Administration, 219 respiratory disorders, 43 dissemination of health and safety information, 11 skin disorders, 49 Du Pent Co., 117, 155 known and unknown health hazards, 52 identified hazards, 52 economic incentives, reindustrialization, and methods for detection of present, unidentified Federal assistance for occupational safety hazards, 54 and health, 327 new hazards, 62 economic incentives and financial assistance, 327 hierarchy of controls, 175-185 injury/exposure taxes, 327 advantages and disadvantages of, 178 tax programs and financial assistance, 329 advantages of engineering controls, 179 Federal aid for research and information problems with engineering controls, 181 dissemination, 338 problems with personal protective equipment, reindustrialization and occupational safety and 179 health, 332 description of, 175

423 424 . Preventing Illness and Injury in the Workplace

consensus standards, 176 Occupational Safety and Health Administration controls for safety hazards, 177 (OSHA), 6, 10, 11, 13, 14, 19, 20, 22, 23, Government standards, 177 25, 29, 34, 35, 45, 86, 90, 92, 93, 104, 110, views of health professionals, 175 145, 189, 197, 219, 220, 224-242, 257, 271, OSHA’S policy, 182 275, 278-288, 298, 359-380 actions by the current administration, 183 enforcement, 232 inspection and enforcement, 14 identification of occupational hazards, 6 regulation, 13 incentives, imperatives, and the decision to standard-setting, 224, 281 control, 297 occupationally associated deaths, injuries, and assessment of existing incentives, 297 illnesses, 4-6 Government regulation, 320 Office of Management and Budget (OMB), 13, labor market forces, collective bargaining, and 280, 281, 287 worker’s rights, 313 options for controlling workplace hazards, 16-26 provision of information, 301 creation of Occupational Safety and Health tort liability, 309 Fund, 24 voluntary efforts, 297 data and hazard identification, 16 workers’ compensation and insurance, 302 education, training, and information Internal Revenue Service (IRS), 16, 331 dissemination, 19 improved control technologies, 18 legislation: incentives and imperatives, 21 Clean Air Act, 237 needs of small businesses, 25 Coal Mine Health and Safety Act, 212, 219 role of OSHA, 22 Federal Mine Safety and Health Act of 1977, 145, 212 personal protective equipment, 143-172 Flood Control Act of 1936, 277 hearing protectors, 154-158 Metal and Non-Metallic Mine Safety Act of acceptance of, 156 1966, 219 conservation programs, 155 Mine Safety and Health Act of 1977, 219 field testing, 158 Noise Control Act of 1972, 157 noise reduction ratings, 157 Occupational Safety and Health (OSH) Act, 4, respirators, 143-154 12, 23, 29, 36, 86, 145, 177, 195, 213, 219, deficiencies in programs, 152 220, 224, 239, 242, 276, 278, 284, 314 dust masks, 148 Toxic Substances Control Act (TSCA), 58, 63, field testing, 149 64 fit testing, 145 Walsh-Healey Act, 86, 213 Los Alamos Scientific Laboratory studies, 147 NIOSH certification, 1s3 National Cancer Institute, 238 third-party testing, 1s4 National Center for Health Statistics (NCHS), 5, personal protective equipment, other types of, 16, 30 159 National Death Index (NDI), 16, 58 NIOSH tests, 159 National Electronic Injury Surveillance System protective eyewear, 163 (NEISS), 32 tests of gloves against chemical hazards, 166 National Highway Traffic Safety Administration, third-party laboratory testing, 170 73 preventing work-related injury and illness in the National Institute for Occupational Safety and future, 345 Health (NIOSH), 6, 9, 11, 16, 18, 21, 22, opportunities for prevention, 346 25, 30, 32, 38, 43, 55, 64, 73, 81, 83, 90, private and public activities, history of, 205 104, 110, 134, 153, 158, 159, 161, 189, 197, government efforts, 207 200, 219, 220, 242-253, 257-271, 359-380 early Federal Government programs, 210 National Library of Medicine, 11 Federal legislation and regulatory programs, National Occupational Exposure Survey, 97 212 National Safety Council (NSC), 4, 29, 69, 205, 298 Federal research and assistance in occupational North Carolina, 86 safety and health, 211 —

Index . 425

State health and safety programs, 209 control at the worker, 85 workers’ compensation, 207 controlling dispersion, 82 voluntary efforts, 205 control technology usage in the United States, American Conference of Governmental 97 Industrial Hygienists, 206 controlling worker exposure to cotton dust, American National Standards Institute, 206 85-89 American Occupational Medical Association, controlling worker exposure to lead, 92 205 controlling worker exposure to silica dust, 89 National Safety Council, 205 technologies for controlling work-related injury, Public Health Service, 211 103-119 Du Pent injury control program, 117 Reagan, President Ronald, 281 fire and explosion prevention, 113 reindustrializa tion and workplace health and injury control programs in industry, 116 safety, 15 preventing construction-related injuries, 106 preventing manufacturing-related injuries, 108 Safety Equipment Institute, 154 preventing motor vehicle-related injuries and safety haza;d identification, 67-74 fatalities, 106 injury causation, 69-74 technologies to control hazards, 7 supplemental information on OSHA and NIOSH, training and education, 10, 189-202 359-380 computer networks, 200 engineers and managers, 199 technologies for controlling work-re ated illness, occupational safety and health professionals, 195 77-99 worker training and education, 189 control systems, 78-85 control at the source, 80 U.S. Supreme Court, 86, 284