Tobacco Documents Online

Special Communications The Environment and the Lung Changing Perspectives Jonathan M. Samet, MD, Mark J. Utell, MD The focus of public health concern and research in regard to environmental lung diseases has changed across the century.,Illustrative agents include radon, indoor asbestos, environmental tobacco smoke, acidic aerosols, and oxidant gases, Tremendous progress has been made in understanding and preventing environmental lung diseases. However, we remain concerned about adverse consequences of breathing polluted outdoor and indoor air. In the persistent concerns about adverse effects of polluted air on the lung, a new emphasis is pervasive; the focus has shifted,from avoiding clinical disease among highly exposed individuals to protecting the population from an unacceptable burden of risk. The technique of quartitative risk assessment has become increasingly important for characterizing the safety of environmental agents. The resulting emphasis on the final risk projection and attendant uncertainties may overly emphasize gaps in our knowledge. (JAiYlA.1991 ;266:670-675) AS THE 20th century ends, we are con- cerned and fearful about the adverse consequences of breathing polluted air, whether outdoors or indoors, and are asking for reduction of known and po- tential hazards. Yet, during this centu- ry the environmental causes of many lung diseases have been identified, the pathogenetic mechanisms underlying the development of some of these dis- eases have been described, and control measures have been implemented with at least partial success for a number of the injurious agents. In the United States and many other developed coun- tries, elaborate regulations and en- forcement mechanisms are in place to From the Pulmonary and Critical Care Division, De- partment of Medicine, and the New Mexico Tumor Reg- istry, Cancer Center, University of New Mexico, Albu- querque (Dr Samet); and the Departments of Medicine and Environmental Health Sciences, University of Rochester(NY) School of Medicine, and the Pulmonary and Critical Care Unit and Occupational Medicine Pro- gram, University of Rochester Medical Center (Dr Utell). Reprint requests to New Mexico Tumor Registry, Medical Center, 900 Camino de Salud NE, Albuquer- que, NM 87131 (DrSamet). ensure that air outdoors and in work- places does not pose unacceptable health risks; indoor air pollution has been recently recognized as a potential threat to health as well and programs have been implemented to address some of its hazards, for example, radon and asbestos. Despite the increasing scientific evi- dence and the far-reaching array of con- trol measures, the public•is still con- cerned about breathing the often visibly polluted air that remains in many cities and even rural locations and is learning that indoor air may be contaminated with the same chemicals that are re- leased by industry and vehicles, and even by such invisible pollutants as ra- don. While regulations have controlled many of the hazards of workplaces heavily contaminated by dust and fumes, new types of manufacturing have introduced novel and uninvesti- gated exposures, and changing work environments have led to the emer- gence of new clinical problems, such as "sick-building syndrome," and concern that some cases of recognized diseases, like asthma, may be caused or exacer- bated by changes in the indoor and out- door environments. In the persistent concerns about ad- verse effects of polluted air on the lung, a new emphasis is pervasive; the focus has shifted from the•avoidance of clinical disease among highly exposed individ- uals toward the protection of the gener- al population from an unacceptable bur- den of disease at much lower exposures, and an attempt to ensure that even the most susceptible persons are not ad- versely affected. This same emphasis extends equally to other environmental exposures and to diseases other than those affecting the lung. Quantitative risk assessment, a four-step process, has become a widely used tool for judg- ing the safety of environmental agents, providing a framework for summarizing the evidence on health risks from toxico- logic studies, controlled exposures of volunteers, and epidemiologic research with information on exposure to injuri- ous agents (Table 1).' The results of risk assessment can be used to identify areas for research, to assign priorities among environmental hazards, and to select approaches for managing risks. It must be recognized, however, that quantita- tive risk assessment is a new method- ology and that its role in regulation is still evolving.' This shift in emphasis from higher exposures producing clinical disease to lower levels projected to increase popu- lation risks has raised new questions and challenges for research on environ- mental lung disease. Providing assur- ance of safety, a level of risk judged to be acceptable,' requires precise and confident characterization of risks, and the scope of research needs to extend 670 JAMA, August 7, 1991-Vol 266, No. 5 The Environment and the Lung-Samet & Utell

Table 1.-The Four Steps of Risk Assessment* Hazard identification: The determination of whether an agent is causally linked to the health effect otconcern Dose-response assessment: The determination of the relation between level of exposure and risk of tfie health effect Exposure assessment: Description of the extent of human exposure Risk characterization: Description of the human risk, induding uncettainties *Based on reference 1. beyond testing for an exposure-disease association to quantification of risk at various levels of exposure and assess- ment of factors that modify the expo- sure-disease relationship. Data from extensive animal studies and large- scale epidemiologic investigations are often required. This study addresses the changing fo- cus across the century of public health concern and research in regard to envi- ronmental lung diseases. We review se- lected agents to illustrate the shift from disease prevention in individuals to risk reduction for the whole population and the difficulty of answering the questions now raised with regard to safety for many agents. We begin by considering the broad groups of environmental agents that produce lung disease and the mechanisms of disease patho- genesis. MECHANISMS OF LUNG INJURY BY ENVIRONMENTAL AGENTS The environmental agents in indoor and outdoor air of greatest contempo- rary concern are diverse, causing both malignant and nonmalignant diseases (Table 2). Continued concern about the risks of these and other causes of envi- ronmental lung disease is justified by the myriad pollutants inhaled in the var- ious indoor'and outdoor environments where time is spent each day, the di- verse mechanisms by which these pol- lutants cause disease, and the wide range of susceptibility to pollutants in the population. Because we inhale 10 000 to 20 000 L of air daily, doses of pollutants present even at low concen- trations may become biologically signif- icant with sustained exposure. Fortu- nately, the lung has physical, chemical, and immunologic defense mechanisms for clearing and detoxifying inhaled agents, although the defense systems may be overwhelmed by large pollutant doses or may not be fully effective against some pollutants. Atmospheric pollutants are present in the form of gases, fibers, or particles. Penetration of pollutants into the lung and retention at potential sites of injury depend on the physical and chemical properties of the agents.s Highly water- Table 2.-Selected Agents Causing Environmental Lung Disease of Current Concern and Associated Adverse Effects Agent Effect(s) Acidic aerosols Exacerbate asthma and COPD,* respiratory symptoms, reduced lung function Asbestos Lung cancer, mesothelioma, pleural disease Environmental tobacco smoke Lung cancer, respiratory infection, respiratory symptoms, reduced lung function Nitrogen dioxide Exacerbate asthma, respiratory infection and symptoms, reduced lung functlon Photochemioal pollution (ozone) Exacerbate asthma and COPD, respiratory symptoms, reduced lung function Radon Lung cancer Silica Silicosis, lung cancer Volatile organic compounds CanceS neuropsychological effects, respiratory irritation *Chronic obstructive pulmonary disease. soluble gases, such as sulfur dioxide and formaldehyde, are almost completely extracted by the upper airway of a rest- ing subject during a brief exposure, whereas less soluble gases, such as ni- trogen dioxide and ozone, penetrate to the small airways and alveoli. Pollut- ants in particulate form are usually found in nature as aerosols. The pene- tration of particles into the lung and the sites of deposition within the lung de- pend on the aerodynamic size of the par- ticle. Those greater than 10 µm are ef- fectively removed in the upper airway, whereas smaller particles penetrate and are deposited in the airways and alveoli. Fibers are defined arbitrarily as particles having a length at least three times the width. The handling of fibers by the respiratory tract depends on fi- ber width and length and susceptibility to dissolution. Exercise increases the amount of air inhaled and the proportion of oral breathing, and thereby increases the dose of inhaled pollutants. The diverse mechanisms by which in- haled gases and particles injure the lung, although not yet fully understood, can be broadly grouped as acute irrita- tion and inflammation, chronic inflam- mation accompanied by a fibrotic re- sponse for some agents, immediate and cell-mediated immune responses, and carcinogenesis (Table 3). The likelihood of an adverse response to an inhaled pollutant depends on the degree of ex- posure to the pollutan', the site of depo- sition and the rate of clearance, and the individual characteristics of the ex- posed person that determine suscepti- bility. The relationship between expo- sure and response may have different forms, depending on the mechanisms by which the pollutant causes disease (Fig- ure). The shape and slope of the expo- sure-response relationship have sub- stantial implications for assessing the risks of environmental agents.' Curves having a threshold that must be ex- ceeded to produce disease indicate that levels below the threshold are without risk; by contrast, a curve without a threshold implies that any level of expo- sure conveys some risk. For example, the linear no-threshold relationship, widely used to assess risks of carcino- gens for regulatory purposes, is consid- ered to be protective of public health because no level of exposure is without effect. The assumption of a linear no- threshold model for carcinogenesis re- mains highly controversial." Distin- guishing among the theoretical curves in the Figure cannot be readily accom- plished using either animal experiments or human data, and exposure-response relationships should be assumed on the basis of biologic plausibility.' 1LLUSTRATIVE POLLUTANTS OF CURRENTCONCERN This section briefly considers the cur- rently available evidence and concerns about risk for several air pollutants, se- lected to be illustrative of the changing emphasis of concern across the century. The pollutants discussed include asbes- tos, radon, environmental tobacco smoke, acidic aerosols and sulfur diox- ide, and oxidant pollutants, including ozone and nitrogen dioxide. Radon Radon, a naturally occurring radioac- tive gas in the decay series of uranium- 238 that is known to cause lung cancer in underground miners, is ubiquitous in indoor environments. "$ The decay prod- ucts of radon are themselves radioac- tive and release cancer-causing alpha particles, which are directly responsible for radon's carcinogenicity. The prob- lem of cancer in underground miners was first reported over 100 years ago,' and radon was considered as a possible cause of the excess lung cancer in these miners by early in the century.' Epide- miologic studies of underground min- ers, initiated in the 1950s and later, soon provided convincing evidence that ra- JAMA, August 7, 7991-Vol266, No.5 The Environment and the Lung-Samet & Utell 671

Table 3.-Principal Mechanisms Associated With Environmental Lung Disease Exposure Examples of theoretic exposure-response relation- ships. Line A shows a linear exposure-response relationship with a threshold, while line B,shows a linear nonthreshold relationship. Lines C and D are examples of nonlinear relationships (reprinted with permisson from reference 2). don caused lung cancer in miners and some information on the quantitative risk of lung cancer in relation to expo- sure. Regulations were implemented during the 1960s and 1970s to protect miners against excess lung cancer.10 By the mid-1980s, it was widely rec- ognized that radon was present in homes, sometimes reaching concentra- tions comparable with levels in uranium mines. To guide the development of public policy by state and federal agen- cies, estimates of lung cancer risk were needed across the range of concentra- tions measured in homes. Because epi- demiologic studies directly addressing these risks could not be quickly per- formed, the risks found in the studies of miners were extrapolated to the gener- al population, yielding estimates that indoor radon may cause approximately 10 000 to 20 0001ung cancer deaths an- nually in the United States.e°ll•'2 Several sources of uncertainty have reduced confidence in the extrapolation of risks from miners to the general popula- tion.'s." Should the risks observed at the higher exposures in miners be extrapo- lated to lower exposures using a linear nonthreshold relationship? Are the quantitative risks higher or lower for the general population compared with the risks for miners? In comparison with evidence from male adult miners, pre- dominantly cigarette smokers, what are risks of exposure for children, for wom- en, and for never-smokers? Some have even questioned the carcinogenicity of indoor radon.15 Reducing the uncertain- ties in assessing the risk of indoor radon poses a complex challenge for biomedi- cal research. Improved understanding of carcinogenesis may lead to better Mechanism Agents Bronchoconstriction SuHur dioxide, acidic aerosols Inflammation Ozone, environmental tobacco smoke Fibrosis Asbestos, silica Cancer F3adon,asbestPS, formaldehyde, active smoking, and environmental tobacco smoke support for a particular model of the relationship between exposure and lung cancer risk. Many case-control studies of indoor radon and lung cancer are now in progress with the objective of direct- ly estimating risk due to indoor expo- sure, and several studies have already been reported.e,1e However, the results of these studies are likely to be affected by difficult methodologic problems, and extremely large and unfeasible studies would be needed to fully address uncer- tainties and provide confident state- ments about risk." As we begin the 1990s, the risks of indoor radon remain extremely contro- versial, even though radon is an estab- lished occupational carcinogen and ex- tensive epidemiologic data from miners provide convergent risk estimates.a'e The continued controversy and wide- spread perception of uncertainty ap- pear to reflect the complexity of the questions that must be answered in sup- port•of policy development rather than weaknesses in the existing data. Risk assessment may highlight the gaps in scientific knowledge and, as illustrated by indoor radon and lung cancer, reduce confidence in good data by calling for answers to questions that cannot be readily answered. Indoor Asbestos Asbestos, a group of naturally occur- ring fibrous minerals, has been widely used in insulation and other materials in schools, public and commercial build- ings, and residences. Man-made fibers are now widely used as a replacement for many of these applications. At the start of the century, clinical cases pro- vided clear evidence that occupational exposure to asbestos caused asbestosis, a fibrotic disorder of the lung.' During the 1950s and 1960s, the results of epi- demiologic studies of workers showed that occupational asbestos exposure also caused lung cancer and mesothello- ma.' Although the information on expo- sures of workers in these studies was limited, quantitative relationships be- tween estimated exposures and cancer risk were addressed in some of the studies. Asbestos fibers can be released into the air of buildings from human activi- ties that disturb asbestos-containing material or by maintenance activities involving asbestos-containing materi- als. Thus, persons potentially at risk from asbestos exposure indoors include persons handling or contacting the ma- terial during job activities or cleaning asbestos-contaminated areas, and gen- eral building occupants if they inhale contaminated air. Because of the well- documented and widely known disease risks in historical cohorts of asbestos- exposed workers, the presence of as- bestos-containing material in buildings has prompted great public concern-and legislative programs to reduce risks of indoor asbestos. The Asbestos Hazard Emergency Response Act requires in- spection of schools for asbestos and sat- isfactory in-place management or, in some cases, removal. New research ini- tiatives have been implemented to ad- dress asbestos in public and commercial buildings. Public and private efforts for manag- ing asbestos-containing material have been undertaken to reduce the expo- sures of custodial and maintenance workers and of general building occu- pants, including general office workers and schoolchildren. Even though con- centrations of asbestos fibers in build- ings are extremely low, persons in the category of general building occupants have been considered to be at risk for lung cancer and mesothelioma21'; re- cent estimates of exposures" are somewhat lower than earlier esti- mates" as more data have become available on indoor concentrations. The risks for the general population can only be estimated by extrapolating from the risks in asbestos workers to the general population. Exposure-disease relation- ships are generalized across exposure scenarios typically differing by orders of magnitude to obtain risk projections for the general population, which are subject to great uncertainty. While in- ~ creasing information is becoming avail- ~ able on concentrations ofasbestos fibers iU indoors, epidemiologic studies cannot ~ directly assess the risks of indoor asbes- ~ tos for such populations as schoolchil- ~ dren and office workers because of the ~ l l i d d arge samp e s zes nee e . I-PA Environmental Tobacco Smoke +~ By mid-century, a marked increase ~ was evident in lung cancer deaths among men, and case-control studies carried out to explain the epidemic quickly provided consistent evidence that cigarette smoking was a strong 672 JAMA, August 7,1991 -Val 266, No. 5 The Environment and the Lung-Samet & Utell

cause of lung cancer. By 1964, sufficient epidemiologic data on smoking and health were available to support a con- clusion by the Advisory Committee to the Surgeon General that cigarette smoking caused lung cancer in men.' Fhrther research has shown that smok- ing is a cause of many malignant and nonmalignant diseases.' Nonsmokers inhale environmental tobacco smoke, a mixture of sidestream smoke and exhaled mainstream smoke. During the late 1960s and early 1970s, several reports suggested that expo- sure of children to environmental tobac- co smoke by the smoking of their par- ents increased their risk for respiratory infections and respiratory symp- toms2',"; in the late 1970s, adverse ef- fects of parental smoking on lung func- tion in children were first reported.' In 1981, reports of two epidemiologic stud- ies, one in Japan and the other in Greece, indicated that never-smokers married to smokers were at increased risk for lung cancer."' Other studies with similar findings were reported during the 1980s, and by 1986 the Inter- national Agency for Research on Can- cer,' the National Research Council,' and the US Surgeon General" had con- cluded that passive smoking caused lung cancer in never-smokers. In reach- ing their conclusions, both the Interna- tional Agency for Research on Cancer and the Surgeon General's Report em- phasized the biologic plausibility of the epidemiologic evidence, assuming that there is no threshold of exposure for respiratory carcinogenesis and thus any exposure conveys some risk. In contrast to the strong and causal associations of active smoking with lung cancer, the risks of exposure to environ- mental tobacco smoke found in epidemi- ologic studies have been lower, and methodologic problems have been dis- cussed as an alternative explanation to causality for the association.' The ex- tent of the lung cancer risk to never- smokers caused by exposure to environ- mental tobacco smoke has been particularly contentious, largely be- cause nonsmokers in public buildings and workplaces involuntarily inhale en- vironmental tobacco smoke, and unac- ceptable risks for this exposure would provide a basis for limiting smoking in these locations. The lung cancer risk of exposure to environmental tobacco smoke has been primarily assessed by generalizing the exposure-response re- lationship from the studies of never- smokers exposed to the smoking of their spouses."' Uncertainties are evident in this approach, including the lack of in- formation on total exposure to environ- mental tobacco smoke and the many as- sumptions inherent in deriving a general exposure-response relationship from studies of never-smokers exposed at home. Because of the methodologic difficul- ties of assessing lifetime exposure to environmental tobacco smoke and pre- cisely describing risks that are not sub- stantially elevated, these uncertainties in assessing the lung cancer risk of envi- ronmental tobacco smoke may never be fully resolved, although they remain a 'subject of research. Yet, full resolution would seem unnecessary for the evolu- tion of public policy on environmental tobacco smoke, a carcinogen with a readily controllable source. In the case of environmental tobacco smoke, it would be unfortunate if potentially irre- solvable scientific uncertainties thwart- ed control. Acidic Aerosols/Suifur Dioxide The sulfur dioxide/particulate matter type of pollution, formed primarily as a result of combustion of sulfur contain- ing fossil fuels, represents a widespread form of pollution in industrialized soci- eties. ' The large-scale mid-century pol- lution disasters in Donora, Pa, in 1948' and in London, England, in 1952g',' probably involved extremely high lev- els, by current standards, of acid aero- sols and sulfur dioxide. During the Lon- don fog of 1952, an estimated excess of 4000 deaths occurred, primarily among the elderly and those with a chronic res- piratory disease. A recent reexamina- tion of London mortality data for the years 1963 through 1972 showed a cor- relation between daily mortality and sulfuric acid aerosol levels on the prior day." Statutory regulations promulgated in the early 1970s by the Environmental Protection Agency under the Clean Air Act resulted in significant reductions in levels of total particles and sulfur diox- ide. However, local reductions in pollu- tion were often achieved by the use of tall stacks, particularly for power plants, which resulted in the pollutants being emitted high into the atmosphere, where prolonged residence time per- mitted their transformation into acid species. An emerging concern about the effects of these acidic aerosols has now extended beyond environmental effects on trees and lakes to human health. Although still limited in extent, new epidemiologic data suggest. adverse ef- fects of acidic aerosols. A consistent as- sociation was reported between hospi- tal admissions for respiratory disease in Southern Ontario and daily levels of sul- fates, ozone, and temperature." The Six Cities Study, conducted by Harvard University in six eastern and midwest- ern US cities, demonstrated links be- tween particle exposure and respira- tory disease in children.'s Chronic cough and bronchitis symptoms were associat- ed with hydrogen ion concentration, a measure of acidity, rather than with sul- fa.te levels or total levels of particles. Furthermore, controlled human studies have established remarkable sensitivity in exercising asthmatics to the broncho- constrictor effects of sulfur dioxide' and acidic aerosols" at concentrations sim- ilar to those at higher outdoor levels. Data showing that acidic aerosols are a widespread form of pollution and the emerging health evidence have led to new research in the United States and elsewhere. While the air pollution epi- sodes earlier in the century provided clear and dramatic evidence that acidic aerosols can increase mortality, the present levels of exposure have prompt- ed questions concerning more subtle ef- fects on mortality and morbidity. Pro- viding certain answers to these questions is a difficult challenge for the scientific community. Epidemiologic studies are limited by the difficulty of measuring exposure and of singling out the effect of acidic aerosols from other factors, particularly for such nonspecif- ic health effects as increased symptoms and reduction of lung function. Con- trolled exposures of volunteer subjects provide information concerning short- term effects, but this approach cannot fully represent the exposures sustained in the community. Nonetheless, the regulatory appara- tus turns to the results of epidemiologic and human toxicologic research as a ba- sis for policy. On a 5-year cycle, the Environmental Protection Agency re- evaluates the "science" to either modify or support the current sulfur dioxide and particulate matter standards. Thus, acidic aerosols may be eventually listed for regulation. ' Oxidant Gases Ozone and nitrogen dioxide (NO) are oxidant gases that contaminate outdoor air in many urban and industrial loca- tions. Ozone is one component of the ~ pollution mixture commonly referred to ~ as "smog," and its concentration is used ~ as an index of the degree of smog pollu- n; tion. Indoor environments may be con- © taminated by oxidants from outdoor air (D and by Nflz produced indoors by com- ~ bustion appliances, such as gas stoves ~ and space heaters. At high concentra-_ tions, oxidants cause extensive lung in- ~ jury, including pulmonary edema and ~ bronchopneumonia in animals and in hu- mans"°; however, effects at levels cur- rently measured in outdoor and indoor air in the United States have been diffi- JAMA, August 7, 1991-Vol 266, No. 5 The Environment and the Lung -Samet & Utefl 673

cult to characterize. A better under- standing of the effects of oxidant pollut- ants is extremely important; despite extensive control efforts, more than one half of the US population still lives in communities where the National Ambi- ent Air Quality Standard for ozone is exceeded,SZ and many homes with gas stoves have NO, levels that approach the standard for outdoor air.' In fact, although outdoor air contamination by oxidants has been the subject of sub- stantial research and of great regula- tory concern, pollution of indoor envi- ronments by N02 has been recognized as the predominant determinant of per- sonal exposure in most locations.' Investigating the health effects of the oxidant pollutants requires muitidisci- plinary research involving toxicologic and epidemiologic approaches.' For ex- ample, studies show that NO, exposure increases the frequency and severity of respiratory tract infections in animals'; we are conducting research to test the hypothesis that NO, exposure also in- creases the incidence and severity of respiratory tract infections in humans. In laboratory studies, volunteer sub- jects are exposed to NO, in a chamber, cells are obtained from the respiratory tract by bronchoalveolar lavage, and the ability of the cells to kill virus is assessed. We recently reported that al- veolar macrophages obtained from healthy volunteers after a 3-hour con- tinuous exposure to 1120 µglm3 of NO, inactivated influenza virus in vitro less effectively than cells collected after air exposure.'s In an epidemiologic study addressing this same hypothesis, over 1000 infants have been followed up from birth with prospective observation for respiratory tract illnesses and monitor- ing of their homes for NO,. ' Persistent questions concerning the long-term effects of residence in smog polluted locations, such as Southern California, will probably require large epidemiologic studies and further stud- ies involving the exposure of volunteers to describe the range of susceptibility and to address the mechanisms of toxic- ity. While the costs of this research may be high, exposure to oxidant pollutants is widespread, and the exposures of a substantial proportion of our population may be associated with adverse effects. CONCLUSIONS Although we have selected a few of the contaminants that cause environ- mental lung disease, many other agents with a similar array of changing con- cerns with regard to disease risk and safety could be listed; for example, sili- .ca, the cause of silicosis, is a suspect carcinogen at contemporary occupa- tional levels of exposure, and concern has been raised about the human carci- nogenicity of man-made fibers. We sug- gest that the lessons to be learned from the example agents may be generalized to other pollutants. Active cigarette smoking, occupa- tional asbestos exposure, radon in un- derground mines, and high levels of acidic aerosols were remarkably strong causes of disease under the exposure conditions originally investigated. In fact, cases of disease caused by expo- sure to these agents were initially iden- tified through descriptive case series rather than more formal epidemiologic investigation. The subsequent re- search, both epidemiologic and toxico- logic, was successful in establishing causal exposure-disease associations and informative in characterizing expo- sure-response relationships. However, current concerns over lower concentra- tions of these same agents cannot be so readily answered, nor can they be an- swered with sufficient certainty to sat- isfy all interested parties, who poten- tially include not only the general public but also involved manufacturers, par- ties to litigation, environmental groups, and regulators. , The technique of quantitative risk as- sessment is increasingly applied to gauge the risks of environmental pollut- ants (Table 1). However, the frequent focus on the final risk projection-for example, stating that radon causes 10 000 to 20 000 cases of lung cancer an- nually in the United States-may inap- propriately heighten debate over the projected numbers, even though the numbers are produced by a simplistic and mathematical representation of complex biological processes. More- over, controversy concerning uncer- tainties in risk projections may detract from less ambiguous research findings. For example, radon is an established carcinogen, although any projection of the risks of indoor radon is subject to diverse uncertainties. Because malig- nancy fits more readily into a risk as- sessment framework than nonmalig- nant outcomes, emphasis by regulatory agencies appears to unduly weight ex- posures causing cancer. In using quanti- tative risk assessment to manage risks, the difficulties of communicating risks may further limit the capability of achieving public health goals.57 It should be recognized that steps can be taken to reduce risks of environmen- tal lung disease despite the types of con- troversies and points of uncertainty that we have considered. For some pol- lutants, the individual can reduce risks. For example, prevention and cessation of smoking control the hazards of both active and passive smoking. Radon con- centrations in homes can be measured inexpensively, and techniques are avail- able for mitigating and avoiding unac- ceptable radon concentrations. Guid- ance is available for other indoor air pollutants, including asbestos. For oth- er pollutants, only national policy and regulation can reduce risk. For exam- ple, reduction of levels of acidic aerosols or of ozone can be effected onlyby multi- faceted regulatory strategies directed at sources. Tremendous progress has been made across the century in understandingand preventing environmental lung dis- eases. The shift of our concern to mor- bidity at lower levels of exposure and more subtle effects on mortality paral- lels strong trends of declining levels for some pollutants. Certain diseases, eg, asbestosis and silicosis, are entirely preventable and the occurrence of new cases is now regarded as a sentinel event, signaling an unacceptable expo- sure. Toxicologic studies have provided many new insights into effects of envi- ronmental agents on the lung, although much remains to be learned about basic mechanisms of toxicity. The present emphasis on risk assessment and risk reduction, which raises many uncer- tainties and new questions, should not detract from these past accomplish- ments. The scientific community has been challenged by difficult questions, some of which may never be answered with complete certainty. Nevertheless, it is the results of research that have shifted our public health emphasis to concerns about lower and lower levels of exposure. We anticipate that research on the environment and the lung will continue to support the evolution of public policy, while raising even more difficult questions relevant to public health. This study was supported by grant ESO2679 from the National Institute for Environmental Health Sciences; contracts 87-4 and 88-8 from the Health Effects Institute, an organization jointly funded by the US Environmental Protection Agency (Assis- tance Agreement X-812059) and automotive manu- facturers; and grant DE-FG04-90ER60950 from the US Department of Energy, Office of Energy Research. References 1. National Research Council, Committee on the Institutional Means for Assessment of Risks to Public Health. Risk Assessment in the Federal Government: Managing the Procesa Washington, DC: National Academy Press; 1983. 2. Lowrance WW. Of Acceptable Risk: Science and the Determination of Safety. Los Altos, Calif: William Kaufmann Inc;1976. 3. Ute11147.I, Samet JM. Environmental]y mediat- ed disorders of the respiratory tract. Med Clin North Am. 1990;74:291-306. 4. Lave LB. The role of quantitative risk assess- ment in environmental regulations. In: Hoel DG, Merrill RA, Perera FP, eds. Risk Quantitation 674 JAMA, August 7, 1991-Val 266, No. 5 The Environment and the Lung-Samet & Utell

and Regulatory Policy. New York, NY: Cold Spring Harbor Laboratory; 1985:3-9. 5. AmesBN, MagawR, GoldLS. Rankingpossible carcinogenic hazards. Science. 1987;236:271-280. 6. Graham JD, Green LC, Roberts MJ. In Search of Safety: Chemicals and Cancer Risk. Cambridge, Mass: Harvard University Press; 1988. 7. Nero AV Jr. Radon and its decay products in indoor air: an overview. In: Nazaroff WW, Nero AV Jr, eds. Radon and Its Decay Products in Indoor Air. New York, NY: John Wiley & Sons Inc; 1988:1-51. 8. Samet JM. Radon and lung cancer. J Natl Can- cer Inst. 1989;81:745-757. 9. Harting FH, Hesse W. Der Lungenkrebs, die Bergkrankheit in den Schneeberger Gruben. Vier- teljahresschr Gerichtl Med 0eff Sanitaetswes. 1879;31:102-132, 313-337. 10. Morgan MV, SametJM. Radon daughter expo- sures of New Mexico U miners, 1967-1982. Health Phys.1986;50:656-662. 11. Lubin JH, Boice JD. Estimating radon-in- duced lung cancer in the U.S. Health Phys. 1989;57:417-427. 12. Puskin JS, Nelson CB. EPA's perspective on risks from residential radon exposure. JAir Pollut Control Assoc. 1989;39:915-920. 13. National Research Council, Committee on the Biological Effects of Ionizing Radiations. Health Risks of Radon and Other Internally Deposited Alpha-Emitters: BEIR IV. Washington, DC: Na- tional Academy Press; 1988. 14. National Research Council, Panel on Dosimet- ric Assumptions Affectingthe Application of Radon Risk Estimates. Comparative Dosimetry of Radon in Mines and Hornes. Washington, DC: National Academy Press; 1991. 15. Abelson PH. Uncertainties about health ef- fects of radon. Science. October 19, 1990;250:352. 16. Samet JM, Stolwijk J, Rose SL. Summary: international workshop on residential Rn epidemi- ology. Health Phys. 1991;60:223-227. 17. Lubin JH, Samet JM, Weinberg C. Design is- sues in epidemiologic studies of indoor exposure to radon and risk of lung cancer. Health Phys. 1990;59:807-817. 18. Lubin JH. Methods for the analysis of radon- exposed populations. Yale JBiol Med. 1988;61:195- 214. 19. Selikoff L7, Greenberg M. A landmark case in asbestosis. JAIfIA. 1991;265:898-901. 20. Murray R. Asbestos: a chronology ofits origins and health effects. Br JInd Med. 1990;47:361-365. 21. National Research Council, Committee on Nonoccupational Health Risks of Asbestiform Fi- bers. Asbestiform Fibers: Nonoccupational Health Risks. Washington, DC: National Academy Press;1984. 22. US Consumer Product Safety Commission. A Report by the Chronic Hazard Advisory Panel on Asbestos. Washington, DC: US Consumer Product Safety Commission Directorate for Health Sci- ences;1983. 23. Mossman BT, BignonJ, Corn M, Seaton A, Gee JBL. Asbestos: scientific developments and impli- cations for public policy. Science. 1990;247:294-301. 24. Corn M, Crump K, Farrar DB, Lee RJ, McFee DR. Airborne concentrations of asbestos in 71 school buildings. Regul Toxicol Pharmacol. 1991;13:99-114. 25. Smoking and Health: Report of the Advisory Committee to the Surgeon General. Washington, DC: US Dept of Health, Education, and Welfare; 1964. DHEW publication (PHS)1103. 26. Reducing the Health Consequences of Smok- ing: 25 Years of Progress: A Report of the Surgeon General. Washington, DC: US Dept of Health and Human Services; 1989. DHHS publication (CDC) 89-8411. 27. Cameron P, Kostin JS, Zaks JM, et al. The health of smokers' and nonsmokers' children. JAl- lergy. 1969;43:336-341. 28. Harlap S, Davies AM. Infant admissions to hospital and maternal smoking. Lancet. 1974;1:529- 532. 29. Tager IB, Weiss ST, Rosner B, et al. Effect of parental cigarette smoking on the pulmonary func- tion of children. Am JEpidemiol. 1979;110:15-26. 30. Hirayama T. Nonsmoking wives of heavy smokers have a higher risk of lung cancer: a study fromJapan. BMJ. 1981;282:183-185. 31. Trichopoulos D, Kalandidi A, Sparros L, et al. Lung cancer and passive smoking. Int J Cancer. 1981;27:1-4. 32. IARC1988: IARC Monographs on the Evalua- tion of the Carcinogenic Risk of Chemicals to Hu- mans: Tobacco Smoking. Lyons, France: World Health Organization, International Agency for Re- search on Cancer;1986;38. 33. National Research Council. Environmental lbbacco Smoke: Measuring and Assessing Health Effects. Washington, DC: US Dept of Health and Human Services, National Academy Press; 1986. 34. The Health Consequences of Involuntary Smoking: A Report of the Surgeon General. Roek- ville, Md; US Dept of Health and Human Services; 1986. DHHS publication (CDC) 87-8398. 35. Lee PN. Misclassification of Smoking Habits and Fassive Smoking. Berlin, Germany: Springer Verlag;1988. 36. Rylander R. Environmental tobacco smoke and lung cancer. N Engl J Med. 1990;323:834. 37. Wu AH, Samet JM. Environmental tobacco smoke: exposure-response relationships in epide- miologic studies. Risk Analysis. 1990;10:39-48. 38. Repace JL, Lowrey All. Risk assessment methodologies for passive smoking-induced lung cancer. Risk Analysis. 1990;10:27-37. 39. SpenglerJD, Brauer M, Koutralds P. Acid air and health. Environ Sci Technol.1990;24:946-956. 40. Shrenk HH, Heimann H, Clayton GD, at al. Air Pollution in Donora, Pa: Epidemiology of the Unusual Smog Episode of Oetober 1948: Prelimi- nary Report. Washington, DC: Public Health Ser- vice; 1949. Public Health Bulletin 306. 41. Logan WPD. Mortality in the London fog inci- dent. Lancet. 1953;1:336-338. 42. Mazumdar S, Schimmel H, Higgins ITT. Rela- tion of daily mortality to air pollution: analysis of 14 London winters, 1958/1959-197111972. Arch Envi- ron Health.1982;37:213-220. 43. Thurston GD, Ito K, Lippman M, Hayes C. Reexamination of London, England, mortality in relation to exposure to acidic aerosols during 1963- 1972 winters. Environ Health Perspect. 1989;79:73-82. 44. Bates DV, Sizto R. Hospital admissions and air pollutants in Southern Ontario: the acid summer haze effect. Environ Res. 1987;43:317-331. 45. Ware JH, Ferris BG Jr, Dockery DW, Speng- ler JD, Stram DO, Speizer FE. Effects of ambient sulfur oxides and suspended particles on respira- tory health of preadolescent children. Am Rev Re- spir Dis. 1986;133:834842. 46. Sheppard D, Wong WS, Uehara CF. Lower threshold and greater bronchomotor responsive- ness of asthmatic subjects to sulfur dioxide. Am Rev Respir Dis.1980;122:873-878. 47. Bauer MA, Utell MJ, Morrow PE, Speers DM, Gibb FR. Inhalation of 0.30 ppm nitrogen dioxide potentiates exercise-induced bronchospasm in asthmatics. Am Rev Respir Dia. 1986;134:1203- 1208. 48. KoenigJQ, Covert DS, Pierson WE. Effects of inhalation of acidic compounds on pulmonary func- tion in allergic adoiescent subj ects. Environ Health PersPect.1989;79:173-178. 49. Lipfert FW4 Morris SC, Wyzga RE. Acid aero- sols: the next criteria air pollutant? Environ Sei Technol.1989;23:1316-1322. 50. National Research Council, Subcommittee on Nitrogen Oxides, Committee on Medical and Bio- logical Effects of Environmental Pollutants. Nitro- gen Oxides. Washington, DC: National Academy Press; 1976. 51. Lippmann M: Health effects of ozone: a critical review. J Air Pollut Control Assoc. 1989;39:672- 695. 52. Samet JM, Utell MJ. The risk of nitrogen diox- ide: what have we learned from epidemiological and clinical studies? Tozicol Ind Health. 1990;6:247- 262. 53. SpenglerJD, Duffy CP, Letz R, et al. Nitrogen dioxide inside and outside 137 homes and implica- tions for ambient air quality standards and health effects research. Environ Sci Technol. 1983;17:164-168. 54. Morrow PE. Toxicological data on NO2: an overview. J 3bxicol Environ Health. 1984;13:205- 227. 55. Frampton MW, Smeglin AM, Roberts NJ Jr, Finkelstein JM, Morrow PE, Utell W. Nitrogen dioxide exposure in vivo and human alveolar macro- phage inactivation of human influenza virus in vi- tro. Environ Res. 1989;48:179-192. 56. Samet JM, Lambert WE, Skipper BJ, et al. A study of respiratory illnesses in infants and NO2 exposure. Arch Environ Health. In press. 57. National Research Council, Committee on Risk Perception and Communication. Improving Risk Communication. Washington, DC: National Academy Press;1989. JAMA, August 7, 1991-Vol266, No. 5 The Environment and the Lung-Samet & Utell 675