Document kmy29rdkp3b7xxGXeJm67Q8nV

.9 Acwce* For Publication la ENVIRONMENT INTERNATIONAL. Pergamon Press, circa April, 1985. J.L. Repace* D.S. Environmental Protection Agency Washington, DC 20460 A.H. Lowrey* Naval Research Laboratory Washington, DC 20375 l / SPI-00203 ABSTRACT This work presents a quantitative assessment of nonsmokers' risk of lung cancer from passive smoking, The estimates given should be viewed as preliminary and subject to change as Improved research becomes available. It Is estimated that U.S. nonsmokers are exposed to from 0 to LA milligrams of tobacco Car per day, and Chat Che typical nonsmoker Is etrposed Co 1.4 milligrams per day. A phenomeno logical exposure-respoase relationship is derived, yielding 5 lung cancer deaths per year per 100,000 persons ex>osed, per milligram dally tar exposure. This relationship yields lung cancer mortality rates and mortality>ratios for a D.S. cohort which are consistent to within 5Z with the results of both of Che large prospective epidemiological studies of passive smoking and lung cancer In the U.S. and Japan. Aggregate exposure to ambiedt tobacco smoke Is estimated to produce about 5000 lung cancer deaths per year In t'.S. nonsmokers aged 35 years, with an average loss of life expectancy of 17 + 9 years per fatality. The estimated loss of life expectancy for the most-exposed passive smokers appears to be about 2/3 of that reported for pipe smokers and 1/|2 of that for cigar smokers. Mortality from passive smoking is estimated to be about two orders of magnitude higher chan that estimated for carcinogens currently regulated as hazardous air pollutants under the federal Clean Air Act. SPI-00204 1. INTRODUCTION Exposure of nonsaokers to Indoor air pollution from tobacco smoke (also known as Involuntary or passive smoki ag) has recently become a public health concern (USSG, 1982) for several reason such exposure Is widespread (Repace and Lowrey 1980; Friedman, et al. 1983); s tudies of the effects of tobacco smoke on smokers worldwide have Implicated It as the most Important cause of lung cancer (USSG, 1982; Doll and Peto, 1981); existence of a threshold for carcinogenesis Is doubtful (USSG 1982; IRLG, 1979; USEPA, 1979a; IARC, 1979; Pitot, 1981), and there Is suggestive new evidence of lung cancer (and ot i<er serious health effects) In nonsmokers exposed to ambient concentrations of to 3acco smoke. (Trichopoulos, 1981; 1983; Hlrayama, 1981a; 1981b; 1983a; 1983b; Gar :inkel, 1981; Correa et al., 1983; Knoth et al., 1983; Glllls et al., 1983; Koo, et al., 1983; Rabat and tfynder, 1984; Miller, 1984; Sandler, et al., a; b. In press > There are three Important fractions of tobacco smoke: mainstream smoke, which the smoker Inhales direct y Into the lung; exhaled mainstream smoke, that fraction of the mainstream smokfe which is not retained In the lungs of the smoker, and sidestream smoke, that frac Ion of tobacco smoke emanating directly from Che burning end of the cigarette in o the air. Nonsmokers are commonly exposed to tobacco combustion products in diluted sidestream and exhaled mainstream tobacco smoke from cigarettes, cigars, md pipes (Repace and Lowrey, 1980). Tobacco smoke contains 60 known or suspect ca cinogens, including 51 in the phase containing particulate matter; the carcino^enic activity of tobacco smoke appears to require this phase(USSG, 1982). Animal bioassays indicate that sidestream tobacco tar is more carcinogenic per unit weigji t than mainstream tar (USSG, 1982). For public health purposes, It will be assumed that mainstream and sidestream smoke have similar human carcinogenic poteh cy. In his 1982 report on cancu r and smoking (USSG, 1982), the U.S. Surgeon General asserted chat despite tjve incompleteness of the evidence, nonsmokers SPI-00205 -2- should avoid exposure co second-hand smoke Co Che extent possible, a risk-management judgement supported by Che World Health Organization and Che National Academy of SdMces (WHO, 1979; NRC, 1981] This raises Che question df whether Che quantity of tobacco car Co which the average nonsmoker Is exposed cieates a significant risk of lung cancer. In order co answer this question, a quantitative risk assessment Is first justified and chen performed. Risk assessment is the use of science to define the health effects of exposure of Individuals or populations Co hazardous materials or situations (NRC, 1983): Risk assessments contain some or all of the following four steps: (1). Hazard identification -- the delerminatlon of whether a particular chemical Is or Is not causally linked to certain health effects. (2). Dose-response assessment -- ehe determination of Che relation between Che magnitude of exposure and Che probability of occurrence of the health effects In question. *(3). Exposure assessment -- the determination of the extent of human exposure before or after application of regulatory controls. (4). Risk characterization -- the description of the nature and often the magiltude of Che human risk, Including attendant uncertainty. In ocher words, quantitative risk assessment deals with the question of how much morbidity and mortality an agent Is likely to produce given specified levels of exposure. Typically utilized In the regulation of carcinogens. It is Important because control efforts cannoC proceed without assurance that Che health gains are worth Che costs (Lave, 1983; Albert, 1983). On the basis of such assessments. Informed risk sanagemenc judgements can be made. This work draws upon the ep Ldemlology of lung cancer (USSG, 1982; Pitot, 1981; USSG, 1979; Ives, 1983) and on i idoor air pollution physics (Repace and Lowrey, 1980; L982; NRC, 1981) to produce a ri ik analysis (IRLG, 1979; USEPA, 1979a; Lave, 1983; COST, 1983; Fischoff, et al., 1931; NRC, 1983) In which nonsmokers1 lifestyles are correlated to exposure to ai;:borne Cobacco tar, and incidence of lung cancer. This analysis first reviews estiuates of the average exposure of Che general SPI-00206 -3- population to ambient-tobacco sitoke. Second, it reviews studies linking tobaccorelated disease In nonsmokers to exposure-related variations In. lifestyle. Third, lc couples these two feet ors to develop a phenomenological estimate for the aggregate lung cancer risk do the (J.S. nonsmoking population, and develops an exposure-response relationship for the estimation of the risk to the most-exposed. Fourth, It compares the estimated level of lung cancer mortality and resultant loss of life expectancy from pas live smoking to those from cigarette, pipe, and cigar smoking. Fifth, It comparts the predictions of alternate exposure-response relationships with the results of two large prospective epidemiologic studies of passive smoking and lung cancer, and performs a sensitivity analysis. Finally, this work compares the estimated risk from ambient tobacco smoke to that from various airborne carcinogens currently being regulated in the U.S. as hazardous air pollutants, to place the significance of the estimated risk In perspective. VARIATION OF EXPOSURE WITH LIFESTYLE .................... In earlier work (Repace and Lowrey, 1980; 1982; 1983; 1984; Repace, 1981; 1982; 1983; 1984; In press; Repac e et al., 19.80; 1984; Bock et al., 1982) factors affecting nonsmokers' exposures t o tobacco smoke were studied, and field surveys of the levels of respirable particles were conducted Indoors and out. In both smoke-free and smoky environments. This work established that ambient tobacco smoke Imposed significant air pollution burdens on nonsmokers, and, using control< led experiments (Repace and Lowre;r, 1980; 1982; 1983), a model was developed to estimate those exposures. Thl.t model predicts chat the exposure of U.S. nonsmokers ranges from 0 to 14 mg of cigarette tar per day (mg/d), depending upon the nonsmoker's lifestyle. As derived In Appendix A and shown in Table 1, the SPI-00207 average population azposure for adults of working age, averaging over the work and hoae microenvironments, la aliout 1.43 mg/d (Repace and Lowrey, 1983) with an 86! expoaure probability. Table 1, derived frota the model la appendix A, eatlaataa probability-weighted exposure to the particulate phase of ambient tobacco smoke for a typical U.S. adult nonaaoker. Exposures received In other (Repace et al., 1980) indoor mlcxoenvlronaenca, outdoors, and in transit, which account for the remaining 12! of people's time, were omitted. Table 1 is derived from considerations that ambient concentrations of tobacco tar have been found to be directly proportional to the smoker density and inversely propor tional to the ventilation rate.(Rjspace and Lowrey, 1980) Ventilation rate tables given by A5HRAZ (1981), can be usad to estimate both the range in ventilation rate (from the design mechanical :rates) and smoker density (from the design occupancies), and thus upper and lower bounds and average concentrations for model workplace and home mlcroentf ronments can be estimated. Table 1 suggests that individuals receiving exposure both at home and at work constitute a high exposure group, with the workplace appearing four times as strong a source of exposure as the home; the reason for this differencial Is the generally higher occupancy (l.e., smoker density) encountered in the workplace (Repace and Lowrey, 1982, ASHRAZ, 1981). This estimate of exposures represents a modeled weighted average taken over the entire population. Including those who are not exposed. Jarvis and Russell (In press), In a study of urinary cotlnlne (a nicotine metabolite) In a sample of 121 self-reported nonsmokers, state that only 12! of subjects had undetectable cotlnlne levels, despite nearly 50! reporting no passive smoke exposure. Macsukura (1984), In a study of 472 nonsmokers, examined the relationship of urinary cotlnlne to the smokiness of their environment, and found that nonsmokers who lived or worke<i with smokers had higher cotlnlne levels than SPI-00208 -5- choa who did noc. Macsukura ec al (1984) also found chac eoclnlne levels increased wlch cha number of smokers p resenc In Che home and Che workplace, although a none of the differences were sea dscically significant except the lowest urinary coelolne level of Che aonsmokers who were noc exposed co cobacco smoke In Che home or Che workplace. These sc idles respectively Illustrate Che widespread exposure of nonsmokers Co aablen|c Cobacco smoke, and Che relaclve Importance of Che domesdc and workplace micro(environments In such exposures. EPIDEMIOLOGICAL EVIDENCE FOR THE VARIATION OF RISK WITH LIFESTYLE PULMONARY EFFECTS Vhlee and Froeb (1980) eva|luaced Che effect of various degrees of long-cerm (>20 yrs) workplace exposure co tobacco smoke on 2100 healchy middle-aged workers., Of Che workers, 83Z held profess|t onal, managerial, or technical posldons, while che remaining 17Z were blue coll ir workers. Relaclve co ehose noc exposed ac home or ac work, passive smokers of b) ch sexes suffered statistically significant declines In mid- and end-explrac jry flow races which averaged abouc 13.5 percenc and 22 percent respectively, and did noc differ significancly from Che values measured In noninhaling or light smokers of cigarettes, pipes, and cigars. They concluded Chac chronic exposure :o Cobacco smoke In Che work environment reduces small airways function Co Che sa: ae extent as smoking 1 Co 10 cigarettes per day. Kauffmann ec al. (1983) comDi ared pulmonary function In about 3800 people In France: 849 male "true" nonsmoke :s (defined as chose noc exposed at home) 165 male passive smokers (defined as Chose exposed at home), 826 female "true" nonsmokers, and 1941 female passive smokers. The authors restricted che analysis to subjects aged 40 years or older (i.e., Co Chose who had been exposed for 15 or more years to smoking by cheir s jouses) and who were living In households with no persons over che age of 18 years except their spouses. They found that nonsmoking subjects of either sex whose spoilses were current smokers of ac least 10 g (about SPI-00209 410 cigarette*) of tobacco a day had mid-expiratory flow races averaging 11.5* lover than those married to nons aokers. For women in social classes with the highest percentage of paid work, the effect of workplace smoking appeared to confound the effect of passive s soklng at home. However, la the large subgroup of voaen without paid work (l.e , not exposed to workplace smoking), a dear dose-respoase relationship to am sunt of husbands' smoking was observed* They conduded that women living vlch heavy smokers appeared to have the same reduc dons In mid-expiratory flow rates as light smokers, and that after 15 years exposure In the home environment passive smoking reduces pulmonary function. A third study by Kasuga (1985) of urinary hydroxyproline-to-creatinlne (HOP-r) ratios as a function of passive nmoklng status shoved that HOP-r levels In nonsmo king vives and children varied In a dose-response relationship with husbands and parental smoking habits, when ad; usted for pre-existing respiratory disease, Easuga (1983) asserts that HOP-r serves as a marker to detect deleterious active or passive smoking effects on the lung, before and after the manifestation of dlnlcal symptoms, and that minary HOP-r in light smoking women Is almost equivalent to HOP-r In nonsmoking wives with heavy smoking husbands. These three epidemiologic studies provide evidence that variations In the ex posure of adult nonsmokers to ambient tobacco smoke at home and particularly, at work, can produce observable pulmonary effects. Like effects have been observed in children exposed at home (Tagcr et al., 1983). CANCZR Thirteen epidemiologic studies have explicitly examined Che lung cancer risk Incurred by che nonsmoking spouses of dgarecte smokers. In all but one study, Che only exposure variable was che st rength of che spouse's smoking hablc. The studies were conducted in Greece (Trichopoulos et al., 1981; 1983), Japan (Hirayama, 1981a; 1981b; 1983a; 1983b), che J.S. (Garflnkel, 1981; Correa, et al., 1983; Kabac and Vynder, 1984; Miller, U84; Sandler, et al. a and b, in press), Germany SP1-00210 -7- (Kaeth at al.f 1983), Scotland (jcillls, et al., 1983), and Hong Kong (Chan and Fung, 1912; Coo, et al., 1983). In the Greek study, Trichopjoulos et. al. (1981, 1983) used Che case-control technique: involuntary exposure to cigarette smoke as measured by the husbands'' daily consumption was found to 1icrease the average risk of lung cancer by a factor of 2.4 (p<0.01) when 77 ling cancer patients were compared to 225 controls, and a dose-respoase relationship vas observed. Divorce, remarriage, husband's death, and change in smoking hab: ts were considered. In the Japanese study (1966-;981) of lung cancer in 91,540 nonsmoking women, Hlrayaaa (1981a, 1981b, 1983a, 1183b) used the prospective technique: relative to those women not exposed at home (controls), involuntary exposure of wives of smokers was found to Increase the average risk of lung cancer by a factor of 1.78 (p<0.001), where che exposure vas also estimated from husbands' daily consumption. The annual lung cancer death (LCD) rate in the controls was 8.7 per 100,000. -Hirayama found that the exposed wives experienced an average annual increase in per 100,000, with a range of from 5.3 to 9.4 per 100,000, in a dose-respoase relationship depending upon che degree of the husband's smoking. Hirayama found further that Che risk of lung cancer death in nonsmoking women increased both v. th Che time of exposure and number of cigarettes smoked daily by che husband. Hiri.yaaa also reported a factor of 2.9 (_+ 0.3, at the 952 conf. level) for increased risk of lung cancer in 1010 nonsmoking husbands with smoking wives. Mors recently, Hirayama extended his earlier work to suggest Increased risk of nasal sinus cancer, and ischemic heart disease in passive smokers, and evidence of decreased lung cancer risk in nonsmoking wives of exsmokers. With respect to canter of Che para-nasal sinuses in nonsmoking wives (n28) , Hirayama found standardized mortality ratios of 1.00, 2.27, 2.56, and 3.44 when husbands were non-smokers, smokers of 1-14, 15-19, and >20 cigarettes SPI-00211 -8- per day respectively (p - 0.01). For Ischemic heart disease, risk elevations for nonsmoking wives (n**494) with the extent of husbands' smoking were reported, with standardized mortality ratios of 1.00, 1.10, and 1.31 when husbands were non-smo kers, smokers of 1-19, and >20 cigarettes per day respectively (p<0.02). For lung cancer, the standardized mortall:y ratio of lung cancer In non-smoking women (n-200) was 1.00, 1.36, 1.42, 1.58, and 1.91 when husbands were non-smokers, ex-smokers, dally smokers of 1-14, 15-19, and >20 cigarettes/day, respectively. In the first D.S. study, Garf\nkel (1981) reported results from an analysis of data collected from the American Cancer Society's (ACS) prospective study of lung cancer risk in 176,739 nonsmoking white women (1960 to 1972) as a function of Involuntary exposure as lndlcate< by their husbands' cigarette consumption. 72Z of the nonsmoking women were mari led to smokers. Three smoking categories were Identified: none, less than a pa< k (20 cigarettes) per day, or greater than a pack per day. Garflnkel reporte< statistically Insignificant risk ratios of 1.00, 1.27, and 1.10 respective!; for the three categories (average 1.20 over the exposed categories). Also r ported were age-standardized death rates, which were respectively 13.8, 12.9, and 13.1 lung cancer deaths per 100,000 person-years for this cohort In 1960-1964, 196 4-1968, and 1968-1972 (average 13.3 per 100,000 person-years for the period 1960- 1972). The death rates were standardized to the dlstrlbutlon of white men and won en combined for the U.S. population In 1965, which decreased the rates for fern ales "slightly'*. More recently, Correa, et al . (1983), studied 8 male and 22 female nonsmoking lung cancer cases and 180 male an d 133 female controls as part of a larger study Including smokers, with 1338 lung cancer cases and 1393 controls, In Louisiana. They reported chac nonsmokers mar rled to heavy smokers had an increased risk of lung cancer, as did smokers whose moch rs smoked. Men wlch smoking wives had a nonsignificant risk ratio of 2.0 compar sd to their counterparts with nonsmoking wives, and women with smoking husbands h id an average risk ratio of 2.07 (p<0.05) compared SP1-00212 -9- to woman vlch nonsmoking husbands. An exposure-response relationship was observed, vlch che peak risk reaching 3.52 (p<0.05). The combined daca for men and women passive smokers was significant (p<0.05) for Che heavier smoking category (>_ 41 pack-years). A third U.S. ease-control study, by Rabat and tfynder (1984), reported on paasive smoking and lung cancer in nonsmokers for 25 male cases and 25 controls, and 53 female cases and 53 controls, where the majority of the patients were from New York City. The controls consisted of patients hospitalized for non smokingrelated diseases, roughly two-thirds being cancer patients. No differences on exposure to passive smoking at hsme or at work were found in the women. However, Che male passive smokers display ed a statistically significant (p-0.05) difference In lung cancer (odds ratio 1.6) relative to the non-exposed group, A fourth U.S. study by Hill tr (1984) of mortality from all forms of cancer in 123 nonsmoking women (only 5 lung cancer cases) as a function of husband's smoking history reported a non-s tgnlflcant odds ratio of 1.4 for all women (p0.15) for women whose husbands smoked relative to chose who did not, and when employed women were excluded Che odds ratlo Increased to 1.94 and was statistically signi ficant (p<0.02). A fifth U.S. study by Sandltr, et al. (in press, a) also examined mortality from all forms of cancer related to passive smoking, in both nonsmokers and smokers (231 cases and 235 controls (702 white and 672 female); only 2 cases of lung cancer in nonsmokers) as a funct on of spouses' smoking habits. Cancer risk -- adjusted odds ratio ---(lung, ireast, cervix, and endocrine) among individuals ever married to smokers was 2.0 limes that among Chose never married Co smokers (p<0.01). This increased risk was not explained by confounding individual smoking habits, demographic characteristics, or social class. In a sixth U.S. study, Sandier, ec al. (in press, b) examined cancer risk in adulthood in 197 cases and 22il controls, 662 female, from early life exposure SPI-00213 -10to parents' smoking. They found hat mothers' and fathers' smoking were both associated with risk for heaatopo: etlc cancers (Hodgkin's dlsesse, lymphomas, and leukemias), and a doee-respona e relationship was seen for the latter two. Th odda ratio for hematopoietic csneers Increased from 1.7 when one parent smoked, Co 4.6 when both smoked (f d0.001). In the first of two studies i rom Hong Kong, Chan and ?ung (1982) found a lover Incidence of passive smoking among 34 female lung cancer cases (40.5Z) than among 66 female controls (47.5Z). All patients and controls were Interviewed conconcerning their smoking habits an 1 thoee of their spouses, their cooking habits, Including types of cooking fuel us ed. Histological diagnoses of tumors were obtalned. Controls were taken from arthopedlc patients. In the second Hong Kong study Koo ec al. (1983) studied passive smoking in56 female lung cancer eases and 85 feoale controls. Passive smoking cases had an excess of 3.8 years of passive i imoklng (workplace plus domestic exposures) compared with controls, but the dll ferences were not statistically significant (p 0.069). However, among a subj Toup of 8 marine dwellers, cases had 11.8 years more exposure than controls p - 0.0003). Knoth et al. (1983) reported n a study of 39 nonsmoking German females with lung cancer. 61.52 were found to h ave smoking spouses. The authors seace that this percentage was threefold that expected on the basis of smoking habits of German males. Glllls et al. (1984) reported preliminary results of a study of passive smoking and lung cancer In 91 male cotttrols (n-2) [the numbers In parentheses give the numbers of cases] vlthouc dome* tic passive smoking and In 90 subjects exposed at home (n-4), and In 40- female cotitrols (n-2) and 58 subjects (n-6). No effects of lung cancer were noted In Che fe.sales, but elevated rates of myocardial lnfarccion were reported (risk ratio 3.0] > In Che males, elevated rates of both lung cancer (risk ratio 3.25) and myocardial Infarction (risk ratio 1.45) were reported. Glllls et al. state chat since 1ns uf fl dent time has elapsed since the beginning SPI-00214 -11of th1* study, no firm conclusloni can b drawn relating to the Incidence of cancer or ocher disease*. Thus there ant now a large number of studies providing evidence for Increased risk of lung cancer from Increased exposure to passive smoking. It might be expected Chat sulgroups of the population which proscribe smoking among thslr membership vould have a lover probability of passive smoking, and therefore should also have a lower! incidence of smoking-related disease chan the general nonsmoking population. One such subgroup Is the Chur:h of Jesus Christ of the Latter Day Saints, popularly known as che Mormon Church, vhlch advises against Che use of tobacco. Enstrom (1978) found chat active Mcrmons vho were nonsaokers had standardized mortality rates for lung cancer vhlch were 21Z of those In the general popu lation vhlch includes smokers. This race was found comparable to che race of 19Z for a sample of the O.S. general population "who had never smoked cigarettes.'* Interestingly, however, this result occurred despite the fact that 31Z of che active Mormon cohort were former suckers. This confounding factor was not present for certain subgroups la the following study. Phillips at al. (1980a; 1980b) have studied mortality (1960-1976) In Seventh Day Adventists (SDAs), a religious group vho also follow rigorous proscriptions against the use of tobacco. As with with the Mormons, SDAs have races of mortality from lung cancer and other smoking related cancers Chat are fractions, respectively 21Z and 66Z, of che rates for a demoj;raphlcally comparable group In the general O.S. population (including smokers) (1980a). A sizable subgroup (35Z) of SDAs report prior cigarette use, especially among men (1980b). SDAs appear to be less likely than che general population to be Involuntarily exposed to tobacco smoke, as children or as adults, at home or (n che workplace, because neither SDA homes nor SDA businesses are likely to be places where smoking Is permitted, and because SPl-00215 -12- che greet majority of SDA family and social contacts are among ocher SDAs who da eoc woke (See Appendix C). Phillips ee al. (1980a; 1380b) compared mortality la two demogTaphlcallyimllar groups of Southern Californians: SDAs (from 1960 Co 1976) and non-SDAs (tram 1960 to 1971). In particular, for two select subgroups of each group, Z5,26A SDAs and 50,216 non-SDAs who vers self-reported nonsmokers who never smoked, age adjuseed mortality ra :es were compared for smoking-related and nonsaoking-relaced diseases, table 2 ccmpares age-adjusted lung cancer mortality ratios for two SDA cohorts relative to aonsaokers in Che general population who never smoked. The first cohort consists of all SDA, and Includes chose who never smoked, ex-smokers, and smokers, the firsc row of table 2 gives the mortality ratios relative to Che never-smoked non-SDAs in the'general population. The second row compares the second SDA cohort (chose who never smoked) to the non-SDA who never smoked. The values given are averaged over both sexes. From Table 2 the results show that the non-SDA group of nonsmokers who never smoked (but who were more likely to suffer involuntary exposure to tobacco smoke) had an average lung cancer mortality rate olf 2.4 times Chat of Che never-smoked-SDAs (the group less likely to have suffered such exposure by virtue of their lifestyle), this concludes the review of evidence relating variations in lifestyle to variations in lung cancer risk la nonsmokers. DOES AMBIZHT TOBACCO SMOKE POSE A cWlKOCENIC HAZARD? The International Agency For Research on Cancer (IARC) criteria for causality to be inferred becween exposure and human cancer state that confidence in causality Increases when 1) independent studies agree; 2) associations are strong; 3) dose-response relationships exist, and 4) reduction in exposure is followed by reduction in cancer inciderce (IARC, 1979). SPl-00216 T 1* There ere new 14 stud:,at, covering 6 cultures, indicating e relationship hetveea exposure to aableac tobacco smoke eada incidence of lung cancer. If the studies a e divided into substudies of asn and women, this fields 20 substudies, ell but of which suggested en Increased cancer mortality from passive smoking, end 12 of which attained statistical significance. Moreover, the mortality ratios based on spouses' smoking as an exposure variable, cluster around the value 2.0. Thus, many Independent studies agree. 2. Mainstream tobacco smoke is strongly associated with lung cancer. The U.S. Surgeon General (USSG, 1982) asserts that mainstream cigarette smoke la e major eauae of cancers of the lung, larynx, oral cavity, and esophagus, and is contributory factor for the development of cancers of the bladder, pancreas, and kidney, where Che term eontr butory factor does not exclude the possibility of causality. Both smokers and tonsmokers are exposed to exhaled mainsteam and sidestream tobacco smoke. Sidestream smoke by animal bioassay has been found to be of grester potency than mainstream smoke. 3. Five of the 14 studies reported dose-response relationships between paaelve smoking and lung cancer, Dose--response relatlonshipe between lung cancer and active cigarette smoking show increasing mortality with increasing dosage of smoke exposure, and an Inverse relationship Co age of Initiation (USSG, 1982). Dose-response relationships are al|*o shown for smokers whose smoking habits are like heavy passive smoking (Vynder and Goodman, 1983; Jarvis and Russell, in press) l.s.. In cigarette smokers rho do not inhdle, and In pipe and cigar smo kers, who also are unlikely to inh.ile (USSG, 1982; USSG, 1979). 4. Reductions in lung cancer Incidence for reduction in exposure have been found in all major studies of active smoking. (USSG, 1982) The one study of passive sasoking and lung cancer which examined this question also found a similar result (Hirayama, 1983b). Purther, the co sparison of the' SDA's who never smoked and who SPI-00217 T  -14- should have reducsd exposure relative co Che non-SDA's who never smoked, also appears Co exhibit this effect. On the basis of the IARC criteria, the evidence appears to be sufficient for reasonable anticipation of an Increase in lung cancer mortality from passive smoking. Justifying a quantltlve risk assessment. The significance of the public health risk will now be estimated. ESTIMATION OF TOTAL LCD RISK A TO A PHENOMENOLOGICAL EXPOSURE-RESPONSE RELATIONSHIP A phenomenological exposur{-response relationship Is now derived based on consistency (Hirayaaa,1983b) o evidence provided by studies of lung cancer In nonsmokers and from our exposu :e assessment. The Seventh Day Adventist Study by Phillips ec al (1980a;1980b appears to provide Che best evidence of the mag-; nlcude of the lung cancer effe< :t from passive smoking among*U.S. nonsmokers. A calculation (Appendix C) based on the age-standardized differences In lung cancer mortality rates bel ween SDAs who never smoked and demographlcally comparable nonSDAs who never si loked (age groups 35 to 854-) from Che studies of Phillips ec al. (1980a; 1980b) yields an estimated 4700 lung cancer deaths for the 62.4 million U.S. nonsmoki xs (USDC, 1980) at risk (USSG, 1979) aged >. 35 years. This in turn yields an exposure-response relationship of 7.4 LCDs per 100,000 person-years (4700 LCDj /yr per 62,424,000 persons), in good agreement with the value of 6.8 per 100,C 00 person-years reported In Che Hlrayama (1981) study. To place che estimated mortality in perspective, 4700 deaths was about 5Z of the total annual LCDs, at d about 30Z of Che LCDs in nonsmokers in 1982 (USSG, 1982). The exposure of nonsmokers In Che U.S. population of working age, taken from the model results In Table I, appears to be a weighted average of about 1.43 mg of tobacco tar per day, including the estimated 14Z of che population who receive no exposure at home or work. The carcinogenic risks will be assumed to apply even to retire d persons, whose exposures are reported SPI-00218  L5- to be less then the employed (Fritdmin, et al., 1983), because the risks of lung cancer from smoking decline onlj slowly even with total aessatlon of exposure (USSG, 1982), and because ttle risks of lung cancer Increase exponentially with age (NCI, 1966). Using the statistical risk of 7. LCDs per 100,000, and dividing by the average exposure of 1.43 mg/d, ve es :lmate a phenomenological exposure-response relation appropriate for the general U.S. population at risk, of about 5 LCDs per 100,000 person-years st risk per 1 mg/d nominal exposure. The range In nominal exposure has been estimated to be 0 to 14 mg/day (Repace and Lowrey, 1980) Studies of lui ig cancer and passive smoking across three cultures have shown an an exposure-ri isponse relationship. Thus, Che assumption of an exposure-response relationship is justified, and a linear exposure-response function (Doll and Peco, 1981; IRLG, 1979; USEPA, 1979; Crump ec al., 1976) is assumed. With zero excess risk fi om tobacco smoke for zero exposure, and applying the exposure-response relatl onshlp derived above, with the maximum exposure of 14 mg/d, s maximum risk cif about (14x5) 70 LCDs per 100,000 personyears Is estimated for the moet-expos>sd lifestyle. This lifestyle has been previously typified by that of a nonslaoklng musician who performs regularly In a smoky nightclub and who resides in i small apartment with a chalnsmoker; many ocher scenarios may be drawn. (Repact and Lowrey, 1980) ESTIMATED LOSS OF LIFE EXPECTANCY Reif (1981 a;b) argues that the.re exists a genetically-determined distribu- cion In natural susceptibility to lur i; cancer in people; the effect of exposure to tobacco smoke is to shift this dls tribution coward death at earlier ages. In ocher words, exposure to tobacco smoke produces a loss of life expeccancy One method of presenting risk data in\iolves calculation of the loss of life SPI-00219 16' expectancy, In units of days of life lost per Individual, averaged over the entire population at risk. Vhen the average life-loss Is multiplied by the nuaber of Individuals at risk, the lopact of the hazard on society In personyears of life lost can be assessed More importantly, one can display the age-specific probabilities of death from the hazard, as veil as the average nuaber of years of life lost by the average victim. Appendix C gives the method of calculation. Averaged over all of the popul itlon at risk, (i.e., including those who di of other causes), the average loss of life expectancy from passive smoking is calculated (appendix C) to be 15 days, which Is equivalent to an ultimate loss of 2.5 ollllon person-years of life for the total at-rlsk U. S. population In 1979 over 35 years of age (&2.7 ail ion persons). The escl'mated worst-case loss of life expectancy Is 148 days again averaged over all of the population at risk. The estimated mean life expectancy lost by a passive-smoking lung cancer victim is 17+9 years. How does the calculated averag^ loss of life expectancy for very heavy average loss of life expectancy found in active smokers? The modeled worst-case lifestyle might be reasonably expected to have lesser exposure, and hence lesser risk than active smokers. Table 3, adapted from Cohen and Lee (1979) gives this comparison. The estimated mostexposed lifestyle has about 2/3 the loss of life expectancy of the average pipe smoker, about 1/2 the loss of the average cigar smoker, and 1/150 of that for active cigarette smoking. SPI-00220 -] 7- tSTVUTZ OF AN EXPOSURE-RESPONSE RE1ATIONSHIP BASED ON RISKS IS SMOKERS Aa alternative extrapolated exposure-response relationship Is now derived fro* evidence provided by studies of lung cancer In cigarette smokers. Using the Surgeon General's estimate that J5Z of all lung cancers are due to smoking (USSC, 1982) a current annual LCD race to smokers at risk of about 316 per 100,000 is estimated (see Appendix B). Assuiting a one-hit model (see Appendix B) for extrapolation of the risk (which in this range Is functionally equivalent to the linear assumption that that a milligram of tobacco tar Inhaled by a nonsmoker produces a response equivalent to tha|t In a smoker) yields an estimate of about 0.87 LCDs/100,000 person-years. This corresponds to an exposure-response relationship of 0.6 LCDs/ 100,000 perion-years per mg/d, and an annual aggregate risk estimate of about S55 LCDs per yisar, an order of magnitude lower chan the phenomenological estimate. DISCUSSION OF ALTERNATE EXPOSURE-RESPC NSE RELATIONSHIPS . Ve now speculate on why these two different methods produce such disparate estimates of risk. One possibility is that nonsmokers may have a reduced tolerance to the effects of tobacco smoke Another possibility is a "large dose" effect (Jarvis and Russell, In press) 'thereby exposure to tobacco tar at the lesser doses experienced by nonsmokers produces a greater risk per unit dose chan the greater doses experienced by active smokers, whose lung tissue is saturated by carcinogenic tar. Large dose effects have been observed in cancer induction by ionizing radiation where the dose-response curve has a linear form at low doses, a quadratic upward (positive) cu:rvature at intermediate doses, buc a downward (negative) curvature at high d 3Ses.(NRC, 1980) Downturns in exposureresponse curves of lung cancer in smokers of more chan 40 cigarectes per day have been observed by Doll and Peco (19'8) and Hirayama (1974). SPI-00221 -18- The affect of a laveling-off or doraturn In the exposure-response curve ac large exposures would be to cause a linear model to underestimate the risk when extrapolated (Hoel, et al., 1975; 983; MRC, 1980) over two orders of magnitude to low exposures. A third possibility Is generat:pd by modeling the dose, as opposed to the exposure, of nonsmokers to tobacco smoke. Nonsmokers' exposure is translated Into dose by means of a simple sinfele-compartment model for lung deposition and clearance (Repace, 1983). This model suggests that tar may accumulate on the surface of nonsmokers' lungs to an equilibrium dose an order of magnitude higher than the nominal exposure, Uo a level of about 16 mg per day, due to the long pulmonary residence times for respirable aerosols. If this 16 mg dose, rather than the 1.4 mg exposure, 1tn the operative factor, then the typical passive smoker would have a risk. according to the one-hit model, of about 9 per 100,000, in agreement with the phenomenological estimate. In fact there Is support for this argument from Matiukura's study (1984), which showed that heavy passive smokers had urinary cotlnlne levels comparable to active smokers of less than 3 cigarettes per day, and from Kasuga's study (1983), which also showed that heavy passive smokers lad urinary hydroxyproline levels almost equivalent to that of light smoker: . Moreover, similar observations have been found Indicating that serum thiocya nate (Cohen and Bartsch, 1980) and benzpyrene (Repetto and Martinez, 1974) level: In some passive smokers were comparable to the elevated levels typically found in smokers. Moreover, the simple model we have proposed Ignores the effect of cancer latency. The long latency period for lung cancer Indicates thac childhood passive smoking may be an Important factor affecting risk In adult life: Doll and Peto (1981) have suggested that the effect of passive smoking may be surpri singly large because lifelong expos se may produce a lung-cancer effect four SPI-00222 19- does as great as chat vhlch Is Halted to adult life (recall ehe observation of Sandler et al.(ln press); childhood passive smoking appeared to elevate the can cer risk of adults). As Bonham and Wilson (1981) have shown from a national study of 40,000 children In 1970, 62E came from homes vlth one or more smokers, Indicating that many adults receive ixposure during childhood. SENSITIVITY ANALYSIS Which of the two exposure-responne relationships derived Is more useful in explaining actual epidemiological data? The Garflnkel (1981) American Cancer Society (ACS) study of passive smoklig and lung cancer, vhlch spanned the years 1960 to 1972, reported a standardlzec mortality ratio of 1.20 and an annual lung cancer rate of 13.3 per 100,000 person-years. Of the 176,739 women In the Garflnkel study, 28Z had nonsnoklng husbands, thus, the "controls'* numbered 49,487 and the total "exposed" were 127,252. According to census data (BOC, 1980), female participation ratces In the labor force ranged from 37. IZ in 1960 to 38.8Z In 1965, to 42.8Z in 19370, and 43.7Z in 1975, and was about 80Z of the 1965 level in 1947. Thus, it ippears that about 38Z of the vomen In this study were In the labor force, aid presumably exposed to passive smoking while at work. It is assumed that fo:r both groups of women, control and exposed, 38Z were employed and exposed to ambient tobacco smoke while at work. As Indicated In table 1, typical U.S. nonsmoking adults are estimated to Inhale 1.82 mg of tobacco tar per average day at work and 0.45 mg per average day at home, an exposure ratio of 4:1; this Is because, although domestic and workplace air exchange rates are similar (appeneix A) workplace smoker densities tend to be far higher. Let the assumed basal rate of lung cancer deaths In these women from causes ocher than passive smoking be 8.7 per 100,000 (the age-adjusted race for nonsmoking women married Co n on-smokers in Hirayama's (1981a) study). SPI-00223 23 The Garflnkel (1981) ACS cohere can low be broken down as shown in Table 4a. The Garflnkel (1981) study can analyzed as follows, using Che phenomeno- logical exposure-response relaclonsht p of 5 LCDs/100,000 person-years-mg/d. The lung-cancer deaths per 100.01 )0 contributed by passive smoking are then 2.25 (0.45 x 5) for the hone and 9. 13 (1.82 x 5) for che workplace. Application of these figures co the numbers of e rue and ralnCed controls and working and nonworking exposed women yields, after iddition of che basal risk of 8.7 per 100,000, che estimated races for lung cancer deaths per 100,000 person-years as shown In cable 4b. The rado of risks (all e;^posed:al2 controls) Is chus 1.19. The rado (averaged over husbands' heavy and 1. ghc smoking categories) In che Garflnkel (1981) scudy was 1.20, less Chan a 1 ! difference. The lung-cancer deach race for Che weighted average of Che "exposed and "control" categories Is 13.8 per 100,000. Over che 12 years of che'G'^rflnkel scudy, Che actual race averaged 13.3 per 100,000, a less Chan 4Z dlff erence. In ocher words, chls analysis (Repace, 1984) appears Co explain bo {h che observed lung cancer deach rate and observed risk-ratio of Che Garflnkel ACS cohort. Could this be due co chance? Suppose Instead of 38Z of women In the workforce, that 100Z of women worked. Then ehe rado of risks would be 1. 1:1 , a 6Z difference from Garflnkel's observation, but che annual lung can4er deach race would be 19.42, a 46Z difference. Suppose 0Z of women worlfed. Then che ratio of risks would be 1.26, a 5Z difference from Garflnkel s result, but Che lung cancer deach rate would be 10.32 per 100,000, a 22Z dl ifference from Garflnkel 's observation, Suppose che exposure-response re] adonship of 0.6 LCDs/100,000 person-years per mg/d yielded by extrapolation wi h the one-hit model from the risks in smokers Is used. The lung-cancer de ^ths per 100,000 contributed by passive smoking are then 0.27 (0.45 x .6) for the home and 1.1 (1.82 x .6) for che workplace. Application of these flgv res to the numbers of true and tainted SPt-00224 -21 controls and working and nonworking exposed women yields, after addition of the basal risk of 8.7 par 100,000, the figures shown in Table 4c.* The ratio of risks (all exposed:all controls) Is then 1.03. Compared with the risk ratio In the Garflnkel (1981) study, this Is t 142 difference. The lung-cancer death rate for the weighted average of the "exposed" and "control" categories Is 9.3 per 100,000, a 302 difference from G irfinkel 's result. Finally, using the phenomenologl cal exposure-response relation, the ratio for "all exposed" and "true" controls Is 1.7. Hlrayama's (1981) average risk ratio was 1.78 from passive smoking, a 4.52 difference. Further, If lung cancer risk rate calculation Is performed with the tainted controls Included as an exposed group, the result Is 14.8 per 100,000, compared with Hlrayama's observed 15.5 per 100,000, a 42 difference. In other words, the effect of moving the confounding tainted controls from Garfl nkel's control group Into his exposed group Is to yield results within 52 of Hlrayama's. When the one-hit model Is used, the ratio of all-exposed to true controls Is 1.09, a 382 difference with Hi.rayamh 's ratio. The corresponding lung cancer mortality rate Is 9.45, a 392 differente with Hlrayama's result, Thus, on the basis of this sensltl pity analysis. It would appear that the phenomenological exposure-response rel atlonshlp Is better able to describe the results of the Garflnkel (1981) study than the one-hit model, and In addition, also appears to be able to explain qua ititatively why the two large prospective studies of passive smoking and lung ca acer yielded different results. SPI-00225 -22- COHPARISON OF THE ESTIMATED RISK OF E ASSIVE SMOKING WITH THOSE-OF HAZARDOUS All POLLUTANTS CURRENTLY UNDER RECULA!noN Although Che quantitative estimate:s presented should be regarded as preliminary ead subject to confirmation by further research, the evidence suggests that passive smoking appears to be responsible for about one-third of the annual lung cancer mortality among U.S. nonsmokers. To place these estimates In perspective, cable 5 gives a comparison of the estimated risk of passive smoking to risks estimated by the U.S. Environmental Protection Agency for the carcinogenic hazardous air pollutants currently regulated under lection 112 of the Clean Air Act (SCEP,1977). As table 5 demonstrates, passive smok|l ng appears to pose a public health risk larger than the hazardous air pollute:hts from all regulated Industrial emissions combined. ACKNOWLEDGEMENTS: The authors are grateful to R.L. Phillips of the Department of Blostatlstlcs and Epidemiology of Loma LI nda University, Loma Linda, CA 92350, for tabulations from his published studl e: i of mortality in members of the Seventh Day Adventist Church. We also thank B. Fisch^)ff, H. Gibb, J. Horowitz, D. Patrick, G. Suglyama, W. Ott, and J. Wells for useif'ul discussions, and J. DeMocker for assistance with computer programming. TThe views presented in this article are tjhose of the authors, and do not necessarily reflect the policies of the agencies name SPI-00226 Al. APPENDIX A; MODELING EXPOSURE OF NONSMOKING U.S. ADULTS TO AMBIENT TOBACCO SMOKE INTRODUCTIM Lifestyle is Ch Integrated way of life of an Individual; aspects of lifestyle which will be considered here have ti> do with the amount of time a non-smoker spends in contact with smokers, and therefore with their effluent. Exposure of nonsmokers to tobacco smoke might be expected to be common in the U.S. because one out of three U.S. adults smokes Cigarettes at the estimated rate of 32 per day (Repaee and Lovrey, 1980), while an additional one oat of six smokes cigars or pipes, and because Indoor air pol utlon from tobacco smoke persists in Indoor environments long after smoking ceases (Repaee and Lovrey, 1980; 1982). Earlier work (Repace and Lowrey 1980) presented a model of nonsmokers' exposure to the particulate phase of ambient smoke which vas supported by control led experiments and a field survey of the levels of respirable particles Indoors and out, in both smokefree and smoky environments. This work, which established that ambient tobacco smoke Imposed s: gnifleant air pollution burdens on nonsmokers, was extended by later work (Repace and Lovrey, 1982) which further demonstrated the predictive power of this model, The model predicts a range of exposure of from 0 to 14 mg of cigarette aerosol per day, depending upon the nonsmoker's lifestyle. Exposures of prototyplca. nonsmokers were modeled, but no attempt was made to estimate the average population exposure. Concentrations of ambient tobacco smoke encountered by nonsmoknrs can be approximated by equilibrium values which are determined by the ratio of the average smoker density to Che effective ventilation rate (Repace and Lowrey, 1980; 1982), and thac in practice, design ventilation standards based on occupancy were useful surrogates for effective ventilation rates. On the average, ,l characteristic value of this ratio can be assigned to a particular microenvironmental class, e.g., homes, offices, restau- SPI-00227 A2. rants, ace. (Rapace, at al., 1980) Therefore, the average daily exposure of Individuals can beesClmated from the time-weighted sum of concent radons encounCared In various microenvironments eontalnlng smoke. (Oct, In press; NRC, 1981; Szalai, 1972; Repace, at al., 1980) EXPOSURE AND LIFESTYLE It Is important to realize that most persons' lifestyles are such chat chey spend nearly 90Z of Chelr time In jist tvo microenvironmental classes, chus affording a great simplification of exposure modeling. Szalal (1972), as part of The Multinational Comparative Time Sudget Research Project, vhleh studied the habits of nearly 30,000 persons In L2 countries (1964-1966), has compiled data reporting the average time spent In various locations or microenvironments. The data for 44 cities la the U.S., as analyzed by Ott (In press) are summarized In Table Al (see also NRC, 1981). Table Al shows that D. S. urbAn people spend an average of 88Z of chelr time In just tvo microenvironments: in homes and In workplaces. Moreover, employed persons In Che U. S. cities are estimated to spend only 3Z of the day outdoors while housewives spend only 2Z outdoors (Ott, in press; NRC, 1981). Assume chat these values ars representative of (he entire population. (In 1970, approximately three fourths of the population was urban) (USDC, 1980). MODELING EXPOSURE OF NONSMOKERS AT *0RX Exposure of the population to the particulate phase of cigarette smoke can be modeled to determine both range of e xposure and the nominal inhaled dose, which is the exposure multiplied by the respi ration rate (Altman and Dlcmer, 1971). Repace and Lowrey (1980, 1982a) have shown chat the ambient concentration of tobacco smoke particles, Q, from cigarette smoking can be usefully represented by an equilibrium model of the form Q 650 Ds/Cv where Ds is the number of burning cigarettes per I00m^, and Cv is Che rentilatory air exchange rate in air changes SPI-00228 u per hour (ach). Rewriting this la ce rms of the occupancy of the space by habitual smokers (Repace end Lowrey, 1980) (for every 3 habitual smokers, there Is one cigarette burning constantly), Dhs( 3:) s): Q - 217 D^/Cr [ug/nr) IA1] , where is Che habitual smoker density In units of smokers per 100 m , and Cy Is Che effective air change rate In units of air changes per hour (ach). Because ASHRAE, The American Society of Heating, Refrigeration, and Ventilating Engineers (Leaderer, at al., 1981), a national et.glneerlng society, sets consensus standards for ventilation races In Che U.S., and because chose standards are tied to expected building occupancy (e.g., ASHRAE, 1981) , Eq. [1] offers Che possibility of modeling Che range of nonsmokers' exposures by estimating the ranges of occupancy and air change rate. Appendix Al estimates that Che average annual exposure to ambient tobacco smoke particles by a typical nonsmoking U.S. worker Is 1.8 mg/day, with a exposure probability of 62.51. MODELING EXPOSURE OF NONSMOKERS AT HOME By reviewing data*from time budget and census studies, the average length of time a person spends in the home micros nvlronment can be calculated. This time differs for gender and employment status Taking into account the different amounts of time spent in the home by employed m en, employed women, and homemakers, an estimate of occupancy--weighted average number of cigareetes smoked in Che home during a 16-hr waking day, assuming chat If the sntire waking day were spent at home, Is 32 cigarettes per day (CPD) smoked In the louse by a smoker of either sex. An escimaced occupancy-weighted average number of cigarettes equal to 22 CPD smoked in SPI-00229 the typical horn* is derived In Appendix A2. Using Eq. 1, multiplied by the ratio 22/32, times a 1 a?fiir respiration ra te for a 16 hr period, the. calculation Is made for single family detached dwelling of 340 m3 volume (see Appendix A3), assuming that on a 16 hr basis, the entire finished volume of the home Is available for dis persion of the smoke. A typical nonssoker of either sex appears to be exposed to an average inhaled dose of 0.45 mg/day, |usumlng that occupancy of the home by smokers and nonsmokers is coincident. MEAN ESTIMATED DOSE TO A TYPICAL ADOli FROM THE MOST-FREQUENTED MICROENVIRONMENTS A probability-weighted average exposure to a hypothetical typical D.S. adult Is estimated by combining the estimated dose to U.S. adults exposed in the workplace and at home, by weighting the exposure received in each mlcroenvironment by the probability of receiving It. Appendix A1 estimates that nonsmoking U.S. workers are exposed on the job to tobacco smoks with a probability of 63Z. Appendix A2 estimates that nonsmoking U.S. adults fere exposed at home to tobacco smoke with a probability of 62Z. Table 1 (main texi:) gives the combinations of these probabili ties, assuming that they are Independent, l.e., that exposure at work Is not corre lated to exposure at home. Table 1 sujgests that only a relatively small percentage (14Z) of the population may escape daily passive smoke exposure. By contrast, individuals having exposure both at hone and at work constitute a high exposure group, with the workplace likely contributing more exposure than the home by a ratio of 4 to 1. On the basis of Table 1 it Is estimated that the mean daily exposure to tobacco car and nicotine from the breat ling of indoor air contaminated by cigarecte smoke, to nonsmoking U.S. adults, is about 1.43 mg/day, averaged over the two mostfrequented microenvironments. This may be compared to the estimate of 14 mg/day to the hypothetical most-exposed individual (Repace and Lowrey, 1980). These results indicate that the typical U.S. 'nonsmoktr" appears to be exposed to a finite, non- SPI-00230 iaro aaount of tobacco aerosol, equivalent in value to three low-tar cigarettes (0.33 eg) per day. In summation, it is possible, baked on ASHRAE standards, doe budget and cen us surveys, Che physics of indoor air pollution transport, and cables of re*pi ration races, to estimate the average exposure of a typical nonsmoking O.S. adult of working age. Using Ctils methodology, estimates of the average exposure of Che U.S. adult populaciin of working age to Che particulate phase of ambient tobacco smoke are made fir the two most-frequented microenvironments the workplace and the home. It is Estimated that 86Z of adults of working age are exposed to ambient tobacco smoku on a dally basis, and 14Z are not. It Is estimated that the range of exposure varies from 0 to 14 mg of tobacco tar per day, and that the typical exposure, averaged over 100Z of the population, is 1.43 mg/day. It also is estimated that Chose individuals who are exposed both at home and at work receive a dally average exposure of 2.4 mg/day, and that 39Z of the adult worker population Is in this category. Those individuals 'exposed only at home receive a daily exposure of 0.5 mg/day, and chac 23Z of the adult population is in this category. Finally, it is estimated that chose individuals exposed only at work receive a dally exposure of 1.8 mg/day, and chac 24Z of Che worker population ij; in this category. Thus these estimates suggest that Che ratio of workplace dose to the exposure received at home is nearly 4:1, indicating that, on the average, the workplace is a more important source of exposure than the home environment. Consistency of these estimates of workplace and domestic exposure vith field data is given in Appendices 1 and 2. SPI-00231 APPENDIX AI: MODELING THE AVERAfrE DAILY EXPOSURE TO CIGARETTE SMOKE FOR A TYPICAL U.S. NONSMOKING WORKER_____________________ It is possible to arrive at an estimated aggregate exposure because the range of occupancies (l.e., smoker densities) is tied to the range of ventilation rates, which in turn determine the rangi of concentration of ambient tobacco smoke Co which nonsmokers are exposed* A font of eq. [Al] is given which can be related directly to the ASHRAE Standards 62-73 /ASHRAE, 1973), promulgated in 1973, which sec standards for natural and mechanical ventilation. The practical range of occupancy given in the ASHRAE Standard <>2-1973 is from 5 persons/1000 ft2 to 150 persons/1000 ft? [5.4 F/100 m2 to 161 P/100 m2], for Commercial and Institutional buildings. [From 1946 to 1973, the opera,ble engineering standard was descriptive of general practice rather than prescriptive: The American Standard Building Code Requirements for Light and Ventilation tS3, Section 8 (ASA, 1946) described typical practice for mechanical ventllat ion based on floor area, not occupancy. Section 8 described minimum values of .! CFM/ft2 for offices, 1 to 1.5 CFM/ft2 [4.4 to 6.6 L/s-m2] for workrooms, and t range of .5 to-3 CFM/ft2 [2.2 to 13.2 L/s-m2] for public and institutional buildings, with the lower value applying to museums, and the upper value to dance lulls. This implies air exchange rates varying from 3 to 18 ach, and at the maximum of 751 recirculation described, this range reduces to .75 to 4.5 ach. In 1970, 60.7Z of the U.S. workforce worked In the white-collar and service occupations which Inhabit such buildings (USDC, 1980). A 1979 survey of 3000 employers in large, medium and small corporations indicated that smoking was prohibited in only 10.5X of white-collar workplaces and in 27.5Z of blue-collar workplaces (tfICSH, 1978). These percentages would likely have been less in 1970. Eq. [A2] expresses the concentration, R, as a function of occupancy, which is now a su rrogate (Repace and Lowrey, 1980, 1982a) for smoker density: R - 25.6 f>a/Cv (ug/m^) [A2] SP1-00232 A.7 where P Is che occupancy la persons ;>er 1000 ft* [100 m*], and Cv is the ventilacory air change race in ach, as bef^re. Exposures can be calculated by multiply lng 1 by Che InCegraced average respi ation rate expected for an adult nonsmoker over an 8-hr workday. A reasonable v.alue Is 8 m^ per workshift, a value corresponding Co alcernace sitting plus llgty t work.(table A3) Multiplying Eq. [A2] by this rate yields the equation for the amount of tobacco Car Inhaled, Nd: Nd - 0.::05 P4/Cv (mg/8-hrs) [A3] where the other parameters are deflnec, as In Eq.[A2]. ASHRAE STANDARDS 62-73 yield the ranges In P4 of from 5 to 150 and In Cy of from 0.13 ach to 18 ach. Table A2 expresses the variation of cl ese parameters for the absolute minimum airchange rate to the recommended mini sum and maximum rates, and enables us to bound che modeled dose for the workplace. The extreme bounds of workplace exposure can be estimated to range from 1.35 <N^< 6.77 mg/8hrs. This assumes that one-chlrd of the occupants are smokers (following the U.S. average that one-third of the adult population smokes (USSG,1979)), and chac they smoke salesweighted-average tar cigarettes at the race of 32 per 16-hour day. (Repace and Lowrey, 1980) Clearly, the true minimum bound Is zero, and the maximum exposure may be higher due to Che presence of chain-smokers or a higher than average number of smokers, but what Is desired is an expected average value for che workplace exposure. At the ASHRAE-recjmmended minimum ventilation, che upper bound for Nd will be 3.38 mg/8hrs. This, the probable average range for Is between 1.35 mg/8hrs and 3.38 mg/8hrs. The average of these two figures ffj 2.37 mg/8hrs represents che mean exposure for U.S. workers who are on-Che-job passive smokers. This value may be transformed into an daily average using table A-l, and considering that In 1972, 382 of che workforce was female: 0.38 x (5.2 workhours/day) and 62Z was sale: 0.62 :: (6.7 workhours/day); Che sum of these is SPI-00233 6.13 workhours/day, dally average, faua, the dally average exposure la Nd (6.13/8)x2.37- 1.82 mg/day. It nov remains to estimate the percentage of workers who'are exposed to cigarette smoke at work. The National Interagency Council on Smoking and Health conducted a survey of Cop management and health officials of 3000 U.S. Corporations In 1978 (NICSB, 1978). A 29Z response rate was achieved. The survey Indicated that of bluecollar companies surveyed, 30.6Z had no restrictions on smoking, 42Z permitted smoking In designated area! , and 27.5Z completely prohibited smoking. The corresponding percentages for the white-collar companies were respectively 74.3Z, 15.2Z, and 10.5Z. Smaller coipanles were less likely to have restrictions. Among companies with restrictions, atout half imposed penalties for violations. 65Z of the respondents Indicated that their policy was established after the release of the 1964 Surgeon General's Report on Smoking. In 1970, white collar workers constituted 48.3Z of the workforce, blue collar workers 35.3Z, service workers 12.4Z, and farmworkers 4Z (USDC, 1980). The largest change in any category from 1960 to 1 979 was that of white collars, Increasing by 7Z. Since about half of the blue collar companies Imposed penalties for smoking, It will be assumed that S0Z of the blue collar nonsmokers were not exposed on the job. By contrast, It will be assumed that only 25Z of the white collars were not exposed. It will further be assumed that half of all workers follow white-collar smoking rules, and the other half, coinsisting of blue-collar workers, service workers,and farmworkers, follow blue -collar rules. Thus, the estimated weighted average percent of nonsmokln ; workers who are significantly exposed to tobacco smoke on the job is: 0.50 x 5DZ + 0.75 x 502 62.5Z. By comparison, a 1983 survey of 1515 white and blue :ollar businesses sampled at random reported that "nearly two-thirds" had no smoki ig restrictions in the workplace. (Tobacco Institute, 1984) SP1-00234 At this stage 1c oust be asked whether the numbers calculated are reasonable In terms of measurements of ambient tobacco- smoke under natural conditions. Repace and Lovrey (1980; 1982a) In i field survey of ambient fobacco smoke In 23 commercial buildings In the metropolitan Washington, D.C. area during 1979-1980, found concentrations ranging from stout 100 ug/m^ to more than 1000 ug/m^. This range Is quite compatible with the the concentrations Q derived In table Al. The average of all values measured under a variety of smoking conditions and ventila tion rates by Repace and Lovrey was 242 ug/m^ (range 100 to 1000 ug/m^) for these 23 locations, corrected for background. This Is compatible with Che values calculated In Table A2. Breathing 242 ug/m^ of ambient tobacco smoke for 8 hours at a rate of .99 m^/hr yields in exposure of 1.92 mg/8hrs or on a dally average basis, 1.92 x (6.13/8) 1.47 mg. In terms of relative exposures, :hese results also appear to be reasonable. In Appendix A2, an average smoking rate of 32 CFD was used (Repace and Lovrey, 1980). At current sales-velghted average tar plus nicotine values (14 mg) (USPTC 1984), the typical smoker would Inhale (14mj/cig) x 32 CPD - 448 mg/d. In Table 1, the typical passive smoker was calculate: to Inhale 1.43 mg/d. This Is a relative exposure ratio of 313:1. Wald et al (1984A), In a study of urinary cotlnlne levels In smokers and nonsmokers, found the ratio (1645 ng/mL)/(6 ng/mL)**`274:1. Thus, the ratio of exposures calculated theoretically using the model derived here differs by only 142 from an experimentally derived value based on a biological marker of exposure. SPI-00235 A10* AFFOfDIX A2. CALCULATION OF THE ESTIMATED DAILY AVERAGE NUMBER OF CIGARETTES SMOKED IN THE AVERAGE HOME Since the source strength depei.ds upon the length of tine smokers spend In Indoor microenvironments, It is necessary to review pertinent Information from time budget (Oct, la press; Szalai, 1972; NRC, 1980 and census (USDC, 1980) studies, which gives Che average length of time that persons spend In various 1eroenvlronmencs. From Table Al, it Is seen chat, allowing for 8 hours of sleep, employed awn spend 34.42 of the waking day In the home; employed women spend 45*92 of the waking day In Che home; "housewives* spend 812 of the waking day In Che home* In 1979, approximately 422 of families ilth both Che husband and wife present, both were employed* Thus for homes occupied by married couples, 662 of the waking day (weighted mean averaged over 422 working wives and 582 homemakers) Che home Is occupied by a wife, and 342 of the day, by a husband. If Che average habitual smoker smokes 32 cigarettes per day (2?D), then the wife will smoke 21 C?D In the house, and Che husband will smoke 11 IPD in the house. Bonham and Wilson (1981) found tliat 622 of U.S. homes with children in 1970 contained one or more smokers, and 252 contained two or more. Thus we may assume that of homes with one or more smoker::, 402 have two smokers, and 602 have one smoker. We have three cases to consicer: a) Husband and Wife Both Smoke, b) Only Wife Smokes, and c) Only Husband Smokes. In 402 of Che smoking homes, (a) Is true, and in 602 of chose homes, either (b) or (c) Is true. 382 of men and 302 of women smoke.(16) Then the probability of (c) being true is 342 (38/68 x 602), and the probability of (b) being true Is 262 (30/68 x 602). The weighted mean of these is given by the sum of the products of the percent of homes with a given number of smokers of either or both se::es, times the number of cigarettes per day smoked by either or both sexes: .40 x '. 2 * .26 x 21 + .34 x 11 - 22 CPD, estimated SPI-00236 All. to be smoked dally In Che average U.S. home-, or abouc a pack per day. Is chls theoretical estimate a reasonable nuaber? Dockery and Spengler (1981a; 1981b) In a one-year study of Indoor air pollu tion In 68 homes In 6 U. S. cities, f|ound that cigarette smoking was the dominant source of respirable particles (RSP), and that In a typical house in Che study, the average 24-hr RSP levels were Increased by 0.88 ug/m^ per cigarette smoked, and in a tightly sealed house, by a value of 2.11 ug/a^. At an estimated occu pancy-weighted average in-the-home smoking rate of 22 cigarettes per day, a 24-hr average RSP level of about 19 ug/a^ (22 CPD x 0.88 ug/m^-CPD) is calculated for Che typical house; In fact, Spengler, et. al. (quoted In NRC, 1981) observed. In 22 of the homes In the study where there was only 1 smoker, a 24-hr average of 19 ug/rn^. This number corresponds to an air exchange rate of 1.5 ach using the model (see Appendix A3). Since this ilr exchange rate Is within the expected range,(Repace and Lowrey, 1980, 1982) an average of 22 CPD smoked In the home provides a reasonable basis for estlmitlng exposure. The theoretical ln-Che-home number of cigarettes smoked In the home Is weighted for occupancy during Che waking day. Since there Is no data differen tiating occupancy for smokers and nonnmokers, it Is assumed chat the statistical occupancy of the nonsmoker Is coincident with that of the smoker, l.e., that there Is a nonsmoker present to receive che exposure. In order to calculate the dally dose received, Eq. Al Is used w:.th Che parameters D^g 0.29 smokers/lOOm^, C7- 1.5 ach, times an occupancy factor of 22/32, times a respiration rate of 0.99 m^/hr, times a 16 hr maximum exposure day, yielding an estimated average exposure of 0.45 mg/day, for an adult nonsmoker, with an exposure probability of 62/1. A reasonable approximation to th probability of a typical nonsmoking adult SPI-00237 Ai i. blag exposed Co ambient tobacco sm<icke at home Is 62Z, the same as Bonham and Wilson (1981) above found for adults vlth children (In 1970, 56Z of families had one or more children under 18) (USDC , 1980) No differentiation Is made between male and female nonsmokers since Fr:Redman et al.(1983) observed that degree of passive smoking had little correlatl on vlth gender. In households with 2 or more smokers, there might not be an adulc nonsmoker to be exposed; In this case, the probability of passive smoking (for a nonsmoker) reduces from 62Z to 37Z. However, the estimated total exposure (Table 1) only decreases by 8Z, from 1.46 mg/day to 1.34 mg/day. In the absence of dats on this point. It will be assumed that a nonsmoking adult Is present. APPENDIX A3. CALCULATION OF THE RATIO OF THE HABITUAL SMOKER DENSITY TO THE EFFECTIVE VENTILATION RATE FOR A TYPICAL U.S. SINGLE FAMIL'Y HOUSE* * The typical range of annual doised-vlndow air exchange rates In U.S. residences is generally considered to be of the order or 0.5 ach to 1.5 ach, vlth the range for the average residence of the ordier of 0.7 to 1.1. ach, and that of the tighter and never residences of the order of 0.5 to 0.8 ach (Fuller, 1981). So-called energy-efficient structures have rat es of the order of 0.3 to 0.5 ach. (Repace, 1982) A typical U.S. single-family detached house Is estimated to have a floor area of 1500 sq.ft. [139 m^] with an 8-ft [2.4 m] ceiling, for a volume of 340 * (NAHB, 1981). Thus, per habitual smoker, the ratio Dhs/Cv " (1/3.4)/ 1.0 - 0.29 habitual smokers per hundred cubic meters per air change per hour. In 1978, nearly 2/3 of occupied housing units were single-family det ached dwellings (USDC, 1980). It is assumed that the ratio calculated above is alid for multifamily dwellings as well (the volume of an apartment in a multi-fa ally building is likely to be less, but the air exchange rate is likely to be gr eater). SPI-00238 APPENDIX B: EXTRAPOLATED ESTIMATE 01' RISK FROM PASSIVE SMOKING An alternative method of estimation of rlslt from passive smoking Is calcula ted as follows. In 1980, 108,504 Individuals In the U.S. were reported to have died from lung cancer (DSPHS, 1983). The 1982 Surgeon General's report on Smoking and Cancer estimated that 8SZ of LCDt are due.to cigarette smoking (USSG.1982) this yields 92,228 LCDs/yr! Lung cancers occur primarily In smokers over the age of 35 (NCI Monograph 19, 1966); In 1980, there were an estimated 29,225,000 smokers of all races and both sexes In this, alge bracket (DSPHS, in press). It follows that In 1980, there were 3.156 z 10~3 LCDs per smoker of lung cancer age. In 1978 Che average cigarette was 17 mg tar, and the average smoker smoked 32 per day (Repace and Lowrey, 1980), for an estimated tar Intake of 544 mg/day--smoker. (A 1980 lung cancer death reflects a 20 to 40 year smoking history, during which smoking rates Increased by, and tar 1 evels decreased by, about 50Z (DSSG, 1979). Thus, 3.156 x 10~3 LCDs/smoker divided by 544 mg/day-smoker yields a rate of about 5.8 x 10"^ LCDs/yr per mg/day >er smoker of lung cancer age. Using a one-hit model (Hoel, et il., 1983; Crump, 1976) for Che extrapolation of the risk from the estimated exposure of smokers down to Che estimated exposure of nonsmokers provides an alternate e qjosure-response relationship. Crouch and Vllson (1981) have used this model wfc.ch saturates at high exposures, but which is linear at low exposures. This model lias the form P(D) - 1 - exp (bD), where ?(D) Is the estimated risk, b Is Che expost ire-response function, and D Is the exposure. This model, because of Its functional form, can be considered as Che first stage of the more complex multistage model. (USEPA, 1983a; Hoel, et al., 1983) Whenever the data can be fitted adequately by the one-hit model, estimates of both models will be comparable (DSEPA, 1983a; Crunp, 1976; Hoel, et al., 1983). SPI-00239 E-2 From above, b " 5*8 z 10"^ LCDs par year per mg/day. D - 1.5 mg/day, from the esti- * mated avarage azpoaura for Cha typl ca1 O.S. nonsmoker (Appendix A), assuming that par milligram, tobacco tar produces the same carcinogenic response In nonsaokers as It does in smokers. This ealeulatlo a yields an estimated annual LCD risk of about 0.87 z 10~5 from passive smoking, or about an-order of magnitude lower than the phenomenological estimate made earll|er. In this exposure range, this result Is essentially the same as would be obt lined from a linear extrapolation. The primary age group at risk o E lung cancer Is that >_ 35 years (keif, 1931a;b). Therefore, In the calculation that f >Hows only nonsmokers 35 yrs will be assumed to be at risk of lung cancer. In 19*0, there were about 63.8 million nonsaokers aged 35 (USPHS, In press). Thus, tine alternative risk estimate Is derived from multiplying 0.87 LCDs/yr per 100,000 passive smokers times 63.8 x 10 passive smokers at risk yielding 555 LCDs ' year In U.S. nonsaokers from passive smoking, using the one-hit model of carcinogenAesls for extrapolation. SPI-00240 APPENDIX C: ACE-STANDARDIZED CALCULATION OF ESTIMATED ANNUAL D.S. MORTALITY AND LOSS Of OP LIFE EXPECTANCY FROM IN? OLUNTA&Y EXPOSURE TO AMBIENT TOBACCO SMQICE Approximately 50Z of SDAs la th cancer age range (>35 yrs old) are adult coeverts Co the church; others vere either bora Into an SDA home or joined the cburch prior eo ge 20, typically i th ocher Immediate family members. A large proportion of SDAs tend to be heavily Involved In church activities. Only eery small proportion of SDAs re port current use of cigarettes (males, 1.7Z; feaales 0.5Z) (Phillips,et al., 1980b) (By contrast. In 1970, 43.5Z of adult males and 31.1Z of adult feaales la the general population aged >.17 years reported soaking) (USDRHS, 1979) Moreover, a substantial portloln of SDAs work for "an organization owned and operated by ehe SDA Church'* (nearly 45Z of SDA females and 40Z of SDA males In the study group, (aged 25 years), reported working for the SDA Church. (Phlllips et al., 1980a; 1980b). Clearly , SDAs are less likely 'than Che general population to be Involuntarily expo sed to tobacco smoke, as children or as adults, at home or In Che workplace , because neither SDA homes nor SDA businesses are likely to be places where saokl|i;g Is permitted, and because the great majority of SDA family and social c jntacts are among ocher SDAs who do not smoke (Phillips et al. 1980b). Table Cl shows the age-standarjllzed calculation of estimated loss of life expectancy and annual lung cancer bi< brtallty from passive smoking. The calculation Is based on the lung cancer mortall :y difference between two Southern California cohorts of self-reported nonsmokers who never smoked. Based on lifestyle differences, they appear to have dl rferenc average levels of Involuntary smoke exposure. The more-exposed group ar<` designated non-SDAs, and the less-exposed group SDAs (see text). Columns 1, 2, 5, and 6 are t:abu lations from which age-adjusted mortality rates were calculated in the study f mortality in the Sevench-Day Adventist SPI-00241 C-2 (SDA) by Phillips et al. (1980s;1180b). Columns 1 and 2 and 3 and 6 give the age-specific lung cancer deaths and person-years at risk respectively for ehe SDA and the non-SDA. The fractional number of LCDs in column 1 Is due to a correction for out-migration of the SDA population from the study area. Columns 3, 7, 10, and 11 shov the average numbers of individuals at risk annually during the study, allowing for those who died during the study. Cols. 4 and 8 show the annual aveiage lung-cancer death rate (LCD) per 100,000 persons, and Col. 9 gives the dlfljerenees between the non-SDAs and SDAs in those races. Col. 12 gives average LCD rates weighted to reflect the fact that there were three times as many women as men In the study, and that the female data attained statistical significance whereas the male did not-----although the combined data were significant. (pjhillips et al., 1980a; 1980b) A common LCD race is assumed for both sexes In Che calculation that follows. Also, it will be assumed Chat Che entire LCD rats difference Is due Co passive smoking (see' discussion on confounding factors In Appendix D). Next, this calculation will bs extrapolated to the entire U.S. nonsmoking population aged 33 years. Col. 13 gives the mean age of the Individuals in the S-year age group, and Col. 14 gives the number of persons alive at that mean age per 100,000 born alive. C>1. IS gives the total number of persons In the 5-year age group (5 x Col. 14) per 100,000 born alive (whites only) from the 1974 U.S. Life Tables (USDHHS, 1975). Col. 16 gives the age-specific LCD rates attributed to passive smoking, standardized to (l.e., veighced by) the age specific population distribution In 1974 for U. S. whites (col. 12 times col. 15). Col. 17 gives che average lifi expectancy corresponding to the mean age SPI-00242 T C-3- glven la Col* 13, which Is taken cs represent that of the etttlre five-year age group. Col. 13, the product of Co Is. 16 and 17, gives the estimated age-speclfl age-standardized person-years of 1 Ife lost due Co lung-cancer from passive molting. The sum of the values of Col. 18 gives an estimated 3932 person-years of life lost due to passive smoklnk per 100,000 persons alive at age 35 in the 0. S. population In 1979. 3932 person-years, vhen divided by the 94,724 persons (USDHHS, 1975) at risk at age 40 (!.CDs were not observed at earlier ages in the SDA study; however, they are observed in the general nonsmoking 0. S. population at age 35) (USSG, 1979) yields 15 days, the mean number of days of life lost, and multiplying by the peak-to-mean exposure ratio, 112 days for the maximum number of days lost (where the risks of the non-white population are taken to be the same as for the white population.) Col. 19 Is col. 16 times 62.4:14 million divided by the sum of col. 15. The sum of Col. 19 gives an estimated age-standardized mortality total of 4,665 LCDs per year in U.S. nonsmol.ers from passive smoking (where there were 93,636,000 persons aged these were nonsmokers). 35 years In 1979, and two-thirds or 62,424,000 of Examining Col. 19, shows that of those Individuals assumed to contract lung cancer from passive smoking, ihac approximately 1-1/22 do so at each year of age from 40 to 69, anc that over age 70, approximately 32 do so each year. Of those who actually contract fatal lung cancer from passive smoking, the mean life expectancy lost Is about 17+9 years, and about 82 lose as much as 33 years. SPI-00243 APPENDIX D; DISCUSSION OP CONFOIfNDINC FACTORS Th XA&C criteria for causality and human cancer specify that possible sources of bias and confounding drror should be considered (IARC, 1979). What factors ocher than passive emokicg could account for a lung cancer difference between two cohorts? The most obvious one is aisdlasalflcatloa. Some of Che individuals clas sified as nonsaotears could have teen smokers or exsaokers, giving rise to a spurious effect. Vorkplace or residential exposure to lung carcinogens or dietary differences between the cohorts might also give rise to spurious differ ences. However, this Is not likely Co be an effect constant over nine positive studies la five different countries, all of which report about a doubling of risk when Che exposure variable 1 i spouses' smoking. Arsenic, asbestos, beryllium, ihloroethers, chromium, coke oven emissions, nickel, radon, and vinyl chloride as well as tobacco smoke, have been lmplieated In the etiology of lung cance (Ives, 1983; Sellkoff, 1981). Possible differences due Co Industrial exposures should be expected primarily in bluecollar workers. Phillips et al. {1980a; 1980b) have stated that the SDA/non-SDA subgroups were demographlcally and educationally similar, suggesting similar occupational distributions, although there Is no Information on this point. There is no reason to believe chat domestic radon levels, which are a property of the soil, would be any differet t in SDA homes than Non-SDA homes. Finally, it should be considered Chat co-exposures to other lung carcinogens (e.g. radon) may Increase the effect of passive smoking (Bergman and Axel son, 1983). It is also possible chat dietary differences between the two groups might have contributed to the SDA/nonSDA lung cancer difference. 542 of SDAs follow a lacto--ovarian diet and 412 rarely use caffeine beverages. However, Hirayaaa S PI-00244 D-2 (1981a; 1981b; 1983a; 1983b) observed a dose-response relationship between exposure eo passive smoking and lung cancer even in those with an apparently cancer-inhibiting diet. Also SDA/non-SDA cancer differences are not significant for other smoking-related cancer s ites; this runs counter to a protective effect of diet as a confounding factor. Finally, Hirayana (1983a) observed that the magnitude of this effect raried from mortality ratio of 1 for passive smoking women who did not follow a protective diet to 0*82 for women who used green-yellow vegetables only occasionally, to 0.72 for women who ace them daily. Thus the magnitude of the effect does not appear to be sufficient to account for the observed SOA/NonSD/. lung cancer difference. Moreover, If 40Z of the SDAs work for church-run organizations, 60X do not:* these 60Z surely oust be subject to some passive smcklng In the workplace, at least partially offsetting Che effects of potential dietary or occupational differences with the nonSDAs. SPI-00245 T TABLE 1. ESTIMATED PROSABILITipS OF NONSMOKERS EXPOSURE TO TOBACCO SMOKE AT HOME AMD AT WORK ( if tar Rapace and Lovrrey. 1983; Appendix A) Non-exclusive probe!,! ill icy of balagexposed at work: 63Z Probablll :y of not balagexposad at work: 37Z Non-exclusive probalilllcy of beingexposad at hoae: 62Z Probablll ;y of not beingexposed at home: 38Z Lifestyle: Dally Average Probability of being exposed (Rounded Values) Exposure (ag) Modeled Dally Dally ProbaAverageblllty-tfelghtec At work and at hoae: Z l>3 X 62 - 39 Neither at work nor at hoae: Z :;7 X 38 - 14 AC hoae but not at work: Z (2 X 37 - 23 At .work but not at hone: Z (3 X 38 24 Total: Z 100 2.27 0.00 0.45 1.82 0.89 0.00 0.10 0.44 1.43 Table 1* The esclaated exposure to Che particulate phase of aablent tobacco saoke for U.S. adults of working age, at work and at hoae (these two microenvironments account for an estlaated 88Z of Che average person's -- both smokers and nonsaokers tlae), determined froa average Concentrations of tobacco saoke calculated for nodel workplace and hoae aicroet^vlronaents, weighted for average occupancy, as derived in Appendix A. SPI-00246 TABLE 2: AGZ-ADJTJSTED SDA-TO-NONSDA RATIO OF LUNG CANCER MORTALITY (a:fit eg Phillips, et al.(1980b} I. All SDAa II. SDAa who Never Sooted Average 0.54 0.41 By HtalCh Hable Index Best Third Average Worse Third Third 0.54 0.40 0.96 0.41 0.32' 0.78 Values shown are adjusted by ManCel-Haenzel procedure (p <_ 0.01).* Lung eancer mortality ratios taken from a prospective study of two demogra phically similar cohorts. The non-SDA come from the general south California population, and were self-reported nonsmokers who never smoked. The SDA come from a southern California subgroup less likely to engage In passive smoking by virtue of lifestyle differences The health habit Index Is a measure of how faithfully Individuals adhered to the Church's teachings; the worst third vere also more likely to have a non-SDA spouse) (Values quoted in text are the reciprocals of numbers given he re.) Phillips, et al.(1980a; 1980b) reported results for all SDA, and report ed replicating these data for SDA who never smoked, as shown (R.L. Phillips , Department of Blostatlstlcs and Epidemiology, Loma Linda University, Lome Lin da, CA 923S0). The SDA subjects and nonSDA subjects for this study consist ed of white California respondents to the sane four-page self-administered questionnaire collected by the American Cancer ted States (NCI Monograph Society study of 1 million subjects throughout the Uni 19, 1966; Garfinkel, 1981; Philips et al., 1980a; 1980b). SPI-00247 TABLZ 3. ESTIMATED LOSS OF LITE E3PECTANCT EROM ACTIVE SMOKING (ALL CAUSES) AND PASSIVE SMOKING (LUNG CANCER OhLY) -- adapted from Cohan and Lee(1979). Causa Cigarette smoking -- male Clgaracce smoking -- female Cigar smoking Pipa smoking Passive Smoking* (Esc. most exposed lifestyle) Passive smoking* (Esc. average lifestyle) 2250 800 330 220 148 15 *Estimated this work (see Appendix C); averaged over all nonsmokers at risk, l.e., chose who are presumed eo die from passive smoking-induced lung cancer, and Chose who do noc. Estimates given for passive smoking are phenomenological estimates. SPI-00248 TABLE 4. NUMBER OF WOMEN IN EACH EXPOSURE CATEGORY IN THE GARFINKEL(1981) STUDY OF PASSIVE SMOKING AND LUNG CANCER Group No. Total cohort 176,739 *Tn' controls: do not work, hush ends do not smoke 30,682 "Tainted" controls: work, husbands do not smoke 18,805 Total "controls" 49,487 "Exposed" workers: work, husbands Jraoke "Exposed" non-workers: do not work husbands smoke 48,356 78,896 Total "exposed" 127,252 TABLE 4b: CALCULATED LUNG CANCER IlISKS FOR EACH SUBGROUP IN THE GARFINKEL (1981) STUD USING THE 5 LCDs/100,000 person-years/mg/d EXPOSURE-RESPONSE RELATION. Group Rate True controls 8.7 Tainted controls 17.8 (8.7 + 9.1) All controls (weighted mean 12.16 Exposed workers 20.05 (8.7 + 2.25 +9.10) Exposed non-workers 10.95 (8.7 + 2.25) All exposed (weighted mean) 14.41 TABLE 4c. CALCULATED LUNG CANCER RISKS FOR EACH SUBGROUP IN THE GARFINKEL (1981) STUDY USING THE 0.6 LCDs PER 100,000 person-years/ag/d EXPOSURERESPONSE RI ELATION. Group Race True controls 8.7 Tainted controls All controls (weighted mean) Exposed workers Exposed non-workers All exposed (weighted mean) 9.8 (8.7 + 1.1) 9.11 10.07 (8.7 + 0.27 + 1.1) 8.97 (8.7 + 0.27) 9.39 SPI-00249 Table 5. COMPARISON OF ESTIMATE :d RISKS FROM VARIOUS HAZARDOUS AIR POLLUTANTS Risks have been assessed fo : non-occupaclocal exposures of the general popula tlon to several hazardous air po lutancs. All are airborne carcinogens; all but passive smoking are being regula ed by society. The statistical mortality given Is bfor control. POLLUTANT Passive Smoking Vinyl Chloride Radionuclides (world-wide Impact from Department of Energy facilities) Coke Oven Emissions Benzene Arsenic ESTIMATED ANNUAL MORTALITY 5000 LCDs per year <27 CDs per year 17 CDs per year <15 LCDs per year '8 CDs per year '5 LCDs per year Reference (this vork) (USEPA,1975) (USEPA,1983b) (USEPA,1984) (USEPA,1979b) (USEPA,1980) CD Cancer Death; LCD " Lung Cancer Death Risks for passive smoking and radionuclides are best estimates, and risks for ocher pollutants are upper bound. SPI-00250 Table Al. TIME SPENT IN VARIOUS IICROENVIRONHENTS BY PERSONS IN 44 U. S. CITIES. EXPRESSES IS AVERAGE HOURS PER DAY. Ott. (In press); KRC M98H- Ssalai( 1972) Mlcroenvlronaent Married HouseEmployed Men, All Days Employed Women, All Days wives. All Days Inside one's home 13.4 15.4 20.5 just outside one's home 0.2 0.0 0.1 at one's workplace 6.7 5.2 In transit 1.6 1.3 1.0 in other people's homes 0.5 0.7 0.8 In places of business 0.7 0.9 1.2 In restaurants and bars 0.4 0.2 0.1 In all other locations 0.5 0.3 0.3 Total 24.0 24.0 24.0 SPI-00251 "ABLE A2 CALCULATION OF THE RANGE OF CON'CENT! A.!TION Q, and EXPOSURE Nd TO WHICH NONSMOKERS ARE SUBJECT UNDER THE MODEL GIVEN BT EQRATIONS 2 and 3 ASSUMING ASHRAE STANDARD VENTILATION A. USING ASHRAE 62-73 RECOMMENDS! MAXIMUM MAKEUP AIR BASED ON OCCUPANCY OCCUPANCY Pa Occupants per 1000 ft2 [per 100 a2] ca air- Exposure' changes/ hour (aeh) (mg/8hrs) Q Concentration (ug/m3) MAXIMUM 150 18 1.69 213 MINIMUM 5 0.5 2.03 256 OFFICES 10 1.5 1.35 170 B. USING ASHRAE 62-73 ABSOLUTE ML fIMUM MAKEUP AIR BASED ON OCCUPANCY MAXIMUM 150 4.5 6.77 853 MINIMUM 5 0.15 6.77 853 OFFICES 10 0.3 6.77 853 C. USING ASHRAE 62-73 RECOMMENDED 1 LNIMDM MAKEUP AIR BASED ON OCCUPANCY MAXIMUM 150 9 3.38 655 MINIMUM 5 0.3 3.38 655 OFFICES 10 0.9 2.26 284 gpi-00252 TABLE A3 RANGE OF TYPICAL ADULT RESPIRA! TON RATES FOR DIFFERENT LEVELS OF EFFORT after Altman and Diener (1971) ACTIVITY LEVEL Resting Sitting Alternate Sitting & Light Work Light Work Heavy Work RESPIRATION RATE (j^/hr) 0.36 0.60 0.99 1.47 2.04 SPl-00253 APPENDIX C AGE-STANDARDIZED ESTIMATION Of ,LUNC CANCER DEATHS FROM PASSIVE SMOKING O o o oo--< eo 0a 0 b b 3Cj & cu >b 0 r- 0W>00 2o m <03ee c00 b a0* <ob 0 0S w 0 uo ofl 0 *0 b0 20> 0 b>> o00 -e0** b0 U Cu < b* 0m3 "o0 in 0 HO 0* OJ w^ b 0 a o o o 0* C0ft. oo~ e0 0 0 *T O b U CJ 00-00- > I&b0 0 MX 000 b0 e3 0 n >< <c 2o I w b0 e O ui 0> 0 0 b0 2 b *x 0- X 06 c< C/3 b jOUo X wr--*<* T*b000*34 C0bO -a b3 X0 in cbj 0^ ^ XN >00 N>0 n* ooo 0O0 CM VI 'O n ^ nv nr* o O o O' m O' M vO rC-4 o S3 >r--0 CM tn cvi >0 in OCO' CVvMi CM CMM -- crrOss-..' o COM m3 rOCM*' r0C---M0i -- foSO*3' *^N^0U^\AvO N -- -* cm-0* ~ OOOpiN<ON<(n| r-. tn ^ vrt 0 ^ On o m O ^ **y <n 03 cm o o o o ao*_*?-o in O a'n O^ *0 ^c* N00 NO O n N N on ffinn *c ^ m in in m *<on o rnC* in o n 0CnM0 CM O' iCnO r6C"M>3 vv-0iDv0-Pi 4^OciOni'nnm'OinCv0*NOG NnN<'^'OC3vflNirtO ^tf>CvxON'u^nn OOOO^G~s#O'T-- ----n o cn -- 0&3 b 030 H0 < VbX1 i O<nK m<In o> o.*I ^(OIoisnii*vmreni *oO'iT^G^^i O'n+ -f'Gofil'Op'J*L J* SPI-00254 ACE-STANDARDIZED ESTIMATION OP LUNG CANCER DEATHS PROM PASSIVE SMOKING ( c a n 1 TJ V.i --ca o e <- -- 03 0 , <--tm. uCO wa *3 aU (J> e No w Co* vO 60 Q. BO --* 3 a. 3? 0t mh ^4 * ^ ^* *0 (N < ^ M 'O ^ N-- WO <-N ps ^ bT ^ r ffl 'O CN b c > e - a O <O8 3' O3 Z b <B b . U- bO Irt c 0* at *-l l xB wO u< MC <bO m O tn o zb o m vn in ir> i/b O " in Ot6\6Ct B ffl fi N *T 'T *0 i a> oo <n O wn o o -- *9 **9 S J ifl <N H 'fl s ot /i O m tn tn ^ n m m <n -m 9o* r> a u tw & 0 CO J0*bO bO3 . X b4) -a >> O <eaU b <M U Xo 0 >b X 0< 0 tn. e O b 3& m a 4i > o X BO W wb mNQ n n ct csi -- m tn tn -r 9<m 3 s ^ 9^n s 9 n n o - r pi 4 8 nt ot p9t^--NOo^<--n-- oooA ^4 oD O W 3 M^w ^ a 9 wn oC wX3 t9a- U0 rt%n <n r*s <nn ^n ^^ r*** nn am r0*9 O o V a ^ X 0 mi *4 4J O 8 <s 93 JmZ e c *4 c -- U 3 <C 59X90. w q **9 t0n --eo r% m 800 cst t<nn so to n*j n o> cn r< 4* <N Ct 9t yO sT f9li Urn A^M o< < U vs a 9 X w 4a)** W Xa u 0i ) c oz os notrNCtvt^ctNnC'tf ^ OMACN ot tfns m9 tcno o* ^09 nr% voi ^<n tn n t<r-t > 00 O m <<Z4 m t<ns o E co *M < --O Wa <Q to&v** cn ^a oPM <9>U <wIb(OI o e m ZO Z*O4 Xu pt/i^OO ao coto ^-- trni cn VoB^ nCNOtOON<rt'OOiN(-N -- *9 nO r* <9 <n n n - 9 a O oO <I -- oo vai a i s u c< <Ce -J u o oawx w -- n m rO N<"A 9* o8 Nn o N OA - n a- O tn v6 e a- n<s *mr m a> -4 >o \n -- o N O 8 8 *7 fS <N O sT sO <00 U a3 >s OU tn o t ^9t 9 9t <n ^^ n m tnIIoIwn I tnf n << tn tn OoIt -# opasI ol i/i O in O /i ^ O8 cN<Hn*'-Of*<o-n-v<,nM0ftni's^f^tlnNtA--*Cen0Nt--n -- CM CS <N pm mm m 9 9 X OOON^4V sC--Mrnv--idnyBC f08Pn-S bff s tn nj t p p4* n3 <c cc < u H IIIIIIIIII /i6 mO'nO'n o ^ C n^^tntn'8'8rNBp,s APPENDIX C SPI-00256 (*SO ao e co ft. 8W ft C CL C ** CO _ Mm mfmt /N, vfftt -o*a u a^ qn O-3 W --ft U Ot OC &Oa3* -4MO1W<>n Oo r* O cn as <(Mn <N Nm N N*sOy n --N -- ft* OO' \s0C sQ U\ I4ri ft 0 33 Owe -fwt Aftt f>ts.--w *y3oJ0 --O wO ft wo* f>t c a fOt, Mofftl O MN %Onc On^w*i C t OS On VOy*Qf\l <N ft *** a>s *I"i 6 ft fftt yft 3ft <ao Ps ft u W & -- Wft &ft UUSft o ->ft * *O >ft* ow O' rs ft ft <n O * 'O sO ** 1i o <N N - ft ft fl APPENDIX C ft to < a U3 > w*i <u03 O' tP"i ^ -sT I 0 O.o as f <# inf wO* o Ii/*i ^y^ ftl Oc O' ft /*I> ft rO%' c*ox 0I I ioI n-f rs rs ao ft T RETERZNCZS Albert, R.E. (1983) "Discussion of APCA Critical Review Paper on Control of Toxic Substances la Ch Atmosp ieric Environment", J. Air Pollution Control Assoc. 33^ 836-837. Anderson, Z.L. (1983) " Quantitati re Approaches In Use to Assess Cancer Risk" Risk Analysis 3.: 277-295. Aleman, P. L.; and Dicmer, D. S (A 971) "Respiration and Circulation" Federation of American Soc. for Experimental Biol., Bethesda, MD. American Soc of Heating, Refrig. ai id Ventll. Engineers (1981) "Ventilation for acceptable Indoor air quality ASHRAZ Standard 62-1981,Atlanta, Ga. American Soc of Heating, Refrlg. ^d Ventll. Engineers (1973) "Standards for Natural and Mechanical Vend lai:i on", ASHRAZ Standard 62-1973, New York, N.Y. * American Standards Association (19|6) "Light and Ventilation" ASA 53, American Standards Association N.Y., N.Y. Bergman, H. and Axelson 0. (1983) Passive Smoking and Indoor Radon Daughter Concentrations'*, Lancet, II: 1 $08-1309. Bock, F.C.; Repace, J.L. and Lowref , A.H. (1982) "Nonsmokers and cigarette smoke: a modified perception of risk" Science, 215: l97* Bonham, G.S. and Wilson, R.W. (198 ) "Children's Health in Families with Clgarett Smokers", American J Public HeAlth, 71_: 290-293. Chan V.C. and Fung, S.C. (1982) 'Lung Cancer In Non-Smokers In Hong Kong," E. Grundaenn (ed): Cancer Campiign Cancer Epidemiology, Gustav Fischer Verlag, Stuttgart, New York Cohen, J.D. and Bartsch, G.E. (1980 ) "A comparison between Carboxyhemoglobln and Serum Thiocyanate as Indl a tors of Cigarette Smoking" American J Public Health, _7<3: 284-286. Cohen, B.L. and Lee, I. (1979) "A Catalog of risks", Health Physics, 36_: 707-722. SPI-00257 R2. Committee on Science and Tecbnolc gy (1983) "A Review of Risk Assessment Methodologies m U.S. House of Representatives , 98th Congress, D.S. Govt. Printing Office. Correa, P.; Pickle, L.V.; Ponham, E.; Lin, Y. and Haenszel, W. (1983) "Peaive Smoking and Lung Can cer". Lancet, II: 595-597. Crouch, E.; Wilson, R. (1981) "Re gulatlon of Carcinogens", Risk Analysis 47-57. Crump, K.S.; Hoel, D.G.; Langley, C.H. and Peto, R. (1976) "Fundamental Carcinogenic Processes and Their Implications for Low Dose Risk Assessment", Cancer Research, ^6.: 2973-2979. Dockery, D. and Spengler, J. D. (|1981a) "Indoor-Outdoor Relationships of Respirable Sulfates and Parti ties". Atmospheric Environ. 15: 335-343. Dockery, D. and Spengler, J. D (1981b) "Personal Exposure to Respirable Particulates and Sulfates", J. Air Pollut. Contr. Assoc. 31: 153-159. Doll, R. and Peto, R. (1981) "The Causes of Cancer", New York: Oxford University Press. Enstroa, J.E. and Godley, F.H. (l9j80) "Cancer Mortality Among a Representative Sample of Nonsmokers In the UiAlted States During 1966-68", J. Natl.Cancer Inst., 65: 1175-1183. Enstroa, J.E. (1978) "Cancer and iotal mortality among active mormons", Cancer, 42: 1943-1951. Flschoff, B.; Lichtenstein, S.; SJ ovic, P.; Derby, S. and Keeney, R. (1981) "Acceptable Risk", Cambridge: Cambridge University Press. Friedman, G.D.; Petlttl, D.B. and Bawol, R.D. (1983) "Prevalence and correlates of passive smoking", Am J Publ ic Health, .73: 401-405. Garfinkel, L. (1981) "Time trends In lung cancer mortality among nonsmokers and a note on passive smoking" , J. National Cancer Inst., 66: 1061-1066. SPI-00258 .22 Carflnkel, L. (1980) "Cancer morta. icy in. nonsmokers: prospective study by the American Cancer Society", National Cancer Inst., 65: 1169-1173. dills, C.R.; Hole, D.J.; Havthorni!, V.M. and Boyle, P. (1983) "The effect of environmental tobacco smoke in two urban communities in the vest of Scotland", Proceedings from Conference enilcled, "ETS -- Environmental tobacco smoke; Report from s Workshop on effects and levels", Unlv. of Geneva, March 15-17. Hammond, E.C. and Sellkoff, I.J. 0981) "Passive smoking and lung cancer vith comments on tvo new papers". Environmental Res., 24: 444-452. Heller, W.D. (1983) "Lung Cancer and Passive Smoking", Lancet, II: 1309. Hlrayama, T. (1974) "Prospective studies on cancer epidemiology based on census population In Japan", Proc. of XI International Cancer Congress, Florence Oct. 20-26. Hlrayama, T. (1983a) "Passive Smokl ng and Lung Cancer: Consistency of Association Lancet, II: 1425-1426. Hlrayama, T. (1983b) "Passive smokl ag and lung cancer", Proc. 5th World Conf. on Smoking and Health, Winnipeg, J ily 10-15, In press. Hlrayama, T. (1981a) "Nonsmoking vl ires of heavy smokers have a higher risk of lung cancer: a study from Japan", Brit Med J, 282: 183-185. Hlrayama, T. (1981b) "Passive smoklig and lung cancer", Brit Med J, 282: 1393-1394 Hoel, D.G.; Kaplan, N.L. and Anderson, M.W. (1983) "Implication of nonlinear kinetics on risk estimation In itardnogenesis", Science, 219: 1032-1037. Hoel, D. G.; Gaylor, 0. W.; Kirschs^ein, R. L.; Safflottl, U. and Schnelderman, M. A. (1975) "Estimation of Riiks of Irreversible Delayed Toxicity", J. Toxicol, and Environ. Health 1: 133-151. Interagency Regulatory Liaison Grouji (1979) "Scientific Bases for Identification of Potential Carcinogens and Est imation of Risks", Report of the Work Group on Risk Assessment, J. Natl. Car cer Inst., 63: 241-268. SPI-00259 R4. International Agency For Research on Cancer (IARC) (1979), "Chemicals and Industrial Processes Associated With Cancer In Humans',.Lyon, Monographs 1-20, Suppl. I. Ives, J.C.; Buffler, P.A. and Cree4berg, S.D. (1983) "Environmental Associations and Histopathologic Patterns od Carcinoma of the Lung: The Challege and Dilemma In Epldlmlologlc Studiois'. Am* Rev. Respiratory Disease, 128: 195-209. Jarvis, M.J. and Russell, M.A.H. (ilnl press) 'Measurement and Estimation of Smoke Dosage to Non-smokers frcam Environmental Tobacco Smoke', Brit. Med. J Rabat, C.C. and Vynder, E.L. (1984) "Lung cancer In nonsmokers". Cancer, S3: 1214-1221. Kasuga, H. (1983) 'Bydroxyprollne sjnd Passive Smoking', presented at Nev Etiologies In Lung Cancer Conference, Honolulu, Hawaii, March 21-23. Kauffmann, F.; Tessler, J.F. and Orlol, P. (2983) "Adult passive smoking In the home environment: a risk facto|r for chronic airflow limitation", American J Epidemiol., 117: 269-280. Enoch, A.; Bohn, H. and Schmidt, F. (1983) "Passive Smoking as a Causal Factor of Bronchial Carcinoma in Female Non-smokers" Med. Rlin., 78: 66-69. Roo, L.C.; Ho, J.H-C and Saw, D. (1983) "Active and passive smoking among female lung cancer patients and :ontrols In Hong Kong". J. Exp. Clin. Cancer Res. , 367-375. Lave, L. (1983) 'Quantitative Risk Assessment In Regulation", Brookings Institution, Washington, D.C. SPI-00260 IS. Leaderer, B* P.; Cain, W. S.; Issairoff, R. and Berglund, L. G. (1984) "Venti lation Requirements In Buildings -- II. Particulate HaCter and Carbon Monoxide from Clgarecte Smelting", Atmospheric Environ. 18: 99-106. Mstaukura, S., et al. (1984) "Effacta of Envlronoencal Tobacco Smoke on Urinary Codnine Excretion In Nonaaokera -- Evidence for Passive Smoking", New England J. Med. 311: 828-832. Meyer, B. (1983) "Indoor Air Quail:y" Reading: Addlaon Wesley. Miller, G.B. (1984) "Cancer, pasa1 re smoking and nonemployed and employed wives". West J Med, 140: 632-6i!5. National Association of Homebuildersa (1982) National Association of Homebuilders, Washington, D.C. National Cancer Institute (1966) Tpldemlologlcal approaches to the study of cancer and other chronic diseases", Haenzel, W. Ed., National Cancer Institute Monograph 19. National Interagency Council on Smoking and Health (NICSH) (1978) "Smoking and Che Workplace" Business Survey NICSH, N.Y., N.Y. National Research Council (1980) "The effects on populations of exposure to lov levels of Ionizing radiation: 1980, National Academy Press, Washington D.C. National Research Council (1983) "Risk Assessment in Che federal government: managing the process." National Academy Press, Washington, D.C. National Research Council (1981) "Iadoor Pollutants", National Academy Press, Washington, D. C. Ott, W. R. (in press) "Human activity patterns: A review of the literature for estimation of exposure to air pollution, U. S. Environmental Protection Agency, Washington, D.C. Pitot, H.C. (1981) 'Fundamentals of Oncology", 2nd Ed. New York, Marcel Dekker. SPI-00261 16. Phillip*, R.L.; Garfinkel, L.; Kuzsa, J,W.; Beeson, W.L.; Lotz, T. end Brin, B. (1980) "Mortality among California seventh-day*adventists for selected cancer sites" J Natl Cancer Inst., 6: 1097-1107. Phillips, R.L.; Kuzma, J.tf.; Beeson, W.L. and Lotz, T. (1980) "Influence of selection versus lifestyle on rlslc of fatal cancer and cardiovascular disease among seventh day adventists", American J Epidemiol., 112; 296-314. Relf, A. (1981a) "Effect of cigarette smoking on susceptibility to lung cancer". Oncology, 38_: 76-85. Relf, A. (1981b) "The Causes of Canter", American Scientist, 6J): 437-447. Repace, J.L. (In press) "Risks of pisslve smoking" Center for Philosophy and Public Policy "To Breathe 7reel;r", University of Maryland, College Park, MD. Repace, J.L. (in press) "The Dosime :ry of Passive Smoking"t 5th World Conference on Smoking and Health, Winnipeg July 10-15. Repace, J.L. (1984) "Consistency of Research Data on Passive Smoking and Lung Cancer", Lancet, II: 506. Repace, J.L. (1982) "Indoor air pol.. utlon" Environment International, 8^:21-36. Repace, J.L. (1981) "The problem of passive smoking". Bull N.T. Acad, of Med, 57: 936-946. Repace, J.L. and Lovrey, A.H. (1984] "A Proposed Indoor Air Quality Standard . for Ambient Tobacco Smoke", proceedings of the 3rd International Conference on Indoor Air Quality and Cllmat e, Stockholm, August 20-24. Repace, J.L. and Lovrey, A.H. (1983] "Modeling exposure of nonsmokers to ambient tobacco smoke", Proceedings of the 76th Annual Meeting of the Air Pollution Control Association, Atlanta, Ju ne 19-24. Repace, J.L. and Lovrey, A.H. (1982) "Tobacco smoke, ventilation, and indoor air quality" ASHRAE Transactions, 88 : 894-914. SPI-00262 R7. Repace, J.L.; Seba, D.B.; Lowrey, 4.H. and Gregory, T.W. (Winter 1983-1984) "Effect of negative Ion genera|c:ors on eabienc tobacco evoke", J. Clinical Ecology II: 90-94. Repace, J.L. and Lowrey, A.H. (1980) "Indoor air pollution, tobacco smoke, and public health", Science, 2)8: 464-472. Repace, J.L.; Ott, W.R. and Wallace, L.A. (1980) "Total human exposure to air pollution". Proceedings of The 73rd Annual Meeting of the Air Pollution Control Assoc., June 22-27, Montreal, 1980. Repetto, M. and Martinez, K. (1974] "Benzopyrene de Cigarettes et son excretion urinal re", J Europeen de Toxicclogie, 7j 234-237. Sandler, D.P.; Everson R.B. and Wil cox, A.J. (In press, a) "Passive Smoking in Adulthood and Cancer Risk", American Journal of Epidemiology. Sandler, D.P.; Everson, R.B.; Wllccx, A.J.; Browder, J.P. (In press, b) "Cancer Risk In Adulthood from Early Life Exposure to Parents' Smoking, Sellkoff, I.J. (1981) "Household rl sks with Inorganic fibers", Bull N York Acad Med, 57: 947-961. Senate Committee on Environment and Public Works (1977) "TJ.S. Clean Air Act, As Amended August 1977", 95th Conglress, 1st Session, U.S. Government Printing Office, Washington, D.C., Serial Mo. 95-11. Szalai, A. (1972) "The use of tlme|: dally activities of urban and suburban Populations In 12 countries", Mouton, The Hague, Paris Tager, I.; Weiss, S.T.; Munoz, A.; ILosner, B. and Speizer, F.E. (1983) "Longitudinal study of the effects of maternal smoking on pulmonary function In children", N Engl J Med, 309 699-703. Tobacco Institute (1984),"New Natioial Survey of Smoking and Productivity in the workplace", Tobacco 0bservr 9: 6-7. SPI-00263 R8. Trichopoulos, 0.; Lalandidi, A. anc Sparros, L. (1983) "Lung Cancer and Passive Saoklng: Conclusion of Creek Study", Lancet, II: 677-67$. Trichopoulos, D.; Kalandldl, A.; Sparros, L. and MacMahon, B. (1981) "Lung cancer and passive saoklng". International J Cancer, 27,: 1-4. 0. S. Dept, of Coaaerce (1980) "Statistical Abstracts of the United States: Table 653, Labor force parti dpRation rates by Race, Sex, and Age: 1960 to 1979", O.S. Dept, of Commerce, Sureau of the Census, Washington, D.C. U. S. Dept of Health and Human Services (1983) "Advance Report of Pinal Mortality Statistics 1980", In Monthly Vital Statistics Report Vol. 32, U. S. Department of Health and iluaan Services, National Center for Health Statistics, Washington, D.C. U. S. Dept of Health and Human Services (1981) "State Legislation on Saoklng and Health 1980", U.S. Dept, of Health and Huaan Services, Public Health Service, Atlanta, Ga. U. S. Dept, of Health and Huaan Services (1979) "Changes in Cigarette Saoklng and Current Smoking Practices Along Adults: United States, 1978" Advance Data No.52, U.S. Dept. Health aid Huaan Services, Washington, D.C. U. S. Dept, of Health, and Welfare <1975) "Vital Statistics of the United States, Life Tables, 1974", U.S. Dept, of Health, and Welfare, National Center for Health Statistics. Washington, D.C. U. S. Environmental Protection Agency (1984) "Carcinogen assessment of coke ovae emissions', Final Report EPA 600/6-82 003F, U. S. Environmental Protection Agency, Washington, D.C. U. S. Environmental Protection Agency (1983a) "Health assessment document for acrylonitrile", EPA 600/8-82-007 F, p. 13-178, U. S. Environmental Protection Agency, Washington, dLc. SPI-00264 R9. IT* S* Environmental Protection Agenty (1983b) "Radionuclides, Background Information Document", leport EPA/52O/1-83-J01, U. S. Environmental Protection Agency, Washington, D.C. U. S. Environmental Protection Agency (1982) 'Coke Oven Zaiealons, Background Information Document", U.S. Environmental Protection Agency, Vaahington, D.C. U. S. Environmental Protection Agency (1980) "Pinal Rlak Assessment on Arsenic", U. S. Environmental Protection jigency, Washington, D.C. U. S. Environmental Protection Agency (1979) "National Emission Standards for Identifying, Assessing, and Regelating Airborne Substances Posing A Risk of Cancer", Pederal Register 44: 5(642-58661. U. S. Environmental Protection Agea<y (1979) "Pinal Risk Assessment on Benzene" U. S. Environmental Protection jgeney, Washington, D.C.' U. S. Environmental Protection Agency (1975) National Emission Standards For Hazardous Air Pollutants: Proposed Standard Por Vinyl Chloride", Pederal Register 40: p. 59532. 'U.S. Federal Trade Commission, (1984) "Report of 'Tar', Nicotine and Carbon Monoxide of the Smoke of 187 Varieties of Cigarettes", Washington, D.C. U. S. Public Health Service (in press) "Lung cancer mortality rates for 1980* U. S. Public Health Service, Division of Health Interview Statistics, Washington, DC, U. S. Surgeon General (1982) "The He|a2th Consequences of Smoking: Smoking and Cancer", U. S. Dept, of Health & Human Services, Washington, D.C., DHHS Pub. No. 82-50179, Govt Printing Off!lice. U.S. Surgeon General (1979), "Smoking and Health", U.S. Dept, of Health, Education, and Welfare, DHEV Pub . No.(PSS) 79-50066. Wald N.J., et al. (1984a) "Urinary c itinlne as marker of breathing other people's tobacco smoke" Lancet I: 230-231. SPI-00265 UO. Wald, H.J. ud Ritchie, C. (1984b) "Validation of acudlaa on lung canear In nonaaokars Berried to aaokars.* Tha Lancet 1: 1067. * Wald. H. J. (2978) "Saoklng aa a cause of disease" A. E. Bennett, Ed., "Recent Advance In Coaaunlty Vol. I", Edinburgh, Church Bill Livingstone White, J.R. and Froeb, H.F. (1980) TSaall alrvays dysfunction In nonsaokers chronically exposed to tobacco naoke", New Engl J Med, 302: 720-723. World Health Organisation (1979) "Hiialth Aspects Related to Indoor Air Quality", Euro Reports and Studies 21, Regional Office for Europe, Copenhagen, Denaark. Wynder, E.L. and Goodaan, M.T. (1983) "Saoklng and Lung Cancer: Soae Unresolved Issues", Epldealologlc Revs, 5: 1177-206. SPI-00266