Tobacco Documents Online

Evidence for Health Effects of Sidestream Tobacco Smoke J.D. Spengler, Ph.D. M.L. Soczek Exposure to tobacco smoke is widespread'; nearly everyone is exposed at one time or another. Many of us are exposed on a regular basis in homes, offices, vehicles, and public places. -- Many of the over 2,000 identified compounds in tobacco smoke are established carcinogens, asphyxiants, and eye, nose, or respiratory irritants. While health effects of smoking on smokers have been clearly and carefully documented, only recently have investigations into the possible health effects of sidestream tobacco smoke on nonsmokers been undertaken. This paper reviews recent research, including studies on the influence of parental smoking on the respiratory health of children and the incidence of lung cancer in women with smoking hus- bands. While results from some of the studies are unclear, consistent relationshins between -respira[ory health and tobacco smoke exposure have been documented. The need for a careful' evaluation of the public health significance of exposure to sidestream tobacco smoke is emphasized. ' . INTRODUCTION V " more women, and particularly teenage giris, are smoking. Additionally, the amount and fre- quency of passive smoke exposure have been shown to vary with other socioeconomic and'ethnic factors. Estimating the population exposed to sidestream tobacco smoke requires knowledge of the smoking population and activity patterns of both the smoking and nonsmoking popula- tions. For example, although recent statistics indicate that 33% of the adult population smokes, a series of studies in several cities across the country revealed that a range of 54% to 76% of children live in homes with one or more smokers (p?RC/NAS 1981). t;'hile the prevalence of regular cigarette smoking in the adult population is decreasing, `health of both adults and children. ''`' In '11978, an estimate&54 million persons smoked a total of 615 billion cigarettes. ,carcinogens; asphyxiants, or eye, nose, and respiratory irritants. Additionally, there is increaoing`evidence that passive exposures to tobacco smoke may be affecting the respiratory Exposure to tobacco smoke is widespread; nearly everyone is exposed at one time or another. aany people.are exposed on a regular basis in homes, offices, vehicles, an&public places. Over 2,000 compounds have been identified in cigarette smoke, many of which are establishe& INDO!3F, CONCENTRATIONS Tobacco smoke can be singled out as an indoor contaminant, as it contributes to indoor con- centrations of respirable particles, nicotine, pol'ycyclilc-aromatic hydrocarbons, carbon Q monoxide, acrolein, nitrogen dioxide, and'many other substances. As with other sources of ~ J.D. :'pengler, Associate Professor, and M.L. Soczek, Research Assistant, Department of ~ Environmental Health Science and Physiology, Harvard School of Public Health, 665 Huntington ~ Avenue, Boston, MA 02115 . }~ CD THIS PREPRINT FOR DISCUSSION PURPOSES ONLY. FOR INCLUSIONYN ASHRAE TRANSACTIONS 1984; V.90. Ptl 1. Not to be reprinted in whole onin part,without written permission of the American Society of Heating„Refrigerating and Air• Condi- tioning Engineers, Inc.. 1791 Tullie Circle NE, Atlanta;,GA 30329. Opinions, findings, conclusions, or recommendations ex. pressed in this paper are those of the author(s)and do not necessarily reflect the views of ASHRAE. 1 4

i n indoor air pollution,, resulting,concentrations of various tobacco smoke contaminants depend on the frequency and amount of smoking, air infiltration rates,, air-cleaning, and air-distri- bution systems. Therefore, it is not surprising that reported indoor concentrations of tobacco smoke vary widely. In-home measurements of tobacco smoke contaminants have found carbon mon- oxide (CO) ranging from 2 to 3'5 ppm and total' suspende&particulate matter (TSP) from 10 to 1,00Oug/m3 (NRC/NAS 1981; Repace and Lowery 1980). Most of these measurements were of short duration and should be considered only as representative of the spatial and temporal range of concentrations that might be encountere&. Repace and Lowery (1980) measured respirabl~e par- ticle concentrations in many work settings and estimated the amount of particulate matter that the smoke-exposed office worker inhale&to be three times that of the nonsmoke-exposed worker. Analyses of 24-hour integrated respirable particle samples from homes with and without smokers indicated that an average "pack-a-day"'smoker contributes 20yg /m3 to indoor particulate I'evels (Spengler et al. 1981). The annual and the 24-hour National Ambient Air Quality Standards (NAAQS) for TSP were exceeded'in several of the homes with two smokers. The health effects of smoking on smokers have been studied extensively, but the health effects on nonsmokers have received liess attention (U.S. Dept. of Health, Education and Wel- fare 1979). Nonsmokers are exposed to two components of passive smoke: the mainstream in- haled and then exhaled smoke and the sidestream~smoke. Sidestream and mainstream smoke account for about equal proportions of tobacco burned, but sidestream smoke has more tar, nicotine, CO, particles, and non-water-soluble organic vapors than does the exhaled mainstream smoke. I'RRITATIiOYAND RESPIRATORY DISTRESS Many of the substances in cigarette smoke are irritants. The degree of sensitivity to the passive smoke depends, not only on the amount and type of smoke an&other environmental fac- tors, such as dryness, but also on the psychological aspects of the setting and persons smoking. In addition, as with other irritants, there is a variation in the degree of sen- sitivity to cigarette smoke. In a survey of 21,366 Social Security Administration employees, Barad (1979) found high percentages of nonsmoking workers who experienced conjunctivial irri- tation (47.7%), nasal discomfort (34:7%)~, and cough, sore throat, or sneezing (30.3%) when exposed to cigarette smoke. These results are compatible with results fromithe chamber . studies of Cain and Lead'erer (11982) indicating that an air-exchange rate of 35 cubic feet per minute (.02 cubic meters per second) per person, the current recommended ventilation standard for environments permitting,smoking, would not be sufficient to prevent an annoyance response over 20%. There have been several chamber or controlled-exposure studies conducted on nonsmokers ,expose&to passive smoke. In genera.'., the observed changes in heart rates, systolic blood pressure, pulmonary functions, carboxyhemoglobin, and psychomotor functions have been slight (Laties and Herigan 1979; Carbon Monoxide 1977; Yavroff et al. 1974). Most of these studies, however, were concerned'onl'y with CO. Surveys of CO in smoke-filled environments reported' values in the 2 to 15 ppm range, slightly less than most urban dri'ving exposures (Cortese and Spengler 1976). With multiplie exposures from passive smoke, however, other sources may be sufficient to elevate COHb in excess of 2%, the level at which changes in acoustic and visual vigilance have been reported (Carbon Monoxide 1977). PARENTAL SMOKING AND HEALTH EFFECTS ON~CHILDREN - .. ., <.rr _ .. . . .. . . . ._.. . The influence of parental, smoking in the home on the respiratory health and postnatal condi- tions of children has been recently investigated by several researchers:. The majority of the studies reported an association between reported!respiratory morbidity in children and paren- tal smoking. However, these relationships did not appear independent of parental symptoms, socioeconomic class, and the smoking habits of the child'remthemselves. On the other hand,, consistent associations between symptoms and liower birth weight and the numb~j r of cigarettes smoked per d'ay by the parents have beenifound. v373S IZO The relationship between parental smoking and childhood symptoms (such as persistent wheeze) and excessive risk of respiratory infection are stronger in the first year or two of life. Tager et al. (1979) reported a dose-d'ependent decline in force expiratory flow (FEF0.25-0.75) in children of smoking parents. They also noted that pulmonary infection in early Life has been shown to adverseliy affect pulmonary functions in childremand adults. This work by Tager et al. supports earlier work by Colley et al. (1974), an&Leeder et aL. (11976a; 1976b; 1976c) in England. Coll,ey, reporting on pneumonia and bronchitis in 2205 children~during their first five years of life, found that a rel!ationship between parental smoking habits and respiratory infection occurred only during the first year. The rate of 1

( respiratory infection increased with~number of smokers and number of cigarettes smoked per day in the home and was independent of parental symptoms, social class, and birth weight. In more recent work, the relationships between childhood illness, lung functions, and parental smoking (maternal) have been further demonstrated with large samples of U.S. children. Hasselblad et al. (198'1i) studied over 16,000 children aged 6 to 13'in 31 communities. A sig- nificant linear relationship between decreased adjusted lung function and maternal smoking, as measured'by packs per day, was found. The effect of maternali smoking appeared in both sexes and age groups, as shown in table 1. It is important to note that Hasselblad et al. (1981) reported the maternal smoking, association with measures of forced'expiratory volume (FEV0.75 and FEV0.25-0.75) for children aged 6 to 13. With initial data on 8000 children aged 6 to 9 years from the Harvard Air Pollution/L,ung Health Study, Speizer et al. (1980) did not find an association between mater- nal smoking and children's FEV or forced vital capacity (FVC). In followup work, however, Ware et al. (1983), using an expanded data base and slightly different adjustments for social status, age,' height, and sex of the children, found that lung functions and several respira- tory symptoms were significantly associated with the presence and amount of maternal smoking. The excessive illness before age 2('as measured by reports of bronchitis, respiratory infection " in the past year, cough, wheeze, alone or in combination with a combined lower-respiratory- tract index of disease)' was 15: to 35% higher for the children liiving withimothers who smoked. Controlling for reporte&parental symptoms lowered the rates, but they remained significantly greater for the smoke-exposed children. No direct exposure data were reported, but the re- lationships appeared to increase with reported'smoking levels. The relative odds for smoke- exposed children having a respiratory illness or symptoms by mothers' daily smoking levels are compared in table 2. The trend'for increased risk of disease with increased maternal smoking is clear. In a further study of the Tager et al. sample of 5- to 9-year-old children, Weiss et al. (1980) reported a linear relationship of parental cigarette smoking to persistent wheezing in chiLdreni(p e 0.012) and'to lower mean forced mid-expiratory flow. The relationship between parental smoking and wheezing iniehildren held when controlling for parental symptoms, as shown initable 3. Evidence that passive smoke exposures may be influencing small-airways dysfunction in adults comes from a 1980 study of nonsmoking office workers conducted by White and Froeb. Slightly lower unadjusted mid-expiratory (FEF0,25-0.75) and'end-expiratory (FEF0.75-0. E5) flow rates were found'lin smoke-exposed office workers than in nonsmoke-expose&workers. This re- lationship persisted when controlling for effects of home smoking and other occupational' ex- posures. Adjusting the data for age, sex, and height reduced the statistical, significance of the relationship, but, nevertheless, the data suggested a deleterious effect of smoke exposure in the workplace on the smoker. CARCINOGENICITY Hirayama et al. (1981), and the other a case-control of 51 women with lung cancer and 163 other patients in Greece by Trichopoulos et al. (1981) investigated the rates of lung cancer in nonsmoking wives with~smoking husbands. Each study found statistically significant associ- ations between wives' relative risks of lung cancer and the amount of husbands' smoking, as indicated by packs per day. The observed lung cancer rates from the Japanese study are shown in table 4. While increased lung cancer risks were noted in nonsmoking spouses in an American study by Garfinkel (1981), only in the Japanese and'Greek studies were clear exposure-response relationships observed. An important caveat for this and other passive smoking studies, particularly those in- volving children, is that some subjects may in fact have been smokers. For example, women in Japan do not smoke freely in public. Since lung cancer rates are much higher among smokers, any misclassification of subjects would prevent accurate examination of actual relationships between totally passive smoke exposure and lung cancer. Additionally, lifetime rates cannot be established until all members of a cohort have died. Hammond and Selikoff (1i981) examined these and other studies and concluded that the reported evidence of exposure-response relation- ships was sufficient to warrant further investigation. Two"important studies, one involving 91,540 nonsmoking wives, aged 40 an&above, in Japan by DISCUSSION 03735121 In summary, children liiving in homes with smoking,parents appear to have higher rates of re- spiratory symptoms and illnesses. The Bri'tish stud'ies found the more significant associations between parental smoking and respiratory symptoms in early chilldhood, while the recent U.S.

C 0 studies reported associations in children 5 to 13 years old. Several studies associated lower lung functions with the amount of maternal smoking. A relationship between chronic air- way restriction and cigarette smoke irritation is possible. Relationships between smoking and childhood respiratory illnesses in generai, however, have not been found to be entirely in- dependent of parental symptoms, parental smoking,during pregnancy, or birth weight of the children. Evidence from studies on adult samples indicates that occupational smoke exposure might adversely affect general respiratory health. In-home exposures may alsolincrease risks of lung cancer. . Unfortunately, in-home and personal exposure data were not available for any studies, so the reported dose-response relationships were based only upon crude categorizations of paren- tal smoking habits. The home-to-home variation in particulate concentrations is quite large among the homes with smokers. Until better direct exposure data (including information on chemical composition of the particulate matter) are available, precise recommendations for acceptable indoor air quality with respect to tobacco smoke can not be made. This does not imply that reduction in exposure would be ineffective in reducing risks. Elimination or re- duction of smoking,, increased ventilation, and particle filtration may be expected to lower indoor exposures. Based on current evidence, health benefits woul&accrue to the nonsmoker. CONCLUSION Estimating parameters of the nonsmoking population exposed to cigarette smoke in an effort to first define, and then perhaps reduce, the population-at-risk is difficult. Developing an overall strategy to reduce exposures and protect the public health, is problematic (Spengler and Sexton 1983)', as institution of source removal/ modification or air purifiration in public buildings to reduce involuntary exposures is feasible, but promulgation a regula- 1tory framework for residential and private buildings, where exposure is largely voluntary, is .questionable. REFERENCES Barad, C.B. 1979. "Smoking on the job: The controversy heats up." Occupational Health and Safet 48 (1), pp. 21'-24. Cain, W. and Leaderer, B. 1982. "Ventilation requirements in occupied space during smoking and nonsmoking occupancy." Environment International 8, pp. 505-514 Carbon monoxide. 1977. Committee on Medical and Biological Effects of Environmental Pollutants, National Research Council. Washington, DC: National Academy of Sciences. Colley, J1.R.T.; Holland, W'.W.; Corkhill, R.T. 1974. "Influence of passive smoking and r- parental phlegm on pneumonia and bronchitis in early childhood." Lancet 1974 2, pp. 1031-1034. - Cortese, A. and Spengler, J.D. 1976. "Ability of fixed monitoring stations to represent personal carbon monoxide exposure." Journal of the Air Pollution Control Asso- ciation 26 (12) pp. 1144-1150. ,Garfinkel, L. 1981. "Time trends in lung cancer mortality among non-smokers and'a note on passive smoking." Jburnal of the National Cancer Institute 66(6) pp. 1061-1066. Hammond', E.C. and Selilkoff, I.Ji. 1981. "Passive smoking and'lung cancer with comments on two new papers." Environmental Research 24 (4), pp. 444-452. Hasselblad, V.; Humble, C.G.; Graham, H'.G.; andiAnderson„ H.S. 1981. "Indoor environmental determinants of lung function in children." American Review of Respiratory'Disease 123 (5)', pp. 479-485. O3'735122 Hirayama, T. 1981. "Non-smoking wives of heavy smokers have a higher risk of lung cancer: A study from Japan." British Medical Journal 282, 6259', pp. 183-185. Lati~es, V.G. and Herigan, W.H. 1979. "Behavioral effects of carbon monoxide on animals and man." Annual Review of Pharmacology and Toxicoliogv 19, pp. 357-392.. Leeder, S.R.; Corkhill!, R.; Irwig, L.M.; Hol'land'!, W.W.; and Colley, J.R.T. 1976a. "In- }

Leeder, S.R.; Corkhill, R.; Wysocki, M.Ji.; Holland, W.W.; and Colley, J.R.T. 1976c. "Influence of personal and family factors on ventiliatory function of chilidren." British Journal of Preventive Social Medicine 30 (4), pp. 219-224. Leeder, S.R.; Corkhill, R.; Irwig,, L.M.; Holland, W.W.; and Colley, J.R.T. 19.'6b. "In- fluence of family factors on asthma and wheezing during the first five years of life." British Journal of Preventive Social Medicine 30 (4) pp. 213-21i8. NRC/NAS. 1981., Indoor pollutants. National Research Council/National Academy of Sciences. Washington, DC: Nationali Academy Press. Repace, J.L. and Lowery, A.H. 1980. "Indoor air pollution, tobacco smoke and public C fluence of family factors on the incidence of lower respiratory illness during the first year of life." British Journal of Preventive Socia3 Medicine 30 (4), pp. 203-212. . health." -Science, 208,,4443, pp. 464-472. Speizer, F.E.; Ferris, B.J.; Bishop, Y.M.M.; and Spengler, J.D.1980'. "Respiratory disease "" rates and pulmonary function in children associated with N02 exposure." American~ Review of Respiratory Disease 121 (1), pp. 3-10. Spengler, J.D.; Dockery, D.W.; Turner, W.A.; Wolfson, J.M.; and'Ferris, B.C., Jr. 198'1i. "Long-term measurements of respirable sulfates and particles inside and outside ; homes."' Atmospheric Environment 15 (1), pp. 23-30. Spengler, J.D. and'Sexton, K. 1983. "Indoor air pollution: A public health perspective." t Science 221, 4605, pp. 9-17. 1 .~'- f r.,, j4^ rcigarette smoking on the pulmonary function of children." American Journal of ' Tager,`iI.B.; Weiss, S.T.; Rosner, B.; and Speizer, F.E. 1979. "Effects of parental EpidemioLogy 110 (1) , pp. 15-26. Trichopoulios, D.; Kalandidil, A.; Sparros, L.;, and MacMahon, B. 1981. "Lung cancer and' ;~and health: A report of the surgeon general." (PHS-79-50066). Washington, DC: ;passive smoking." International Journal of Cancer 27 (1), pp. 1-4. U.S. Department of Health, Education and Welfare, Public Health Service. 1979. "Smoking Ware, J.M.; Dockery, D.W_; Spiro, A.; Spei~zer, F.E.; and Ferris, B,Ji. 1983. "Passive Government Printing Office. Weiss, S.T.; Tager, I.B.; Speizer, F.E.; and Rosner, B. 19801. "Persistent Wheeze. Its ;,smoking, gas cooking and respiratory health of children living in six cities."' American Review of Respiratory Disease, in press. • White, J.R. and Froeb, H.F. 1980. "Sma1i1 airway dysfunction in nonsmokers chronically relation to respiratory illness, cigarette smoking and level of pulmonary function . in a population sample of children."' American Review of Respiratory Disease 122 .(5), pp. 697-707. exposed to tobacco smoke." New England Journal, of Medicine 302 (13), pp. 720-723. Yabroff, Ii.;'Meyers, E.; Fend, V.; David, N.; Rotiertson, M.; Wright, R.; and Braun, R. 1974. The role of atmospheric carbon monoxide in vehicle accidents. Menlo Park, CA: Stanford Research Institute. ACKNOWLEDGEMENTS - Work for this paper was conducted in part under Electric Power ResearchlInstitute and TRC Environmental Consultants, Inc. Contract RP1948-1: "Evaluation of Indoor Air Quality Data for Making Risk Assessment" and under Harvard University and1UT.S. Environmental Protection Agency Cooperative Agreement CR808530101. ti

C TABLE 1 Mean FEVp,.75 Measurements Adjusted for Age and Height by Covariates Maternal Smoking Boys Girls 6-9 9-13 6-9 9-10 Never or ex-smoker 1.315 1.833 1.211 1.734 (sample size) (11933) (2949) (1777) (2408) i < 1 pack/day 1.301 1.825 1.195 1.729 i (650), (890) (597) (772) + d from never -0.014 -0.008 -0.016 -0.005 i 1 k d z pac per ay 129 5 1 806 195 L 72 '2 1. ~. (1016) (1470) (977) i (1310) A from never -0.020 -0.016 -0.016 -0. 12 .. .~);cr... , . . .. (Source: Hasselbliad, V.; Humble, C.G.; Graham, H.G.; and Anderson,, H~.S. 1981. "Indoor environmental determinants of lung function in children." American Review of Respiratory Disease 123,5, pp. 497-485. TABLE 2 Relative Odds of Respiratory Ililness and Symptoms by Mother's Reported Daily Smoking N 0 MATEKNAL S4iOK]NG Current Daily Smoking 1-5 6-15 16-25 26-35 3G-45 4fi+ Lincnr Trcnd ~ P Doctor Ding 7273 1.00 1.U'1 1.0`) 1.42 1.415 1.C>h 2.(i''.) 1 ILE12! .fD000 Resp III Before 2 History of Bron- 7944 1.00 1.03 1.19 1.20 1.42 1.26 2.04 13.68 .0002' chitis Resp Illness . 7507 1.00 1.58 .97 1.15 1.56 1.01 1.07 2.95 .086 Last Year (1st Exam) Resp Illness 6613 1.00 1.06 1.07 1.15 1.42 1.64 1.26 ' 13.13 .0003 Last Year (2nd Exam) Bronchitis Last 6813 1.00 1.74 1.24 1.25 1.08 1.63 1.58 6.34 .012 Year Cough Last 6594 1.00 1.01 1.30 1.11 1.46 2.12 1.24 11.52 .00U1 Year ~_~._ . . Wheeze Last 6576 1.00 1.17 1.11 1.27 1.39 2.31 1.35 24.94 ' .0000 Year Lower Resp In- 8578 1.00' 1.15 1.09 1.20 1.29 1.78 1.E3'r' 2;i.E52 .0000 dex (Source: Ware, J.M.; Dockery, D~.W.; Spiro, A.; Speizer, F.E.; and Ferris, B.C. 1983. "Passive smoking, gas cooking and respiratory heal!thlof chil- dren living inisix cities." American Review of Respiratory Disease, in press.) '

C TABLE 3' Percent of Wheeze in Children by Level of Parental' Smoking Childhood Parents Smoking Symptom None One Both~ Current Persistent Wheeze 1.9 6.8 12 Without Wheezing Mothers 0 1'.8 7.7 Without Wheezing Fathers 0 6.7 14 (response/sample size) (1/57) (10/146) (20/169) (Source: Weiss, S.T.; Tager, I.B.; Speizer, F.E.; and Rosner, B. 1980'. "Persistent wheeze, it's relation to respiratory illness, .cigarette smoking, and level of pulmonary functionlin a population sample of children."' .american Review of Respiratory Diseases 122,5, pp. 697-707. TABLE 4, Lung Cancer Incidence in Wives with Smoking Husbands Husbands' Smoking Wives' Lung Cancer Rates j 0 packs per day 1.58 x 1'0"4 per year 1 pack per day 2.44 x 110"{ per year 1+ packs per day 2.96 x 110'4 per year (Source: Hirayama, T. 1981. "Non-smoking wives of heavy smokers have a higher risk of lung cancer: A study from Japan." British Medical Journal 282, pp. 183-185.)

RECE1~ rE8,41984 A~E!'E~rs