lndcJur Air 1993 , 3: 219-226 Cvpyrighc <°'; 1\tlunksgaard I 993 Prim ..· J in D~muurk · all ri'ghts res11n:cd Indoor Air ISSN 0905-6947 Indoor Air : A Public Perspective

Jonathan M. Samet, M.D. 1

Abstract Introduction In developed and developing countries, indoor Although indoor environments have long been is gaining increasing prominence as a public health problem. known to be associated with clinically evident ad­ Time-activity studies and exposure surveys have shown the verse health effects, only recently has indoor air pol­ dominant contributions of indoor environments to popula­ lution received widespread recognition as a public tion exposures for many pollutants. Mounting epidemio­ health problem (National Council, 1981; logical evidence documents adverse health effeas of indoor Sigerist, 1943; Spengler and Sexton, 1983). Earlier pollutants and assessments indicate that indoor carcino­ concerns about specific clinical entities such as car­ gens ma_v contribute substantially to the 's burden of lung and other . Unacceptable bon monoxide poisoning and have been has also been identified as a common cause of symptoms. supplanted by a broader and unifying view of indoor This paper addresses the public health problem posed by air pollution as a single threat to public health. The indoor air pollution, offering a schema for categorizing emergence of this conceptualization of indoor air adi:erse health effeas of indoor air pollution, and considers pollution as a single public health problem followed the co111plexity of estimating the full scope of the problem. the reduction of outdoor air pollutant concen­ trations in many developed countries and the recog­ nition of the dominant contributions of indoor ex­ posures to total personal exposures for many pol­ lutants (National Research Council, 1981; Spengler and Sexton, 1983). Beginning in the 1970s, evidence from epidemiological studies provided affirmation that indoor air pollutants cause adverse health ef­ fects (Samet and Spengler, 1991). The continuance and growth of the International Conferences on Indoor Air Quality and re­ flect the emergence of public health concern about indoor air pollution. These and similar conferences have facilitated interactions among the diverse disciplines involved in indoor air quality issues: engineers, architects, clinically-oriented , epidemiologists, persons with expertise in and , and persons from regulatory and public health agencies and from the many industries with interest in indoor air quality. In the 15 years that have passed from the First International Indoor Climate Symposium to the most recent conference, Indoor Air '93, there have been substantial advances in our knowledge of the KEYWORDS: adverse health effects of indoor air pollution. We Air pollution - indoor, Air pollution/ae (adverse effects), are beginning to gain understanding of the broad , Risk factors, susceptibility array of health concerns associated with indoor air pollution. The progression of our knowledge can be 1 Department of and the New Mexico Tumor Regis­ try, Research and Treatment Center, University of readily demonstrated by a review of the proceedings New Mexico, Albuquerque, NM 87131, U.S.A. of earlier conferences in this series. The 47 papers 220 Samet: Indoor Air Pollution: A Public Health Perspective

at the first conference in 1978 included only a few explicit definition, and the boundary between "ad­ titles suggesting a focus on health, with thermal verse ' and 'non-adverse" has been a subject of comforc and odor receiving the greatest emphasis controversy. Both researchers (see Ferris, (Fanger and Valbj0rn, 1979). By contrast, health 1978 and Higgins, 1983 for example) and a commit­ was the pervasive theme in the hundreds of presen­ tee of the American Thoracic ( 1985) have tations at the Fifth International Conference held offered conceptual frameworks for defining adverse in Toronto in 1990 (Walkinshaw, 1990). health effects. 1n a 1978 review of the health effects This overview highlights the now extensive of exposure to regulated air pollutants, Ferris ( 1978) literature on the adverse health effects of indoor air began by considering the definition of "health ef­ pollution and offers a classification scheme for these fects ' and "how does one categorize adverse?" He effects. My purpose is to systematize a seemingly acknowledged the difficul'ry of interpreting effects endless listing of health effects that range from de­ ar low levels of exposure and the considerable judg­ creased comfort and .irritation and possibly neurotic mental element in any determination of adversity. concerns about the quality of indoor air to death In a 1983 paper, Higgins (l 983) offered the follow­ from asphyxiation. Initially, I consider concepts of ing: "An adverse health effect may be defined as a health and adverse health effects that are fundamen­ biological change that reduces the level of w 11- tal to any classification system. Subsequently, a being or functional capacity". This definirion shares classification system and examples are provided. emphasis on well-being with che prevailing defi­ The scope is limited to the problem of indoor air nitions of health. -.... pollution in developed countries. Smith and others A committee of the American Thoracic Society have addressed the public health consequences of ( 1985) offered guidelines on what constitutes an ad­ indoor air pollution in developing countries (Smith verse respiratory health effect in a 1985 report. The 1987; Chen et al., 1990). committee defined adverse respiratory health effects as "medically significant physiologic or pathologic changes generally evidenced by one or more of the Concepts of Health and Adverse following: 1. interference with the normal activity Health Effects of the affected person or persons, 2. episodic respir­ Concepts of health have been variable across his­ arory illness, 3. incapacitating illness 4. permanent tory, changing in response to societal expectations respiratory , and/or 5. progressive respiratory and understanding of the nature and causes of dis­ dysfunction". The committee emphasized the inter­ ease (Last, 1983; Sigerist, 1943). The World Health pretation of ep.idemiological data and implicicly fo­ has described health as "a state of cused on ourdoor air pollution. The committe complete physical, mental, and social well-being, noted that all changes are not adverse and described and not merely the absence of disease or infirmity". a spectrum of response extending from poUurion Last (1983) has defined health as "a state character­ exposure through mortality. In this continuum, the ized by anatomic integrity, ability to perform per­ boundary between adverse and non-adverse effects sonally valued family, work, and roles; was placed between "physiologic changes of uncer­ ability to deal with physical, biologic, and social tain significance" and "pathophysiologic changes". stress; a feeling of well-being; and freedom from the An editorial commencing on these guidelines risk of disease and untimely death". If the broad pointed to the potentially variable interpretation of construct of health explicit in these definitions is the language and the inherent difficulty of establish­ accepted, then adverse effects of air pollution in­ ing criteria for "medical significance" (Samet, clude not only clinically evident disease but more 1985). subtle symptomatic and physiological responses and The U.S. Environmental Protection Agency has compromise of well-being and an increased risk of not directly defined an adverse health effect. In con­ disease. sidering the evidence on the health effects of ozone Any definition of an "adverse health effect" is in the 1989 Office of Air Quality Planning and Stan­ also made in the context of prevalent societal values. dards (OAQPS) Staff Paper, Agency staff acknowl­ The need to define this context has been made clear edged the need for a framework for categorizing in interpreting the language of the Clean Air Act in lung function responses to ozone as either adverse the . While the Clean Air Act uses the or non-adverse (U.S. Envirorunental Protection term "adverse health effect," it does not offer an Agency, 1989). Normal exposed to Samet: Indoor Air Pollution: A Public Health Perspective 221 ozone during respond with a reduction of In regulating outdoor air quality, the U.S. En­ lung function as assessed by spirometry and respir­ vironmental Protection Agency has explicitly atory symptoms questionnaires. Responses placed attempted to set standards to protect the health of into a "mild" category according to Agency criteria all susceptible groups within the population (except were not considered adverse. Recent studies of those requiring life-support systems) (Frank, 1988). ozone toxicity in further illustrate the diffi­ Approaches to controlling indoor air pollution need culty of classifying responses as adverse or non-ad­ to recognize the heterogeneity of . It verse. Abnormalities indicative of injury in bron­ may not be practicable to assure protection in in­ choalveolar lavage fluid can be found in volunteer door environments for the most susceptible persons. subjects exposed to ozone concentrations near the For example, some severe asthmatics may respond current U.S. National Primary Ambient Air Qual­ adversely to levels of biological pollutants or par­ ity Standard (Devlin et al., 1991); does evidence of ticulate matter commonly found in indoor environ­ any response, i.e., a marker of injury, constitute an ments. The recent emergence of persons who ident­ adverse health effect? ify themselves as sensitive to contaminants in indoor Increased risk for cancer (or other ) esti­ environments merits consideration; regardless of mated by risk assessment methods has not been ad­ underlying pathogenetic mechanism(s), persons dressed in these definitions. The increased for with "multiple chemical sensitivity" are a growing cancer estimated for individuals exposed to carcino­ and vocal group who are asking for uncontaminated gens have no detectable correlates at present and air in indoor environments. Persons who are immu­ the ultimate manifestation of the adverse health nocompromised by acquired immune deficiency effect is discrete (i.e., the development of malig­ syndrome (AIDS) or other diseases, or from thera­ nancy). We still lack biological markers indicative peutic agents, represent another growing suscep­ of a particular cancer being caused by a particular tible population at risk for transmitted environmental agent, and there is no universally indoors. Guidelines for indoor air quality have ac­ recognized level of increased risk that is deemed knowledged heterogeneity of response in regard to unacceptable. For example, at present within the the perception of indoor air quality as "acceptable". U.S. Environmental Protection Agency, risk levels Standards of both the American Society of Heating, at which regulatory responses are invoked vary, and Refrigerating and Air-Conditioning Engineers and no effect has yet been made to establish a worldwide of the Commission of the European standard for the level of unacceptable risk. Further­ define indoor air quality as acceptable on the basis more, risk levels are dependent on the assumptions of acceptance by 80% of the members of a panel made in performing risk assessments, and changes (ASHRAE, 1989; Commission of the European in the assumptions can have a substantial impact on Communities, 1992). the risks assigned to carcinogens. Lacking a clear definition of an "adverse health Varying susceptibility to indoor pollutants effect", any schema for classifying adverse health further complicates any schema for classifying ad­ effects must be based on an assumption of prevail­ verse health effects. Responses to indoor air pol­ ing societal views and is, therefore, subject to review lutants are not uniform within populations. The and criticism. Even the designation of selected term "susceptible" has most often been applied to groups within the population as susceptible requires groups of people who share one or more character­ such assumptions. Nevertheless, proposal of a istics that place them at increased risk compared to classification scheme is a starting point for the move people without these characteristics. Even within a toward better public health protection. susceptible class of persons, a range of susceptibility can be assumed for many determinants of suscepti­ bility and environmental agents. Many susceptibil­ A Classification of Adverse Health ity factors are potentially relevant to indoor air pol­ Effects of Indoor Air Pollution lution; for example: the underlying degree of air­ Introduction ways responsiveness or the presence of asthma; the As with outdoor air pollution, a spectrum of health presence of cardiac or ; abnormal responses to indoor air pollution can be identified lung function and the presence of chronic obstruc­ (Table 1). Table 1 provides a classification of these tive pulmonary disease; the presence of atopy; and responses including categories for disease, impair­ inherently increased risk for respiratory cancer. ment, symptoms, increased risk, and perceptions. 222 Samet: Indoor Air Pollution: A Public Health Perspective

Table 1 A classification of the adverse effects of indoor air commercial building environments, this category of pollution. adverse effects, e.g., hypersensitivity pneumonitis, Clinically evident diseases: Diseases for which the usual methods has been referred to as specific building-related ill­ of clinical evaluation can establish a causal link to an indoor nesses (American Thoracic Society, 1990; Marbury air pollutam. and Woods, 1991). However, the distinction be­ Exa.cerbation of disease: The clinical starus of already established disease is exacerbated by indoor air pollution. tween specific building-related illnesses and the Increased risk for diseases: Diseases for which epidemiological non-specific syndrome referred to as sick (or tight) or other evidence establishes increased risk in exposed individ­ building syndrome rests on the capability of estab­ uals. However, the usual clinical methods indicalive of injury typically cannot establish the causal link in an individual pa­ lishing a clinical diagnosis and this group of adverse tient. effects is better recognized as unified on this basis. Physiological impairmenr: Transient or persistent effects on a The occurrence of the entities included in this cat­ measure of physiological functioning which are of insufficient magnitude to cause clinical disease. egory is also not limited to commercial environ­ Symptom responses: Subjectively reported responses which can ments. be linked to indoor pollucancs or are attributed to indoor pol­ The public health burden posed by diseases in lutants. Perception of unacceptable indoor air quality: Sensing of indoor this category is potentially estimable by using hospi­ air qualicy as uncomfortable to an unaccepra ble degree. tal discharge and other medical care data bases along Perception of e.-.:posure to indoor air pollucnnrs: Awareness of with population survey techniques. To date, how­ exposure to one or more pollucants with an unacceptable level of concern about exposure. ever, estimates have not been published.

Exacerbation of Established Disease

Table 2 Selected examples of clinically evident disease linked Conditions that may be exacerbated by indoor air ta indoor oir pollution. pollution are common in the population. Asthma, a chronic respiratory disease characterized by hyper­ Carbon monoxide poisoning responsiveness of the lung's airways to environmen­ Hemorrhagic pneumonitis from high levels of N00 Hypersensitivity pneumonitis and humidifier fever tal factors, affects approximately 5 to 10% of Legionella pneumonia children and adults. Indoor exposures to Cat- and mite-induced asthma danders, molds, and allergens from house dust mi­ tes and other insects may both cause and worsen the clinical status of persons with asthma. Environ­ Each category is treated below and examples pro­ mental tobacco smoke may increase the non-specific vided. responsiveness of the lung to environmental stimuli and even trigger attacks of asthma (Samet, 1991 ); Clinically Evident Disease the U.S. Environmental Protection Agency has re­ While exposures co indoor air pollutants are univer­ cently estimated that exposure to environmental to­ sal, clinically evident cases of pollucion-related dis­ bacco smoke exacerbates symptoms in about 20% ease appear to be relatively infrequent. ln the case of the 2 million to 5 million U.S. children with of such a clinically evident disease, a link can be asthma and is a "major" exacerbating factor in ap­ established to an indoor pollutant by specific diag­ proximately 10% (U.S. Environmental Protection nostic tests (Table 2). For example, an appropriate Agency, 1992a). clinical picture and an elevated serum precipitin tit­ er are sufficient to document hypersensitivity pneu­ Increased Risk for Disease monitis due to thermophilic actinomycetes contami­ Many pollutants in indoor air are associated with nating an air-conditioning system ~ eissman and increased risk for a variety of malignant and non­ Schuyler, 1991). The level of carbon monoxide malignant diseases (Table 3). The evidence sup­ bound to hemoglobin (carboxyhemoglobin) pro­ porting the relationships between exposures to these vides a marker of exposure to concentrations of car­ agents and increased risk comes from epidemio­ bon monoxide associated with carbon monoxide logical studies, short-term exposures of volunteer poisoning. Skin tests and serologic tests can provide subjects, animal studies, and in vitro toxicological evidence of sensitization to antigens that produce studies. The population burden of disease attribu­ disease through immediate hypersensitivity re­ table to such agents is often estimated using quanti­ sponses. tative risk assessment, a technique for integrating In classifying illnesses associated with public and information on the population pattern of exposure ftrmmr•

Samet: Indoor Air Pollution: A Public Health Perspective 223

Table 3 Selected examples of exposure-disease associations However, risk assessment methods can be used to for indoor air pollutants describe the population's increased risk of lung can­ Radon: Lung Cancer cer from radon and even to apportion this risk be­ Environmental Tobacco Smoke: Lung cancer, increased lower tween radon and the combined effect of radon and respiratory illness in smoking (National Research Council, 1988; U.S. Benzene: Leukemia : Lung cancer and Environmental Protection Agency, 1992b). For ra­ Formaldehyde: Nasal cancer don, risk estimates are made by applying the ex­ posure-response relationship observed in the miners to the distribution of exposures received by the gen­ Table 4 Prevalence of selected symptoms in office workers by eral population (Samet, 1992). Adjustment can be type of ventilation system:. made for physical differences between mines and residences and for physiological differences between identification: The determination of whether an agent is the two circumstances of exposure (National Re­ causally linked to the health effect of concern Dose-response assessment: The determination of the relation search Council, 1991 ). Key uncertainties include the between level of exposure and risk of the health effect extrapolation from higher exposures and exposure Exposure Assessment: Description of the exrent of ex­ rates in mines to generally lower exposures and ex­ posure Risk characterization: Description of the human risk, including posure rates in residences and the pattern of the uncertainties combined effect of smoking and radon. * Source: Based on data from the National Research Council, For some agents in this category, risk assessments 1983. provide estimates of the numbers of attributable cases of disease or death in selected populations. For example, the current distribution of indoor ex­ and the exposure-response relationship to obtain an posures to radon in the United States is estimated estimate of the attributable number of disease cases to cause approximately 14,000 lung cancer deaths or deaths associated with the exposure. The U.S. annually (U.S. Environmental Protection Agency, National Research Council (1983) has characterized 1992b). Exposure-response relationships can also be risk assessment as a four-step process (Table 4). used to describe the risks associated with specific t The results of risk assessment guide develop­ exposures and thereby project the risks for individ­ ment and implementation in the separate activities uals. However, an individual case of lung cancer in of risk . The risk assessment approach the general population could not be attributed to also supplies a framework for characterizing uncer­ indoor radon with a high degree of certainty; the tainties in the risk estimates and identifying research best estimate of the probability of causation in an needs. To date, risk assessment has been most wide­ individual would be the attributable risk for a popu­ ly applied to carcinogens, although the methodology lation of similar individuals. Risk assessments have could be applied to noncarcinogens as well. also been published for environmental tobacco The problem of indoor radon and lung cancer smoke and lung cancer (Repace and Lowrey, 1990). is illustrative. Radon, a ubiquitous contaminant of indoor air in homes, has been causally linked to lung Phy siological Impairment cancer by epidemiological studies of underground Exposures to indoor pollutants can impair physio­ miners and confirmatory animal investigations (Sa­ logical functioning, although not to a degree neces­ met, 1989). To date, there is little evidence that lung sarily associated with or disease. For ex­ cancers in radon-exposed miners differ from lung ample, exposure to environmental tobacco smoke cancers in the general population, and lung cancers during childhood reduces the rate of lung growth associated with radon cannot be separated directly and the maximum level of lung function achieved; from those caused by other factors; moreover, the average estimated effect is not anticipated to be studies of underground miners suggest that ciga­ clinically detectable nor to be associated with re­ rette smoking and radon exposure synergistically duced functional capacity (Samet and Spengler, increase lung cancer risk (National Research Coun­ 1991). Similarly, low levels of carbon monoxide ex­ cil, 1988). Thus, lung cancers in individuals cannot posure transiently impair oxygen delivery to tissues; be designated as caused by radon or by cigarette however, the impact on exercise capacity is limited smoking or another factor. In fact, the majority of and likely to be manifest only during maximal activ­ radon-associated cases occur in cigarette smokers. ity (Coultas and Lambert, 1991). On the other hand, 224 Samet: Indoor Air Pollution: A Public Heal th Perspective reduced oxygen transport in the carbon-monoxide­ Perception of Unacceptable Indoor Air Quality exposed individual with coronary artery disease may The perception that indoor air quality is unaccept­ increase the likelihood of clinically significant myo­ able should be considered as distinct from the cardial ischemia. The public health relevance of this symptoms caused by indoor air pollutants. To the category of adverse effects of indoor air pollution extent that unacceptable indoor air quality reduces has received little consideration to date. well-being, the perception of indoor air quality as unacceptable should be classified as an adverse Symptom Responses health effect in the context of current concepts of Epidemiological evidence links specific indoor air health. Judgements as to the acceptability of indoor pollutants to a variety of symptoms. Environmental air quality presumably integrate multiple character­ tobacco smoke exposure is associated with increased istics of the air, including the presence of odor and risk of respiratory symptoms in children (U.S. De­ irritants, humidity, air movement, and temperature partment of Health and , 1986). (Berglund and Lindvall, 1990; Spengler and Samet, The sick-building syndrome is a non-specific con­ 1991 ). Undoubtedly, there is a range of responses stellation of symptoms characteristically affecting and expectations across the population. Physical and multiple occupants of a building (American Tho­ psychological aspects of the environment not di­ racic Society, 1990). In some outbreaks of sick­ rectly related to indoor air quality may also influ­ building syndrome, specific pollutants are found to ence judgments as to the acceptability of indoor air cause the symptoms, but in the majority only gen­ quality. eral etiological factors reflecting building operation The findings of a nationwide survey of U.S. office and maintenance can be identified (American Tho­ workers suggest that dissatisfaction with the air racic Society, 1990; Samet et al., 1988). quality in offices is common (Woods et al., 1987). Estimates of the burden of symptoms associated Of 600 workers surveyed by telephone in 1984, 20% with indoor air pollutants have not been made; how­ perceived that their work performance was affected ever, this burden is likely to be substantial because "often" or "sometimes". Aspects of indoor air qual­ of the high prevalence rates of exposure to agents ity that were found to be "very serious" or "serious" associated with symptoms. Surveys of the preva­ by at least 50% of the affected respondents included lence of work-related symptoms, although not con­ lack of air movement (67%), being too hot in sum­ ducted in random samples of buildings, indicate mer (61 %), stagnant or still air (55%), cigarette high prevalence rates for symptoms. Burge and co­ smoke (54%), being too cold in winter (53%), and workers (1987) described symptom rates in 4,373 being too humid in summer (50%). office workers in 42 different buildings. Symptoms were considered work-related if they occurred more Perception of Exposure to Indoor Air Pollutants than twice during the previous 12 months and im­ The perception of exposure to indoor pollutants proved on days away from the office. Using this should also be regarded as an adverse health effect definition, the mean number of work-related symp­ if the perception reduces well-being. The range of toms varied across the sample of buildings from responses to the perception of exposure is broad, approximately 1.5 to 5. Symptoms of eye and upper extending from annoyance because of an odor to airway irritation and headaches were common the sometimes disabling symptom complex now fre­ (Table 5). quently referred to as "multiple chemical sensi-

Table S Prevalence of selected symptoms in office workers by type of ventilation system:.

Ventilation type Symptoms Dry eyes Blocked nose Dry throat Headache Natural 18 (%) 40 (%) 36 (%) 39 (%) Mechanical 20 32 33 33 Local induction/fan coil 34 58 56 52 Central induction/fan coil 31 57 54 47 All air 31 45 46 43 Whole group 27 47 46 43 *Source: Data from Table 5 in Burge et al., 1987. Ot:l15M'11l

Samet: Indoor Air Pollution: A Public Health Perspective 225 tivity". While the pathogenetic mechanisms under­ to this question is affirmative. The emergence of lying multiple chemical sensitivity remain unknown indoor air pollution as a unified public health con­ and may be multiple, some cases are likely to repre­ cern has fostered the interdisciplinary interactions sent somatization triggered by awareness of ex­ that are needed to find solutions. If the conse­ posure to environmental pollutants (Miller, 1992). quences of indoor pollution exposures are frag­ The numbers of persons who are adversely affected mented into a series of seemingly unrelated prob­ by the perception of exposure· cannot presently be lems (e.g., combustion products, hypersensitivity estimated. pneumonitis, comfort, and radon), the imperative to achieve solutions through actions needing coordi­ nation of a wide range of professionals and organiza­ The Public Health Burden of Indoor tions is diminished. More information on the public Air Pollution health burden posed by indoor air pollution is cen­ This broad classification scheme makes clear the tral to maintaining research on sources, exposures, potential scope of the public health problem posed health effects, and control measures. by indoor air pollution and the difficulty of fully estimating the magnitudes of adverse effects in each of the six categories (Table 1). Within this schema, Acknowledgements some adverse effects have clear definitions (e.g., This work was supported by a contract to the death from carbon monoxide poisoning) while Health Effects Institute (HEI), an organization fun­ others are defined on the basis of subjective re­ ded by the U.S. Environmental Protection Agency sponses, and any criteria for placing responses into (EPA) (assistance agreement X812059), automobile a particular category require the assumption of a manufacturers, and the Gas Research Institute societal framework for separating adverse from non­ (GRI). The contents of this paper do not reflect the adverse responses. Risk assessement has been used views of the HEI, nor do they reflect the to quantify the hazard associated with some carcino­ of the EPA, automobile manufacturers, or the GRI. gens of current concern and a few exposures associ­ Support was also given by the U.S. Department of ated with non-malignant respiratory effects. How­ Energy, Office of Energy Research, under Grant ever, symptom and perceptual responses to indoor No. DE-GF04-90ER60950. air pollutants can only be addressed by directly in­ vestigating exposed populations. To date, few studies have addressed these responses in popula­ References tion-based samples, and we thus lack any compre­ American Thoracic Society (1985) "Guidelines as to what consti­ hensive and population-based assessment of the full tutes an adverse respiratory health effect, with special refer­ scope of the public health consequences of indoor ence to epidemiologic studies of air pollution", American Re­ air pollution. view of Respiraiory Diseases, 131, 666-669. American Thoracic Society ( 1990) "Environmental controls and We also lack societal guidelines for describing the lung disease", American Review of Respiraiory Disease, 142, magnitude of the problem. In the United States, 915-939. for example, has evolved for single ASHRAE (1989) Ventilaiion for Acceptable Indoor Air Qua/icy, , GA, American Society of Heating, Refrigerating and pollutants without broader consideration of more Air Conditioning Engineers (ASHRAE Standard 62-1989). fundamental principles. Even for single agents, e.g., Berglund, B. and Lindvall, T . ( 1990) "Sensory criteria far radon, conflicting views among involved regulators, healthy buildings". In: Walkinshaw, D.S. (ed.) Proceedings of Indoor Air '90, Ottawa, Mortgage and the Congress, and scientists have led to persistent Corporation, pp. 65-79. controversy (Cole, 1993). Without any firm guide­ Burge, S., Hedge, A., Wilson, S., Bass, J.H. and Robertson, A. lines, the scope of the problem of indoor air pol­ (1987) "Sick building syndrome: a study of 4373 office workers", Annals of Occupaiional , 31, 493-504. lution can be readily manipulated, as underlying Chen, B.H., Hong, C.J., Pandey, M.R. and Smith, K.R. (1990) assumptions in risk assessments are varied. A pro­ "Indoor air pollution in developing countries'', World Health cess is needed for establishing a conceptual frame­ Staiistics Quarterly, 43, 127-138. Cole, L.A. (1993) Elements of Risk; The Politics of Radon, Wash­ work for indoor air pollution (Nero, 1993; Spengler ington, DC, AAAS Press. and Samet, 1991). Commission of the European Communities (1992) Indoor Air Would the goal of public health protection be Quality and its Impact on Nian: Guidelines for Ventilaiion Re­ quirements in Buildings, Luxembourg, Office for Publications better achieved by more complete information on of the European Communities (Report No. 11). the health effects of indoor air pollution? My answer Coultas, D.B. and Lambert, W.E. (1991) "Carbon monoxide". 11 I 226 Samet: Indoor Air Pollution: A Public Health Perspective

In: Samet, J.M. and Spengler, J.D. (eds) Indoor Air: a Health Samet J .M. (1985) "Defining an adverse respiratory health ef­ Perspective, Baltimore, MD, Johns Hopkins University Press, fect" American Review of RespiraJory Disease 131, 487. pp. 187-208. Samet, J.M. (1989) "Radon and lung cancer" Joumal of rhc Devlin, R.B., McDonnell, W.F., Mann, R. Becker, S., Hause, Nazio11al Cancer l11st.it111c, 81, 745- 75i. D .E., Schreinemachers, D. and Koren, H.S. (1991) "Ex­ Samec, J.M . (1991 ) "Environmemal tobacco smoke". In: posure of humans to ambient levels of ozone for 6.6 hours Lippmann, M. (ed.) Critical Reviews of E1ro fro111111mtai Toxi­ causes cellular and biochemical changes in the lung", American cams: Human Exposures and their Health Effects, New York, Joun1a/ of Raspfralory Cell and Molecular Biology, 4, 72-81. Van Nostrand Reinhold Company, Inc. pp. 231-265. Fanger, P.O. and Valbjarn, 0 . (1979) fodbor Climaie: Effeas Samer, J.M. (1992) " Indoor radon and lung cancer: estimating 011 Human Comfort, Performanca, a11d Hcalih i11 Rosidcmial, the risks'', ~\rmcrn Joumal of Medicine 156, 25-29. Commercial, and LigJ11-/11d11sr.ry B11ildh1f1S, Copenhagen, Samet, J.M. Marbury, M .C. and Spengler, J.D. (1988) "Health Danish Building Research lnstirure. effecrs and sources of indoor air pollution. Part II", A111erica11 Ferris, B.G. (1978) "Health effects of exposure to low levels of Review of RespitaLory Disease, 137, 221-242. regulated air pollutants: a critical review", Journal of the Air Samet, J.M. and Spengler, J.D. (1991) Indoor Air Pollution. A Pollution Control Association, 28, 482-497. Health Perspeaive, Baltimore, MD, Johns Hopkins University Frank, R. (1988) "Susceptibiliry to air pollution: the clean air Press. act". In: Brain, J.D., Beck, B.D., Warren, A.J. and Shaikh, Sigerist, H.E. (1943) Civilizarion and Disease, Chicago, Univer­ R.A. (eds) Variations in S11scep1ibili1J1 to illhaled Polluiams, sity of Chicago Press. Bakimore, MD, Johns Hopkins University Press, pp. Smith, K.R. (1987) Biofuels, Air Pollution, and Health. A Global 443-459. Review, New York, Plenum Press. Higgins, 1.T.T. (1983) "What is an adverse health effect?" Jour­ Spengler, J.D. and Samet, J.M. (1991) 'A perspective on indoor ·I nal of the Air Pollution Control Association, 33, 661--663. and outdoor air pollution". In: Samet, J.M. and Spengler, Last, J.M. (1983) A Dictionary of , New York, Ox­ J.D. (eds) Indoor Air Pol/1aion, a HeaiLlt Perspective Baltim re, ford University Press. MD, Johns Hopkins University Press, pp. 1- 29. Marbury, M .C. and Wfoods, J.E.J. (1991) "Building-related ill­ Spengler, j .D. and Se:Kton, K. (1983) " Indoor air pollution: a ness". In: Samet, J.M. and Spengler, J.D. (eds) Indoor Air public health perspective', Scic11cc, 221, 9-J 7. ,, Pol/wion: a Health Perspeaive, Baltimore, MD, Johns Hop­ .S. Department of Health aod Human Services ( 1986) The I· kins University Press, pp. 306-322. Health Co11scg11e11ce.t of Iuvolzmzary Smoking, a Report of the I Miller, C.S. (1992) "Possible models for multiple chemical sensi­ S11rgeo 11 General, Rockville, MD, .S. Govemmem Printing I tivity: conceptual issues and role of the limbic system". In: Office. 1· Rest, K.M. (ed) Advancing the U11dcrn.a11diiig of Multiple U.S. Environmental Protection Agency (1989) Review of zlzc Na­ Chemical Se11si1ivity, Princeton, NJ, Princeton Sciemific Pub­ cional A111bie111 Air QualilJ• S1a11dard for Ozone, Assessmenr of ll lishing Co., Inc., pp. 181 -202. Scie111ijic and Technical b ifor111a/io11, Research Triangle Park, National Research Council, Committee on Indoor Pollutants NC, EPA (EPA-450/ 2-92). (1981) Indoor Pollutams, Washington, D .C., National Acad­ U.S. Environmemal Protection Agency (1992a) Enviro11me1ual emy Press. Equity: Reducing Risk for All Co11111111111iies. Report lo zhc ad111i11- National Research Council, Committee on tl)e Biological Effects istracor fro1111hc EPA E11viro11111ezual Equity V."orkgroup. Draft, of Ionizing Radiation (1988) Heallh Risk.( of Radon and Other Publication ·o. 230-DR-92-002. buemal{)' Deposited Alpha-Eminers: BE/ R IV, \XTashington, U.S. Environmcmal Protection Agency (1992b) Tech11ical up­ DC Press. pon Dowmeru for the 1992 Citi:;e11's G111d.· w Rt1tlo11 \'ti'ash­ ational Research Council, Comminee on the Institutional ington, DC t:.S. EPA (EPA 400-R-9'.!-001 ). Means for Assessment of Risks to Public Health (1983) Risk Walkinshaw, D.S. (ed.) (1990) indoor Air '90 . Procadl11gs of 1/z.: Asses1;111a111. i11 the Federal G