Risk Management Peer-Reviewed

Safety Expectations Finding a Common Denominator

By Vladimir Ivensky

he spectrum of achieved levels of OSH ership might treat all incidents the same may be a care and results (occupational mortal- presumption that all OSH risks must be controlled Tity, injury and illness rates) varies between equally no matter their severity potential in order countries, industries and to achieve the zero harm expectations that are em- companies with an overall bedded in many OSH programs. IN BRIEF tendency for improvement. This article discusses OSH •Over the past decade, occupational However, the occupation- techniques as applied to zero harm programs, injury rate reductions have primar- al injury rate reductions and describes ways to bring OSH expecta- ily affected low-severity incidents, achieved in the past decade tions and strategies to a common denominator while serious injury and fatality rates and earlier are affecting among all project stakeholders to better pre- decline more slowly. more the low severity end vent serious incidents. •One reason may be that operational of the incidents spectrum, leadership presumes OSH risks must while serious injury and fa- Safety Expectations: Parties Involved be controlled equally regardless of tality (SIF) incident rates are In today’s marketplace, companies’ expectations severity potential. declining at a slower pace of employee and subcontractor safety performance •This article discusses OSH risk as- (Mangan, 2015; Manuele, are driven by both ethical obligations and market sessment techniques as applied to 2003). for competitiveness, as safety perfor- zero harm programs, and ways to bring The strategy to accelerate mance has become an important differentiator in OSH expectations and strategies to a the decline of serious and many industry sectors. common denominator among various fatal occupational injuries For the purpose of this discussion, the following parties involved in a project. is the subject of ongoing parties are considered as having some vested inter- debate in the OSH profes- est in OSH, and having specific expectations from sion. According to Mangan OSH programs: (2015), the discrepancy between the reduction •project owner/client; achieved with SIF rates versus that achieved with •prime contractor management; relatively minor cases exists in part because com- •prime contractor employees; panies’ operational leadership and safety practi- •subcontractor management; tioners treat all incidents the same, while in reality •subcontractor employees; only about 20% of incidents have the potential to •public and regulators. become an SIF. One reason that operational - The safety-sensitive client expects excellent safe- ty performance at its project sites, relying mostly on its prime contractor (and its own enforcement Vladimir Ivensky, CSP, CIH, has more than 25 years’ experience in OSH. of safety programs). For the prime contractor, ex- He is a corporate vice president of safety, health and environment for a global cellent safety performance is, therefore, a project multidisciplinary consulting and construction management firm. Ivensky holds a deliverable. The prime contractor would commu- master’s degree and doctorate (equivalents) in Occupational Safety and Health nicate the client’s and its own safety expectations and Environmental Protection from the Rostov Civil Engineering Institute in the former Soviet Union. and requirements to its subcontractors and expect

excellent safety performance from its own employ- ©iStockphoto.com/Shade O N 38 ProfessionalSafety july 2016 www.asse.org ees and the subcontractors’ employees. Possible OSH expectations are summarized in Table 1. Table 1 Employers’ (and clients’) interest in managing the outcome of relatively minor incidents within Company & Individual OSHA’s definition of or their willingness OSH Expectations to apply modified/restricted work programs to avoid lost-time incident cases may conflict with the Company Individual injured person’s interests (when overzealous case •Low (or zero) OSHA •Safe work environment management approaches are applied). The associ- recordable case rate, discounts •No work -­‐related OSH ated efforts may also be time consuming, and the effort may be better spent on serious injury incident on workers’ compensation incidents (whether OSHA prevention. It may be in the OSH industry’s inter- insurance, no regulatory recordable or not) that would est to overcome that controversy, potentially by citations, no litigation impact personal health modifying OSHA’s recordkeeping system toward •Enhanced competitiveness beyond the day of the emphasizing high-potential incidents (no matter (reputation) incident the outcome) and de-emphasizing low-potential •Low (or zero) financial loss •Zero personal financial loss and low-severity cases as a competitive differen- tiator by utilizing, for example, approaches similar to those used in the U.K.’s Reporting of Injuries, 2) The zero harm goal is unachievable and Diseases and Dangerous Occurrences Regulations to misallocating safety resources to multiple minor (RIDDOR) 2013 (HSE, 2015b; Ivensky, 2015). , increasing the risk of missing the major hazards. An example of this point of view: Ultimate Safety Expectation: Zero Harm We are kidding ourselves if we think we can The popular zero harm OSH philosophy sug- prevent all personal damage; why have a goal gests the ultimate goal of OSH: zero occupational that is unrealistic? The reality is that while we injuries and illnesses for all workers and for any- are spending time, effort and money on minor one affected by a company’s business. Many com- damage we are misallocating resources that are panies and groups have adopted this philosophy better directed at serious personal damage. Re- in their safety (and quality) programs. Zero harm, alistically there are limitations to what [company] however, has not been achieved consistently any- resources can be applied; let’s get the biggest where and the ways to achieve it, the possibility bang for our buck. of achieving it, or even an ultimate benefit of zero While the first point of view is widely represented in harm programs for the prevention of SIFs are the many companies’ zero harm safety policies and pro- topics of discussion within the OSH profession. grams, the second argument appears with increased For example, a recent opinion survey among the frequency in the OSH literature (Burnham, 2015) and readers of Safety+Health, who are predominantly blogosphere (NSC, 2014; Safety Institute of Austra- OSH professionals, indicates that only 45.5% of lia Ltd., 2009), and noticeably in Long’s (2011; 2012) respondents believe that zero injures is a realistic articles and monographs, which provide an excellent goal (NSC, 2014). literature review on the topic. The Australian company Human Dimensions While beneficial, the on high-fre- surveyed more than 21,000 participants across in- quency/probability, simple and easily observable dustries such as construction, mining, manufactur- events (e.g., speeding, not wearing hard hat, not ing and government, and found that only 35% of wearing hearing protection, wrong lifting tech- the workforce believes in the concept of zero harm niques) may obscure addressing more sophisticat- (Long, 2011). ed, technical and not easily observable hazards that Comments submitted to the Safety+Health sur- require professional safety support, and may delay vey (NSC, 2014) and to other OSH discussion implementation of needed engineering and sys- forums [e.g., Safety Institute of Australia’s (2009) temic controls, the ultimate area of concentration. OSH discussion forum] reveal the active, emotion- Manuele (2003) elaborates on a similar concern: al debate generated by zero harm safety programs. The opinions expressed fall into two general [U]nfortunately, many safety practitioners con- tinue to act on the premise that if efforts are camps: concentrated on the types of accidents that oc- 1) Zero harm is the only ethical approach to cur frequently, the potential for severe injury will safety. We cannot have any other goal than zero. also be addressed. That results in the severe in- Anything else would be the declaration of an ac- jury potential being overlooked, since the types ceptance that it is okay to hurt people. An example of accidents resulting in severe injury or fatality of this point of view: are rarely represented in the data pertaining to Statistically speaking, the likelihood of zero harm the types of accidents that occur frequently. A is extremely low. However, the vision of zero sound case can be made that many accidents harm is not about numbers. It is about a philoso- resulting in severe injury or fatality are unique and phy that no one should get hurt. It is about peo- singular events. ple. If the vision is not zero harm, what should Manuele (2003) also states that Heinrich’s relat- the vision be? Should we aim for two harm or ed premise that the predominant cases of no-injury three deaths? www.asse.org july 2016 ProfessionalSafety 39 Accident”; “Zone Zero”; “Zeroing on Safety”; “Mission Zero”; “Reaching Zero”; “Vision Zero”). The zero harm philosophy is typically defined in a company’s safety policy statements, supple- menting traditional safety program systems such as those described, for example, in ANSI/ASSE Z10-2012, Occupational Health and Safety Man- agement Systems (ANSI/ASSE, 2012). That stan- dard includes a review and comparison of major modern occupational safety and health manage- ment systems, illustrating significant similarity in OSH management around the world. The Z10 standard (as well as many similar sys- tems) includes sections on management leadership and employee participation; project and task OSH planning; risk assessment; OSH program imple- mentation and operation; evaluation and correc- While many companies’ safety policy tive action; encouraging employee participation; statements declare that all incidents are management review; roles and responsibilities, policy statements; assessment and prioritization, preventable, only about 45% of OSH incident investigation and audit. The hierarchy of professionals and 35% of the workforce controls is also a part of the Z10 standard. surveyed elsewhere share that belief, Additional elements typically present in zero harm programs include: according to data reviewed. 1) Expressed belief that all incidents are pre- ventable and a goal of zero incidents. Example: incidents are identical to the predominant cases of “We believe all incidents are preventable and that incidents resulting in major injuries is invalid. working injury-free is possible.” The statement can Burnham (2015) suggests asking eight questions include some supporting clauses, such as “Our ul- “before using zero as a safety target”: timate goal is to operate injury-free. We know this 1) How well will zero as a target motivate em- is possible because most of our facilities already ployees? operate injury-free on a rolling 12-month basis.” 2) What is the organization’s true purpose? It appears that those declarations apply to all com- 3) To what extent is the company willing to sup- pany employees and management, but it is difficult port perfection? to confirm whether that belief is actually shared by 4) How can the company control all the factors everyone in the company, or whether it is an im- that contribute to injuries? plied condition of employment. 5) What cognitive limitations make people sus- 2) Program uniqueness in finding a correct ceptible to error, even in the best management sys- key to injury-free work: “ABC Zero is not a typi- tems? cal safety program. As we are all aware, there are 6) How could zero as a target undermine the many other ‘zero’ programs out there, but ours is ability to lead? different in the responsibility and accountability 7) How could zero as a target hinder the organi- it places with each of you. We will not simply be zation’s ability to learn? talking about safety; we will be living it, measuring 8) Instead of zero, what approaches can a com- for it and planning to reach our ultimate goal: zero pany focus on to achieve better results? incidents.” Burnham (2015) concludes: 3) Local origin: “ABC Zero was born out of your [S]etting zero as a target may seem logical, but feedback and direction from the survey many of it could harm an organization’s safety efforts. you completed.” It may have questionable effects on employee 4) Reliance on behavior-based safety: “ABC Zero motivation; it may involve assumptions that are is based on our belief that human factors are the either wrong or unreliable; and it may prevent an biggest area of opportunity when it comes to being organization from improving. and working safely.” It would be logical to review and discuss how 5) Ambition: “ABC Zero is (or is striving to be) a zero harm programs differentiate from traditional world-class safety program.” safety programs and the ways zero harm programs Another important aspect is the purpose of the envision achieving the ultimate goal of an injury- company’s OSH program. One company states the and illness-free workplace. purpose of its zero-harm-driven corporate safety program as follows: Zero Harm vs. Traditional Safety Programs To ensure that every employee holds health, Zero-harm safety programs come under many safety and environment as a cultural value and brands. Searching via Google for “zero harm” re- believes every incident is preventable, and that sults in an array of corporate zero harm safety pro- employees work safely, protect the environment gram brands and logos (e.g., “Target Zero”; “Zero and lead the company to a sustainable future. 40 ProfessionalSafety july 2016 www.asse.org A more traditional (and OSHA’s General Duty Zero Harm: Necessary Conditions Clause-driven) safety policy purpose would sound Theoretically, succeeding at zero harm would more like the following: require a reliance on comprehensive integrated To share and communicate our commitment to safety management systems including leadership, providing a workplace free from hazards. management and technical (engineering) com- ponents. The systematic approach to zero harm The preceding statements mark the fundamental would include several steps: difference in philosophies and approaches to safety 1) Determine the incident to be prevented by management. While any employer is obligated by the program. Would it be OSHA-recordable cas- regulation to provide a workplace free from rec- es? Would first-aid cases also be included? What ognized hazards to its employees, zero-harm pro- about high-potential dangerous occurrences and grams strive to ensure a belief among all employees near-hits? that all incidents are preventable. That aspect was 2) Ensure that operations (e.g., tasks, projects) are characterized by Long (2011) as an “ideology with conducted in closed systems (i.e., where all hazards the elements of a cult.” are completely controllable). This means that any The task of transforming people’s beliefs on any uncontrollable third-party-related risks may need subject, including the subject of preventability of all to be excluded. Driving is an example of operating incidents, can be challenging at best, considering in an open system, where third-party drivers can that no consensus on that topic exists even among cause motor-vehicle incidents that DOT officially scientists or OSH professionals. It is hard to expect classifies as nonpreventable (e.g., a rear-end col- all employees and managers to subscribe to any lision of a company driver who was driving in full company-driven ideology (whether correct or not). compliance with traffic regulations). According to This is not to diminish the critical importance of the zero harm assertion that all incidents are pre- developed safety cultures and effective senior lead- ventable, rear-ending incidents are also preventable ership involvement for the overall success of any by the offenders. However, the zero harm company safety program. has no control over those offenders. Laboratory op- Appendix A of ANSI/ASSE Z10-2012 provides erations or plant operations are good examples of several examples of policy statements: closed systems in which an employer can identify •Create a safe and healthy workplace, and build and control to a specific degree all work parameters a respect for the environment. including exposure to any OSH . •Continually improve its OSH performance. 3) Eliminate completely all safety risks within the •Conform to the spirit as well as the letter of ap- closed system. In this case “no compromising on plicable laws and regulations. safety” would mean that no compromise is made •Integrate OSH considerations into business on costs or efforts to achieve 100% risk elimination planning, decision making and daily activities. for any incident, defined as a subject of a zero harm •Provide resources and training to carry out this program (step one). This may contradict traditional policy. OSH programs that are based on the premise of •Communicate our OSH policy. reasonable care and acceptable residual risk. No Another OSH policy statement example pro- modern approach in occupa- vided in Z10 includes language such as “no com- tional safety guarantees complete elimination of all promise of an individual’s well-being in anything occupational risk. Elimination of safety risk would we do” and “the implementation of actions to help include elimination of any unsafe conditions and realize a healthy, injury-free work environment is a acts, including human errors or eliminating their leadership responsibility.” influence on the outcome of an incident. From the hierarchy of controls perspective When subcontractor safety performance is in- (ANSI/ASSE, 2012; NIOSH, 2015), “managing cluded in the prime contractor’s or hiring con- workplace culture in a way to instill a belief that tractor’s zero harm program, the cited conditions all incidents are preventable” may be classified as should be applied to the subcontractors. This an administrative control (i.e., changing the way creates an additional significant challenge as the people work). At the top of the hierarchy are the level of control over subcontractors’ employees is most effective control elements such as elimination lower than over a company’s own employees. Fur- of the hazard, substitution with a less severe haz- thermore, the level of the subcontractor’s safety ard (e.g., selecting a less toxic chemical) and engi- program maturity, while it may be acceptable by neering controls. common prequalification criteria, may not be suf- From the perspective of appealing to improved ficient to achieve zero harm performance. , changing people’s beliefs on in- cident causation and preventability, and con- Zero Harm & ALARP centrating on the human factors as the greatest Traditional OSH programs are not designed opportunity for improvement. Zero harm pro- to eliminate 100% of risk. For example, in indus- grams belong primarily to a category of behav- trial hygiene, the ACGIH (2015) threshold limit ior-based safety (BBS) programs and ongoing value (TLV), time-weighted average (TWA) for discussions on zero harm effectiveness, applica- occupational exposures for chemical substances tions and further development are part of wider (significantly more stringent than OSHA discussions on BBS programs. permissible exposure limits) is defined as: www.asse.org july 2016 ProfessionalSafety 41 establishes controls to reduce the hazards to an ac- Figure 1 ceptable risk level. The risk assessment code (RAC) matrix in the AHA process requires classification of Risk Assessment Codes in task-specific risks by severity and probability as ex- Activity treme (E), high (H), medium (M) and low (L). The USACE OSH program requires RACs clas- Overall Risk Assessment Code (RAC) (Use highest code) L sified as extremely high or high after implemented controls to pass additional review and acceptance RAC Matrix from the government designated authority. There is Probability no expectation that all residual risk will be classified Severity Frequent Likely Occasional Seldom Unlikely as low and there is no classification of zero risk in Catastrophic E E H H M the AHA risk assessment process or in any known Critical E H H M L similar processes (e.g., , job hazard Marginal H M M L L analysis). When residual risk is permissible within Negligible M L L L L the standard of care, the probability of an incident Step 1: Review each Hazard with identified safety Controls and determine RAC (see above) is always present and never equals zero. Zero harm Probability is the likelihood to cause an incident or near hit, within the existing AHA and similar processes is not and is identified as Frequent, Likely, Occasional, Seldom or RAC Chart Unlikely. achievable unless all risks are effectively eliminated Severity is the outcome/degree if an incident or near hit did E = Extremely High Risk and driven to zero. In the author’s view, high risk occur and is identified as Catastrophic, Critical, Marginal or cannot be accepted, but neither can moderate or Negligible H = High Risk low risk be accepted in any activity hazard analysis Step 2: Identify the RAC (Probability/Severity) as E, H, M or L M = Moderate Risk if the zero harm target is adopted. for each Hazard on activity hazard analysis (AHA). Annotate the overall highest RAC at the top of AHA. L = Low Risk The concept of a tolerable risk, in the author’s opinion, contrary to the zero harm concept, is well Note. Adapted from Safety and Health Requirements Manual (EM 385-1-1), illustrated by the principle of as low as reasonable by U.S. Army Corps of Engineers, 2014, Washington, DC: Author. practicable (ALARP), developed by the U.K.’s Health and Safety Executive (HSE, 2015a) (Figure 2). TLV-TWA. The TWA concentration for a con- ALARP divides OSH risks into three categories. ventional 8-hour workday and a 40-hour work At the top is intolerable risk: that which cannot week, to which it is believed that nearly all work- be accepted no matter how high the controllable ers may be repeatedly exposed, day after day, costs. At the bottom is the negligible risk: that for a working lifetime without adverse effect. which is considered acceptable residual risk. The tolerable risk is located in between intolerable risk Note the terms believed and nearly all. They indi- and negligible risk. Tolerable risk is that which can cate an uncertainty, typical in any risk assessment, be tolerated when cost-effective measures have including human health and safety risk assess- been implemented. ment. Another example of such uncertainty are Anyone operating in the tolerable risk region variable permissible exposure limits for the same must demonstrate that effective risk review has chemical substance in various regulations. been conducted and that the cost-effective con- EPA’s (2005) risk assessment guidance and tools, trols are in place that provide the lowest reason- for example, Guidelines for Risk Assess- ably achievable risk possible. While the definitions ment, is another example of a scientific approach to of intolerable, tolerable and negligible risks are calculating human health risks, in that case based open to interpretation, the ALARP approach con- on the exposure duration and adsorbed or ingested tradicts the zero harm expectations as residual risk doses of carcinogenic and noncarcinogenic chemi- is tolerated in ALARP and the cost-benefit analysis cals. The document discusses and considers multi- on controls, to some degree, is a “compromising ple uncertainties in great detail. The acceptable risk on safety” while zero harm programs are declaring defined by the guidance is never a zero. “no compromising on safety” as a core principle. For known or suspected , accept- able exposure levels are generally concentration Conclusion levels that represent lifetime cancer risk to an in- -4 -6 •There is an ongoing discussion on zero harm dividual of between 10 (1 in 10,000) and 10 (1 safety program with the arguments ranging from in 1,000,000) using information on the relation- “Zero harm is the only ethical approach to safety” ship between the exposure dose and response. to “The zero harm goal is unachievable and leads (EPA, 2005) to misallocating safety resources to multiple minor U.S. Army Corps of Engineers’ (2014) Safety and hazards, thereby increasing the risk of missing the Health Requirements Manual EM 385-1-1 requires an major hazards.” activity hazard analysis (AHA) (Figure 1) conducted •Zero harm programs belong primarily to a cat- “for each work activity involving a type of work pre- egory of BBS programs and ongoing discussions senting hazards not experienced in previous project on zero harm effectiveness, applications and fur- operations or where a new work crew or subcontrac- ther development are part of wider discussions on tor is to perform the work.” The AHA process is an these programs. The argument on misallocating OSH risk assessment tool. It defines the activity or safety resources to numerous minor hazards with work to be performed, identifies the hazards and the existing risk of an incident caused by a unique 42 ProfessionalSafety july 2016 www.asse.org Figure 2 As Low as Reasonably

Practicable (ALARP) so could harm an organiza- Risk magnitude tion’s safety efforts. It may have questionable effects on employee motivation; it may involve assumptions that are Risk cannot be tolerated either wrong or unreliable; Unacceptable (except in and it may prevent an orga- extraordinary High region nization from improving. PS risk circumstances)

References Risk tolerable only if reduction is impracticable American Conference of or cost is grossly disproportionate to the Governmental Industrial Hy- improvement gained gienists (ACGIH). (2015). 2015 TLVs and BEIs. Cincinnati, OH: Tolerable ACGIH Signature Publications. Medium region ANSI/ASSE. (2012, June 27). There risk needs to be a Occupational health and safety system in place to (ALARP) ensure that risks are management systems (ANSI/ periodically reviewed to ASSE Z10-2012). Des Plaines, examine whether IL: ASSE. further controls are Burnham, M. (2015, April). appropriate Risk tolerable if cost reduction would exceed Targeting zero: Eight questions the improvement gained to ask before using zero as a safety target. Professional Safety, No additional measures 61(4), 40-45. Increasing individual risks and societal concerns societal and risks individual Increasing Broadly are necessary except EPA. (2005, March). Guide- maintaining usual acceptable lines for carcinogen risk assess- precautions ment (EPA/630/P-03/001F). Low region risk Washington, DC: Author. Health and Safety Executive (HSE). (2015a, June 20). ALARP “at a glance.” Retrieved from www.hse.gov.uk/risk/theory/ Note. Adapted from www.risk-assessments.org. alarpglance.htm HSE. (2015b, Sept. 29). RIDDOR—Reporting of inju- unrecognized major hazard is also present in BBS ries, diseases and dangerous occurrences regulations program evaluations. 2013. Retrieved from www.hse.gov.uk/riddor •While many companies’ safety policy state- Ivensky, V. (2015, Aug.). Reporting and recordkeep- ments declare that all incidents are preventable, ing requirements: Their influence on safety manage- only about 45% of OSH professionals and 35% of ment in the U.S. and the U.K. Professional Safety, 61(8), the workforce surveyed elsewhere share that belief, 34-37. Long, R. (2011). The zero aspiration: The maintenance according to data reviewed. of a dangerous idea. Kambah, Australia: Human Dimen- •Traditional OSH programs are not designed sions. to eliminate 100% of risks. Reviewed risk assess- Long, R. (2012). For the love of zero: Human fallibility ment and risk management approaches contradict and risk. Kambah, Australia: Scotoma Press. the zero harm expectation in parts where residual Mangan, M. (2015, Feb. 21). Safety leadership: Ap- risk and uncertainties are tolerated, such as in hu- plying behavior-based safety to serious and fatal injury man health risk assessments, AHA processes or in prevention. Safety+Health. Retrieved from www.safety ALARP systems. andhealthmagazine.com/articles/11855-safety-leader •Attempts to eliminate 100% of risk from opera- ship-safety-in-practice-applying-behavior-based-safety -to-serious-and-fatal-injury-prevention tions would be cost-prohibitive in many cases. Ac- Manuele, F.A. (2003). On the practice of safety (3rd ceptable and unacceptable risks should be clearly ed.) Hoboken, NJ: Wiley-Interscience. defined and ways to achieve the standard of care NIOSH. (2015, April 21). Hierarchy of controls. Re- should be established, designed and funded. trieved from www.cdc.gov/niosh/topics/hierarchy •As “the vision of zero harm is not about num- NSC. (2014, July 16). What’s your opinion: Is “zero bers, and is about a philosophy that no one should injuries” a realistic goal? Retrieved from www.safety get hurt,” one potential to the debate is to andhealthmagazine.com/articles/10755-whats-your ensure that a high level of OSH care is provided -opinion-is-zero-injuries-a-realistic-goal to employees, subcontractors, the public and envi- Safety Institute of Australia Ltd. (2009, Feb. 25). Zero ronment, focusing on control or avoidance of me- harm. Retrieved from www.sia.org.au/forums/showpost .asp?ContentID=43081&ContainerID=2960&TopNodeI dium- to high-potential severity hazards no matter D=43081&CurrentPage=1 their probability. U.S. Army Corps of Engineers (USACE). (2014, Nov. •Other authors have concluded that setting 30). Safety and health requirements manual (EM 385-1-1). zero as a target may seem logical, but that doing Washington, DC: Author. www.asse.org july 2016 ProfessionalSafety 43