Tobacco Documents Online

I I I I ! LAW OFFICES SHOOK, HAIrY& BACON I I I ENVIRONMENTAL TOBACCO SMOKE & INDOOR AIR QUAI.ITY COMPILATION 1 I I I I I I I I I I I SHB

I I I I I I l I I I I I I I I i I I l -- TABLE OF CONTENTS -- LUNG CANCER .................................................................................................... .................. 1 CARDIOVASCULAR ISSUES ................................................................................... 11 RESPIRATORY DiSEaSES aND CONDITIONS - ADULTS ............................................... 16 RESPIRATORY DISEASES AND CONDITIONS - CHILDREN .......................................... 24 OTHER CANCER .................................................................................................... .............. 41 OTHER HEATH ISSUES .................................................................................................... ...... 45 ETS EXposuP~ aND MONITOPdNG ................................................................................ 52 INDOOR BAR QUMATY .................................................................................................... ...... 62 SMOKING POLICIES AND RELATED ISSUES ....................................................................... 75 STATISTICS AND RISK ASSESSMENT ................................................................................... 79 INDEX .................................................................................................... ......... 85 Please note: This document is not intended to be an encyclopedic summary of scientific literature relating to ETS. Rather, it contains exclusively those studies included in Reports on Recent ETS and IA Q Developments. This document has been prepared for use by Shook, Hardy & Bacon attorneys and analysts. Please, do not distribute this document to persons outside the firm.

I VOLUME. 1[, ISSUES 1-24 JULY 1992 I I ENVIRONMENTAL TOBACCO SMOKE & INDOOR AIR QUALITY i I i i I I I I I i I I i I I LUNG CANCER [I] Letters to the Editor Regarding "Lung Cancer in Nonsmo "klng Women: A Multicenter Case- Control Study," E.T.H. Fontham, P. Correa, A. Wu-Williams, P. Reynolds, R.S. Greenberg, P~. Buffler, V.W. Chen, P. Boyd, T. Alterman, D.F. Austin, J. Liff and S.D. Greenberg, Cancer Epide- raiolo~y, Biomarkers & Prevention 1." 35-43, 1991 [Issue 23, Item 17] Cancer Eioidemiolog7, Biomarkers &Prevention recently published four letters concerning this ardcle, which is a preliminary report of an ongoing U.S. case-control study that currently includes data on 420 female lung cancer cases. As discussed in issue 12 of this Report, December 10, 1991, the authors reported a statistically significant odds ratio for adenocardnoma assodated with spousal smoking. The series begins with a letter from Nathan Mantel, followed by a reply from the authors of the study, then continues with a letter from Peter N. Lee, also followed by a reply from the authors of the study. The letters appear in Cancer Epidemiolag7, Biomarkers & Prevention 1: 331-334, 1992. Mantel stated that certain cotinine data presented in the study indicated misclassification of smoking status. Mantel also noted that possibly "extravagant" bias could be introduced in the study because next of kin provided information for 34 percent of the cases, but did so for only 10 percent of one control group. Moreover, when next of kin provided information, urinary cotinine levels could not be measured in the study partidpants. Finally, Mantel critidzed Fontham, et al., for focusing on only adenocarcinoma, instead of treating all lung cancer cell types: "[I]fthese investigators have had their choice of which type of lung cancer to emphasize, their statistical significance levels should be modified to take the mul- tiple-testing aspect into amount." In their reply to Mantel, Fontham, et al., stated that their identification and exclusion of women with high urinary cotinine values "can only be considered a strength of the study." They also noted, with regard to the next of kin concern, that "estimates of relative risk did not differ in analyses restricted to self or proxy respondents." Finally, Fontharn, et al., stated that they did not have a choice of which histological ~¢pe to emphasize "because most cases turned out to be adenocardnomas a_fi:er histological review." They pointed out that the number of cases with other cell types of cancer was too small to allow reasonable statistical power in specific analyses. Legs letter commented that some data on cotinine and on lung cancer cell type were incomplete in this interim report. He noted that possible confounding factors (e.g., occupation, diet, medical history and other exposures) had not been taken into account. Lee said that confound- ing could also be due to inclusion of unmarried women in the analysis of spousal exposure, never-employed women in the analysis of occupational exposure, and to an unadjusted index ofsodal exposure. According to Lee, this could lead to "an inevitable confusion of possible effects of ETS with possible effects of marital status, occupation, and sodabiliry." Lee also stated that he calculated a relative risk for nonadenocardnoma lung cancer which was not significantly different statistically from the relative risk for adenocardnoma calculated by the authors. According to Lee, this failed to "justify the spedal attention given to the adenocarcinoma results." Finally, Lee questioned the biological plausibility of an elevated risk of adenocarcinoma associated with ETS exposure, "given that the assodation of active smoking with adenocarcinoma is so weak." Fontham, et al., replied to Lee by stating that "the availability of a large data set with which to address an unresolved issue of great public health importance was compelling justification for publishing a report" that represented only three years of a five-year study. They stated that a number of potential confounders, including age, race, geographic region, respondent type, income and education had been considered, and that other potential risk factors "will be examined in further analyses." Specifically, Fontham, et al., commented on an ongoing analysis of dietary factors, and stated that g-carotene has not appeared to be related to spousal smoking habits in this study. N4rith regard to the inclusion of unmarried women in spousal smoking calculations, Fontham, et al., calculated risk estimates with those subjects excluded that were only slightly louver than the original estimates. Exclusion of never-employed women from workplace calculations resulted in risk estimates ~vhich were elevated somewhat compared to the original calculations. Finally, Fontham, et al., defended their "special attention" to adenocarcinoma because of the large proportion of SHB

ETS AND IAQREFERENCE JULY 1992 adenocarcinoma in the study~They also proposed that "exposure to sidestream smoke might result in a distribu- r.ion of histological types of lung cancer different from that associated with exposure to mainstream smoke." [2] "Passive Smoking and Canine Lung Cancer Risk," J.S. Reif, K. Dunn, G.K. Ogilvie and C.K. Harris, American Journal of Epidemiology 135 (3): 234- 239, 1992 [Vol. I, Issue 19, April 10, 1992] This article, published in a major peer-reviewed epidemiologic journal, reports on a case-control study (51 cases, 83 controls) of lung cancer in pet dogs. The authors reported a weak relation between exposure to a smoker in the home and lung cancer in pet dogs (odds ratio of 1.6, 95% CI 0.7-3.7). The authors propose that epidemiologic studies of pets may "add to our under- standing of environmental tobacco smoke effects in human populations." One of the authors of the study, John S. Reif, was coauthor with Ross C. Brownson of the 1987 study "Risk Factors for Adenocarcinoma of the Lung," which discussed ETS. EXCERPTS~ "A case-control study was conducted to determine whether household exposure to environmental tobacco smoke is assodated with an increased risk for lung cancer in pet dogs. Lung cancer cases and controls with other forms of cancer were obtained from two veterinary teaching hospitals during 1958-1987. Exposures assessed included the number of smokers in the household, the amount smoked, and the proportion of time spent indoors by the pet." "A weak assodation was found between exposure to environmental tobacco smoke and the risk of canine lung cancer .... After adjustment for age, sex, skull shape, time spent indoors, and hospital, the odds ratio rose slightly to 1.6 (95 percent CI 0.7-3.7)." "Evidence of a dose-response relation for passive smoke exposure was largely lacking." "The increase in risk found in this study in dogs corresponds reasonably well with the estimate of an increased risk for lung cancer in humans of 1.35 that was calculated in a meta-analysis of the first 13 studies of lung cancer risk and passive smoking conducted worldwide. The current study suffers from some of the same limita- tions found in the studies done in humans, i.e., small sample sizes, impredse risk estimates, and difficulties in measuring exposure." "The finding that increased canine lung cancer risk is restricted to dogs with short and medium length noses is -2- consistent with the hypothesis that the relatively effident air filtration of the tong-nosed breeds may exert a protec- tive effect for lung cancer." "The rarity of lung cancer in dogs makes a collaborative multicenter case control study the design of choice for further studies of canine lung cancer. Studies of nasal cancer, a more common form of cancer in dogs, are under way to examine the effects of environmental tobacco smoke on the nasal epithelium." [3] "Lung Cancer in Nonsmoking Women: Dietary Antioxidants," E. Fontham, R. Coates, A. Dilley, P. Reynolds, Pall. BuYer, A. Wu-Williams, V. Chen, R. Greenberg, P. Boyd, T. Alterman, D.F. Austin and P. Correa, abstract submitted for American Society of Preventive Oncology meeting, Cancer Epidemiology, Biomarkers & Prevention 1(3): 429, 1992 [Vol. I, Issue 19, April 10, 1992] This abstract presents additional data from the ongoing multicenter case-control study of lung cancer in female nonsmokers which has also reported on ETS (see Issue 12, December 10, 1991, of this report). The abstract reported that lung cancer cases consumed fewer vegetables than did controls. Dietary intake of Vitamin C or alpha- carotene was reportedly associated with a small reduction in lung cancer risk, but no such effect was reported for total Vitamin A, Vitamin E, total carotenes or beta- carotene. In the initial, preliminary report of this study, the authors reported (i) no statistically significant risk associated with spousal smoking or childhood exposure to ETS for all lung cancer types; (ii) an OR of 1.38 for adenocarcinoma assodated with spousal smoking; and (iii) statistically significant ORs for all cancers and for adenocarcinoma associated with workplace exposure to ETS. Seelssue 12 of this Report, December 10, 1991. "Most epidemiologic studies which have found a protective assodation between lung cancer risk and con- sumption of fruits, vegetables and thek micronutrient constituents have been conducted in male smokers. This large, on-going, multicenter case-control study of lung cancer in female lifetime never-smokers provides the opportunity to evaluate these dietary factors in the absence of possible confounding by smoking. A total of 273 primary lung cancer cases and 774 controls selected from the general population, all of whom were lifetime never-smokers and self-respondents, were included in this analysis. Cases consumed fewer vegetables and some fruits. Dietary intake i I ! I i 1 I i I I I i I I

"qO~'UM2., I, ISSUES 1-Z4 JULY 1992 I I i of vitamin C, with or without supplementation, was assodated with a small reduction in risk, while no such effect was noted for total vitamin A, vitamin E or total carotenes. Individual carotenes were also evaluated. No reduction in risk of lung cancer was noted for any of the three highest quardles of consumption olin-carotene, but about a 30% decreased risk was assodated with b_igh consumption of_- carotene. All dieta~ risk estimates were adjusted for age, race, study area, education, exposure to environ_mental tobacco smoke, family history of lung cancer, occupational exposures, cholesterol and total calories." [4] "Lung Cancer in Women in the Niagara Region, Ontario: A Case-Control Study," EJ. Holowaty, H.A. Risch, A.B. Miller and J.D. Burch, Canadian Journal of Public Health 82: 304-309, 1991 [Vol. I, Issue 18, March 20, 1992] In this study, 51 female patients with lung cancer and 45 matched controls were interviewed. Information was collected about acdve and "passive" smoking, occupation and residential history. The authors reported a srzong statistical assodation between active cigarette smoking and lung cancer. No assodadon was reported between lung cancer and air pollution or residential history. With regard to ETS, the authors reported no statistically significant odds ratios for lung cancer. These results are not directly comparable to the studies often cited for ETS and lung cancer because the authors did not restrict their analysis to nonsmoking women. Rather, their analyses included all cases, regardless of smoking history. "We report the findings of a case-control study about the possible etiologic assodation between the above factors (i.e. tobacco smoking, ambient air pollution, passive smoking and occupational exposures) and lung cancer among female residents of Niagara Region." "Cases [51] and controls [45] were interviewed in their own homes by an experienced interviewer, using a standardized questionnaire. Information was collected on lifedme residential and occupational history, lifetime tobacco smoldng history, exposure to passive tobacco smoke, personal and spousal exposure to occupational substances and suspect industries, and various socio- demographic variables." "Cigarette smoking was strongly associated with risk of lung cancer in Niagara women." "Overall, 90% (46/51) of cases and 78% (35/45) of controls reported some past exposure to passive tobacco smoke in their homes. After adjustment for personal lifetime cigarette consumption, there was insufiqdent statistical evidence supporting an association between tobacco smoke in the household and lung cancer (OR = 3.6; p = 0.24). Further, there was no association between lung cancer and having a mother, father or husband who smoked in the same household. Passive smoking in the household was also examined in terms of duration (years) that each subject reported exposure as a child, and as an adult. Neither exposure was significantly associated with risk after adjusting for active smoking. There were no evidence that passive smoking exposure in the workplace was assodated ,with lung cancer." "The previously reported excess of farad lung cancer in Niagara females is most likely attributable to active cigarette smoking. This study failed to demonstrate a strong association between air pollution or residential history and risk of lung cancer." [5] "Lung Cancer in Nonsmoking Women: A Multicenter Case-Control Study," E.T.H. Fontham, P. Correa, A. Wu-Williams, P. Reynolds, R.S. Greenberg, P.A. Buffler, V.W. Chen, P. Boyd, T. Alterman, D.F. Austin, J. Lift and S.D. Greenberg, Cancer Epidemiology, Biomarkers &Prevention 1: 35-43, 1991 [Vol. I, Issue 12, December 10, 1991] Th~s preliminary report of an ongoing U.S. case-control study includes data on 420 female lung cancer cases. The authors report no statistically significant risk assodated with spousal smoking or childhood exposure to ETS for all lung cancer types. They do, however, report a statistically significant ORof 1.38 for adenocardnoma assodated with spousal smoking. They also report statlsti,-~lly significant ORs both for all cancers and for adenocardnoma assodated with workplace exposure to ETS. Several individuals contributing to this article were primary authors of other ETS and lung cancer papers. Pelayo Correa reported on a small case-control study from Louisiana in a 1983 paper; Elizabeth Fontham collaborated on that study. Earlier this year, Anna Wu-Willlams published a paper on a large case-control study conducted in northeam China. Patrida Buff!er reported on a case-control study done in Texas in a 1984 paper and additional results were published by Shaw, et al., in 1991, as reported in Volume I, Issue 4 (July 12, 1991) of this report. "The association be~veen exposure to environmental tobac.m smoke and lung cancer in female 1Ketime nonsmok- ers was evaluated using data collected during the first 3 years of an ongoing case-control study. This large, multicenter, population-based study was designed to minimize some of -3-

ETS AND IAQREFERENCE JULY 1992 I the methodological problems which have been of concern in previous studies of envh'onmental tobacco smoke and lung cancer. Both a cancer control group and a population control group were selected in order to evaluate recall bias. A uniform his~opathological review of diagnostic material was conducted for case confirmation and detailed classification. Biochemical determination of current exposure to tobacco and screening of multiple sources of information to deter- mine lifetime nonuse were utilized to minimize misdassification of smokers as nonsmokers." [Although data on possible confounding factors, e.g., diet, were apparently collected, they were not considered in this paper.] "For all histopathological types of lung cancer combined, a 30% increase in risk is observed (OR = 1.28 and 1.29 with colon cancer and population controls) .... The estimated relative risk of pulmonary adenocarcinoma assodated with cigarette smoking by spouses was 1.36 (1.02-1.84) with the population controls as comparison and 1.31 (0.94-1.84) with the colon cance~ controls as comparison. No associa- tion between spouses' tobacco use and lung cancers other than adenocarcinoma was observed." "Exposures to cigarette smoking from spouse(s), other household members, on the job and in other activities of adult life ('social') are each associated with an overall 40- 60% significant elevation in the risk ofadenocarcinoma of the lung." "No association was found between risk of any type of lung cancer and childhood exposure to cigars, pipes, cigarettes, or all types of tobacco combined .... No significant elevations in risk were found at any level of smoking by household members during childhood." "Although fl'tis report represents the findings of the first 3 years of a 5-year szudy, it is neverthdess the Largest case-control study reported to date on this topic The findings provide additional evidence in favor ofa ~,,~ relationship between exposure to ~ and lung cancer in women who have never used tobacco themselves. A dose response, not likdy due to dmnce, was apparent for exposure to mbac-m smoke during adult life fi'om a variety of exposure sources. The assodafion was specific for both adenocardnoma of the lung and for all lung cancers combined compared to colon cancer." [6] "Passive Smoking and Cancer Risk: The Nature and Uses of Epidemiological Evidence," A. Woodward and A.J. McMichael, European Journal of Cancer 27(l l): 1472-1479, 1991 [Vol. I, Issue 12, December 10, 1991] The authors summarize the epidemiologic studies on spousal smoking and lung and other cancers, describing the purported assodation with lung cancer as "most probably causal." The authors also discuss the Sean Carroll and AFCO cases with regard to establishing "proof" of lung cancer risk in courts of law. Alistalr Woodward recently published another article, this one focusing on ETS in the workplace, which is summarized in Volume I, Issue 11 (November 22, 1991) of this newsletter. Anthony McMichael chaired the working group which prepared the Australian National Health and Medical Research Coundl report on ETS in 1986. He was also a witness for the applicant (AFCO) in AFCO v. Tobacco Institute of Australia. "The cancer risk of passive smoking is not a subject of mere sdentific curiosity. The risk of cancer in non- smokers is often the main reason given for prohibiting or restricting smoking in public places. From an economic point of view, much hinges on such prohibitions or restrictions. An obvious example is the reduction in cigarette consumption that is likely to follow workplace smoking bans: Chapman et al. estimate that if half the white collar worksites in Australia were to ban smoking the Australian tobacco industry would lose sales of $6.5 million annually. Because the stakes are so high, the issue of passive smoking and cancer also casts a spotlight on the scientific and legal assessment of risk, revealing different views on how cause and effect should be judged, and what constitutes 'sdentific proof.'" "The apparent effect of passive smoking on cancer risk has become an irnporrant social and political issue. For this reason alone the strength of the epidemiological evidence warrants dose examination. The research published to date indicates a positive assodations of passive smoking with lung cancer, but there is no consistent evidence ofassodations with cancer at other sites. We have sumrr~lrized the epide- miological evidence, and examined the major critidsms raised against these studies. These critidsms include alleged bias arising from misdassification of exposure to environ- mental tobacco smoke (ETS) or of personal smoking history, and from differential publication of positive findings. In their strongest form, these critiques challenge the ability of epidemiology to establish causation on any issue. We argue that epidemiology is not inherently different from other branches of sdence --in each of which sdentific 'proof' of cause and effect involves judgment based on measurement and logical inference. We also describe the application of epidemiological data to esr.ablishing proof, in courts of law, of the lung cancer risk of passive smoking." "The significance of this case [A/CO] was that it produced the first legal judgment anywhere in the world in relation to I I I I i i I i I i i I I °i i

l i I the tor.al,iry of the scientific evidence pertairfing to the effects - of ETS upon human health." "A review of the [AFCO] judgment indicates that Mr. Justice Morling accepted the primary argument that the interprerabilky of the epiderniological evidence depended in particular upon the pattern of findings acxoss studies, and on its coherence with the substantial evidence from studies of active smoking and cancer and from corroborative animal experimental and other bioassay studies -- and did not depend on the compilation of a scorecard of statistical signhqcance tests in a series of individual studies." [7] "Is Passive Smoking in the Workplace Hazardous to Health?" A. Woodward, Scandinavian Journal of Work and Environmental Health 17." 293-301, 1991 [Vol. I, Issue 11, November 22, 1991] This article concludes that, although little information directly addressing workplace ETS exposure and disease is available, the "evidence" from other sources is suffident to conclude that smoking should be reduced at work. The author reviews epidemiological studies and existing exposure data both for the home (from which he extrapo- lates to the workplace) and the workplace itself. "First, passive smoke contains many potentially toxic substances. These include the obvious tars and respiratory irritants. There is also a range of carcinogens and substances shown to be genotoxic. For example, cigarettes are the major cause of public exposure to benzene. Passive smoking is estimated to be responsible for 5 % of the national exposure to benzene in the United States. (This level is orders of magnitude greater than the hazard that led recently to the recall of Perrier water contaminated with traces of benzene.)" "Second, there is no known safe level of active smoking. Ira lot of smoke causes a big increase in the risk of disease, it makes sense to most people that a little smoke, as inhaled by the passive smoke, v,411 be responsible for a small but definite increase in risk." "A similar point is that there is evidence, from the epidemiologic studies, that this exposure is associated with cancer in humans. Many occupational exposures have been regulated in the past solely on the basis of animal evidence. The questions of whether tiffs association is causal, and what the strength of the association is, are less important, in many minds, than the observation that exposure is associated with cancer in human populations." "Third, the effects of passive smoking upon the nose, throat, and eyes make the exposure obvious. Furthermore, tolerance appears to be related to frequency of exposure. As passive smoking becomes less common, nonsmokers are less willing to put up with the smell and irritation of tobacco smoke." "Finally, there axe economic benefits to regulation, and few costs. The benefits not only include fewer fires and lower cleaning costs, but also lower smoking rates in the work force, and hence, it may be predicted, less absentee- ism. The costs of regulation of smoking at work are more difficult to identify." "The recent scientific evidence on passive smoking is consistent with the conclusions of the major reviews of 1986, which were that breathing other people's tobacco smoke is a cause of serious disease, including lung cancer. There is relatively little information available on whether breathing other people's tobacco smoke at work causes disease. However, it appears reasonable to extrapolate from what is known about health effects of passive smoking in other settings (predominantly the home) to the likely health effects in the workplace. It is difficult to quantify the risks involved." [8] "Smoking and Health: A Review Prepared By the Smoking and Health Subcommittee of the To- bacco Industries Council," a Council Formed By the Minister of Finance of Japan, H. Kasuga, H. Katsuld, O. Miyagl, W. Mori, S. Takayama and T. Yanagita, The International Journal of the Addic- tions 26(4): 423-440, 1991 [Vol. I, Issue 11, November 22, 1991] In this review of smoking and health issues, the authors report that evidence is insufficient to claim that "passive smoking" and disease are assodated. They state that irrita- tion "cannot be ignored," but find exposure information insuffident to establish "relations" with other diseases. "Any definition of health is inevitably broad and contains various elements that may differ from one individual to another. Recent studies on the effects of smoking on physical and mental health have progressed remarkably and have great value in the fields of epidemi- ology, pathology, clinical medicine, and psychiatry. This report concludes that while smoking may have benefidal psychological effects on smokers, it may pose a risk to physical health." "Recent research findings on the health effects of exposure to environmental tobacco smoke, what is known as passive smokin~ have aroused concerns about public health. However, the concentrations of tobacco smoke inhaled passively are slight in comparison with mainstream smoke -5-

ETS AND IAQ.REFEtLENCE JULY 1992 I • that is inhaled directly by smokers. Thus, if any correlation between passive smoking and lung cancer exists, it is thought to be ex'a'emdy slight, but at present there is not enough evidence of suffident probability to support the proposition. Similarly, in the case of general prevalence rates for respira- tory diseases, the evidence is not strong enough to prove that a dear distinction can be made between diseases resulting from passive smoking and those mused by other environ- mental factors." "Epidemiology is simply a tool to duddate suspicious factors in diseases affecting the general public. Thus, just because a corrdation exists, there is no reason to conclude that a risk factor constitutes a direct cause. Fsr~lishing causality requires the support of basic medical and clinical "As for the health effects of passive smoking, a certain unpleasanmess to the senses such as irritation to the eyes, nose, and throat cannot be ignored. However, at present the rdations between passive smoking and diseases such as lung cancer have not been confirmed because accurate evaluation methods of exposure levds have not been established." [9] "An Epidemiological Study of Lung Cancer in Xtmn Wei County, China: Current Progress. Case-control Study on Lung Cancer and Cooking Fuel," X. He, W. Chen, Z. IAu and R.S. Chaprrum, Environmental Health Perspectives 94: 9-13, 1991 [Vol. I, Issue 9, October 17, 1991] The information in this paper is nearly identical to that presented in the Idu, et al., paper, which appeared earlier this year (See Vol. I, Issue 2 (May 17, 1991) of this Report). The authors report that indoor polycyclic aromatic hydrocarbon pollution, produced by the burning of coal in unvented pits, plays an important role in the etiology of lung cancer. No association of lung cancer with ETS exposure was reported. [The Liu, et al., 1991, study, which reports nearly identical information, was surnmarized in Vol. I, Issue 2 (May 17, 1991) of this Report.] "Data gathered to date in Xuan Wei discloses [sic] a stronger assodation of lung cancer with domestic coal use than any other risk factors so far assessed, induding tobacco smoking .... Smoking may be contributing to lung cancer in males in Xuan Wd [more than 40% of whom smoke], but it is difficult to explain the marked differences of lung cancer mortality among Xuan Wei communes and the high lung cancer mortality in femal.es." A case-control study of 110 individuals (only 17% cytologically confirmed) and 426 controls matched on age, sex and village of residence was conducted. "No relation- ship between lung cancer and ever having smoked tobacco is observed .... In contrast, a statistically signifi- cant dose-response trend of lung cancer with smoking index is observed." "No association of lung cancer with passive smoking is observed [OR = 0.74, 99% C.I. 0.32-1.68; females only, 45 cases and 176 controls]." "[A] statistically significant relationship of lung cancer with age at which the woman began to cook food is observed, though none of the individual odds ratios is significant." "In both sexes, lung cancer is significandy assodated with family history of lung cancer and personal history of chronic bronchitis." "Conclusion. PAH (BaP as index) pollution in indoor air induced by smoky coal burning in unventilated, shallow pits plays an important role in lung cancer. Smoking is not a main risk factor of lung cancer in Xuan Wei County, China." [10] "Mainstream and Environmental Tobacco Smoke," G.B. Gori and N. Mantel, Regulatory Toxicology and Pharmacology 14: 88-105, 1991 [Vol. I, Issue 6, August 27, 1991] This review describes the inconsisten'des between the dose to nonsmokers from ETS and the purported increased risk of lung cancer based on epidemiology. Describing "irreparable deficiencies" in the epidemiologic studies, the authors focus on the numerous factors for which lung cancer relative ris "ks have been reported but which were not considered in the ETS epiderniologic Sl~udles. a~,XCERP TS "- "Some epidemiologic studies of nonsmokers presumably exposed to ETS have suggested a marginal increase of risk for some diseases previously assodated with active mainstream smoking (MSS). These reported risks, however, border on statisticaJ and epidemiologic insignifi- cance, and could easily derive from numerous and documented biases and confounders." "[T]he lung cell doses for average ETS-exposed non- smokers are probably between 1/10,000 and 1/100,000 of equivalent cell doses for average mainstream active smokers. In practical terms, this implies an annual retained dose of tobacco smoke components equivalent to far less than the dose for the active smoking of one cigarette somehow evenly disbursed over a 1-year period." I i I I I i ir I ! I I

VOLUME I, ISSUES 1-24 JULY 1992 I "IT]he active smoking of 4-5 dgarettes/day is not likely to be statistically assodated with elevated lung cancer risk. . .. It suggests that since ETS retained doses are several thousand dmes less than MSS doses from this level of cigarette smoking, they appear insuffident to generate elevated risks of lung cancer." "ETS is a very elusive entity, undergoing continuous transformations at extremes of dilution that make efforts to define its chemical, physical, and biologic characteris- tics highly difficult. While the components of MSS/SSS may also be present in ETS, it is also dear that with few exceptions they are undetectable by the most sophisti- cated analytical procedures." "[I] f epidemiologic investigations of MSS and lung cancer had been confined to the effects of exposure to a few dgarettes daily, they would have failed to yield significant risk signals .... At the same time it is apparent that subjects included in ETS epiderniologic studies were probably exposed to equivalent MSS-RSP doses below even a single cigarette peryear. Therefore, marginal RR values associated with ETS exposures should be imputed to biases, confounders, and other weaknesses of the investigations, and any judgment that ETS exposure leads to lung cancer and other diseases would flow from argument, not from credible data." "In fact, the majority of epidemiologic studies of ETS suffer from what appear to be irreparable defidendes." "An even greater prejudice to the credibility of ETS epidemiologic studies derives from their failure to account and control for the possible confounding by many independent risk factors." "Since many of the RRs... are substantially larger than any reported for the association of lung cancer and ETS, even weak contributions by combinations of these confounders would be cumulative and could be more than suffident to explain the marginal lung cancer risks that some epidemiologic studies Of ETS have reported... • [W]hese studies have not controlled for the factors.., in any meaningful or comprehensive way, while other investigations provide evidence that several of those risk factors cluster and selectively segregate in families with smokers." "It should be clear that the seemingly insurmountable difficulties in measuring ETS exposures and doses, unresolved classification bias, and the inability to control for numerous independent confounders explain the inconsistency of weak ETS epiderniologic results and speak against scientifically credible conclusions about a risk that, if real at all, remains imponderable." "Indeed, the only justhqable conclusion is that this issue cannot be resolved sdentifically on the basis ofcurrendy available information. Moreover, exposure and dose considerations alone seem to indicate that ETS is an insignificant entity among the substantial mass of exogenous and endogenous challenges to health that we continually face." [11] "Weaknesses in Recent Risk Assessments of Environmental Tobacco Smoke," P.N. Lee, Environmental Technology 12(3): 193-208, 1991 [Vol. I, Issue 6, August 27, 1991] Several risk assessments based on epidemiologic data have given higher risk estimates than those based on dosimetry. The author considers six potential sources of bias in the epiderniologic studies of ETS and lung cancer and offers additional criticisms of risk estimates for heart disease and for cancer other than the lung. "IT]here has been an increasing tendency to carry out risk assessments to estimate annual numbers of deaths due to ETS. The purpose of this paper is to underline a number of problems in conducting such risk assessments, and to comment critically on three that have recendy been published [Wells, Kawachi, et al.; and Repace and Lowrey]." "IT]here is certainly strong evidence of a marked discrepancy between the epidemiology and dosimetry .... [T]he discrepancy seems very large, by two or even three orders of magnitude." "One implication is that risk assessments based on dosimetric evidence are likely to give substantially lower estimates than those based on the epidemiologic evidence. Another implication is that it gives reason to doubt the epidemiology, and to look for sources of bias." "Epidemiology is imprecise. Various sources of bias can produce spurious relative risks of 2 or even more. Since the relative risks seen for ETS exposure are well within this range, and since they seem inconsistent with the dosimetric evidence, it is important to examine the epidemiological evidence critically. Six potential sources of bias are considered below." The author discusses in some detail, misclassification of diagnosis, misdassification of ETS exposure, publication bias, poor design of some studies, confounding, and misdassification of active smokers as nonsmokers. He then concludes: "The epidemlology has indicated a magnitude of risk in relation to spouse smoking that is implausibly large compared with what is known about the -7-

AND IAQREFERENCE JULY 1992 I • extent of ETS exposure involved. There ire clear weak- nesses and sources of bias in the epidemiology which could invalidate risk assessments based on it." "All three risk assessments criticized in this document take the epidemiology virtually at face value, with no real discussion at all of its weaknesses .... No reasonable scientific criteria are used to decide what constitutes a valid study before it can be included in a risk assessment --studies conducted with complete disregard of basic scientific principles ire included as if they were as valid as carefully designed studies." Regarding heart disease, the author claims to demon- strate "that the risk assessment for heart disease essentially rests on the results from two studies [Helsing, et al., and Hirayama], both of which seem unreliable." He makes the following points, "First, there are a very large number of risk factors for heart disease .... Second, the extent of the assodation seen in some of these studies, which in some cases is close to that reported in relation to active smoking, is implausibly high when viewed against the extent of the assodation seen in relation to active smok- ing. Third, there is a major danger of publication bias." "The overall evidence for cancer other than the lung is clearly remarkably unconvindng in demonstrating any effect of ETS exposure." "[Xvc']hile estimates based on retained particulate matter give tens of deaths and those based on nicotine or respirable suspended particulates give hundreds, the epidemiologically based estimates all give thousands of deaths, x,grl~ich answer, if any, one accepts depends to a large extent on the faith one places on the different types of evidence. Wells, Kawachi et al and Repace and Lowrey accept the epidemiology essentially at face value and pay little or no attention to its poor quality and very obvious weaknesses. They either ignore the dosimetric evidence, do not make it clear that it gives different answers and/or dismiss it as inconsistent with the epiderMology, or invoke mechanisms to explain the discrepancy which ire scientifically unappealing. It seems to this author that the epidemiological evidence is untrustworthy and that, between the two, the dosimetric evidence is preferable." "When one restricts attention to lung cancer, to never smokers and to ETS exposure from the spouse, one is at least operating in an area where the epidemiological evidence indicates an association. When one extends risk assessment to other diseases, to ex-smokers and to ETS exposure in the workplace one is stretching the limits of what is sdence .... There is some evidence on ETS exposure in the workplace, but this shows no assodation at all with lung cancer risk." -8- [12] "'Passive Smoking' and Lung Cancer:. A Criticltl Ana/ysis," G. Fetter and D.J. Ecobichon, Modern Medicine of Canada46(4): 26-29, 1991 [Vol. I, Issue 6, August 27, 1991] The authors reviewed the available information on purported health effects of ETS and determined that it is inconclusive. They note that any conclusions as to whether ETS increases the risk of disease will have to wzit until reliable and appropriate data ire available on rnisdassification error, marital concordance and the establishment of spousal smoking as a valid surrogate index of ETS exposure. EXCERP'P$: "Cireful review of the pertinent information relating to the health effects of ETS reveals that the data remain inconclusive with respect to any causal relationship between ETS exposure and lung cancer. The finding of the NRC and other investigators of a statistically signifi- cant composite OR [odds ratio] that links spousal smoking with a modest (20-50%) in,eased risk of lung cancer in the nonsmoker is provocative." "However, a number of reviewers have questioned some of the assumptions on which that finding was based, as well as the appropriateness of using meta-analysis for this particular data base. Whether the increased OR reflects an effect of ETS or statistical artifact will have to await the availability of reliable and appropriate data on misdassificarion error, risks of lung cancer assodated with active smoking, and marital concordance. It also remains to be established that spousal smoking is a valid surrogate index of ETS exposure." [13] "Environmental Tobacco Smoke and Lung Cancer in Never Smokers," H.G. Stockwell, E.C. Candelora, A.W. Armstrong and Pak. Pinkham, Meeting Abstract, Society for Epidemiologic Research Conference, June 11-14, I991 [-Vol. I, Issue 5, August 9, 1991] This ongoing, population-based, case-control study investigated risk factors for lung cancer among never smoking women in Florida. When the abstract was submitted, 124 cases and 241 controls were included; data were reported at the conference based on 148 cases and 265 controls. (A complete copy of the Abstract is reprinted in Appendix A.) Risk factors for lung cancer among women who never smoked cigarettes were examined in an ongoing, popula- tion based case-control study conducted in Florida. One i I i I i I i I i

VOLUME I, ISSUES 1-24 JULY 1992 hundred and twenty four primary carcinomas of the lung, and 241 control women who had never smoked were included. Results suggest that childhood and adult exposure to environmental tobacco smoke may increase the risk of lung cancer among women who never smoked cigarettes. Having a husband who smoked dgarettes resulted in a statistically significant increase in risk of lung cancer among women who never smoked, with an odds ratio of 1.8 (95% C.I. 1.1-2.9). A 40% increase in risk was observed among women with less than 25 years exposure to a spouse who smoked, when compared to women who repor'~ed their spouse had never smoked, with the risk increasing to 60% among women exposed 25 years or longer. When exposure to tobacco smoke in childhood was considered, the data were less consistent. Having a parent who had smoked during the respondent's childhood did not increase the risk of lung cancer. However, among those respondents with high levels of exposure to parental smoking, an excess risk, although not statistically signifi- cant, was observed. Never smoking women who accumu- lated 25 or more exposure years experienced a 70% increase in risk (OR=l.7, 95% C.I. 0.9-3.6) of lung cancer compared to women who reported neither parent had smoked dgaretres. [14] "Lung Cancer Risk Associated With Cancer in Relatives," G.L. Shaw, R.T. Falk, L.W. Pickle, T.J. Mason and P.A. Buffler, Journal of Clinical Epidemiology 44(4-5): 429-437, 1991 [Vol. I, Issue 4, July 12, 1991] The authors examined data from a population-based case-control study of lung cancer in the Gulf Coast of Texas, and reported that exposure to ETS showed a minimal excess in lung cancer risk after adjustment for smoking. They reported an adjusted OR of 1.2 in females and 1.1 in males. Data from population-based case-control study of lung cancer in the Gulf Coast of Texas. For derails of the study the reader is referred to a paper by Buffler, et al., which was one of those included in the meta-analysis of ETS and lung cancer in the 1986 NRC report. "Exposure to passive smoke showed a minimal excess in lung cancer risk after adjustment for smoking." [Adjusted OR = 1.2 in females (435 exposed cases) and 1.1 in m~es (363 exposed cases). No 95% CI given, but neither OR was statistically significant.] "Consistent with previous studies, we found a modest increase in lung cancer risk assodated with reported cancer in first-degree relatives. Risks were similar for both men and women. Risk was greatest for those subjects with first-degree relatives with lung cancer, followed by first- degree relatives with other TRCs [tobacco-related cancers]." "Differential cancer risk by histologic type, and between first-degree relatives and spouses is consistent with the hypothesis that genetic characteristics may influence risk by increasing host susceptibility to an environmental carcinogen." "The absence of smoking data on farrfily members limits the interpretation of our results, since this may contribute to a number of tobacco-related cancers among relatives." [15] "Dietary Patterns of Female Nonsmokers With ~nd Without Exposure to Environmental Tobacco Smoke," L. Le Marchand, L.R. Wilkens, J.H. Hankin and N.J. Ha/ey, Cancer Causes and Control2: 11-16, 1991 [Vol. I, Issue 4, July 12, 1991] In this study, 82 female nonsmokers were interviewed to assess the possible relationship between dietary confound- ers of lung cancer risk and ETS exposure. The authors concluded that diera.ry beta-carotene and, possibly, dietary cholesterol a_re potential confounders of the ETS/ lung cancer assodation. Therefore, according to the authors, these dietary factors should be carefully measured in future studies. Study subjects were selected from population controls who took part in two lung cancer case-control studies conducted in Hawaii between 1979 and 1985. Eighty- two female nonsmokers were interviewed in 1986 to assess the possible relationship between dietary confound- ers of lung cancer risk and ETS exposure. ETS exposure was assessed by measurement of urinary cotinine and also by marriage to a smoker. "Intakes of vitamin A... decreased as exposure to ETS [urinary cotinine] increased. A similar inverse relationship was observed for each of the two food sources of vitamin A, namely, retinol and carotenoids, including beta- carotene.', "Cholesterol and fat intake were also negatively associ- ated with lETS/urinary cotinine] exposure level; however, the test for trend was statistically significant for choles- terol intake only (P = 0.05)." "Women married to a smoker had lower intakes of total vitamin A and its components, as well as lower intakes of -9-

ETS AND IAQ PdSFERASNCE JULY 1992 cholesterol and fat, than women married to a 'never smoker'.... [W]omen married to an ex-smoker had a dietary pattern which is intermediate between those of the other two groups of women." "We conclude that dietary beta-carotene and, possibly, dietary cholesterol are potential confounders of the passive-smoking/lung cancer assodation. Although the distorting effects of these dietary factors appear modest, they should be carefully measured in future studies since they may interfere with the confirmation of this impor- tant assodation." [16] "Serum Beta-Carotene in Persons With Cancer and Their Immediate Families," A.H. Smith and K.D. Waller, American Journal of Epidemiology 133(7): 661-671, 1991 [Vol. I, Issue 4, July 12, 1991] This study was conducted in New Zealand to test the hypothesis that serum beta-carotene would be lower in the families of patients with cancer than in the families of control patients. The authors reported that low levels of serum beta-carotene were observed in both the cancer patients and their families for cancers of the lung, stomach, esophagus, small intestine, cervix, and uterus. The strongest findings for an individual cancer site were those for lung cancer. EXCERPTS: No mention of ETS. Study conducted in New Zealand to test the hypothesis that serum beta-carotene would be lower in the families of patients with cancer than in the families of control patients. This was accomplished by incorporating family members into a standard case-control study design. "[L]ow levels of serum beta-carotene were observed in both the cancer patients and their families for cancers of the lung, stomach, esophagus, small intestine, cervix, and uterus. The strongest findings for an individual cancer site were those for lung cancer .... Our findings for lung cancer applied to all cell types other than adenocarcinoma." "We found no evidence that confounding by current smoking status could account for the findings." "[M]ost of the cancer sites that were associated with low levels of serum beta-carotene are sites for which smoking is a strong risk factor (with the exception of cancers of the bladder and kidney)." -10- [17] "Diet and Lung Cancer in California Seventh-Day Adventists," G.E. Fraser, W.L. Beeson and R.L. Phillips, American Journal of Epidemiology 133(7): 683-693, 1991 [Vol. I, Issue 4, July 12, 1991] This study looked at data based on six years of follow-up (1977-1982) in the prospective study of Seventh-Day Adventists in California. The authors reported that the dietary item that showed a substantial assodation was all types of fruit. The authors indicated that the protective association with fruit appears to be more marked for adenocarcinomas. EXCERIVl~: No mention of ETS. Data based on 6 years of foLlow-up (1977-1982) in the prospective study of Seventh-Day Adventists in Califor- ilia. "The dietary item that did show substantial assodation upon stratified analysis was fruit (all types). Here a relatively powerful apparent protective effect was seen that is unlikely to have been due to chance." "The protective assodation with fruit appears to be more marked for adenocarcinomas (Kreyberg group II), but it could not be clearly differentiated from a relatively similar effect in the Kreyberg group I tumors [squamous cell, large cell, and small cell]. This result agrees closely with the observations of Koo among Chinese females in Hong Kong." "Foods included in our data set that have substantial beta-carotene content showed no consistent associations." [18] "Smoking and Other Risk Factors for Lung Cancer in Xuanwei, China," Z. Liu, X. He and R.S. Chapman, International Journal of Epidemiology 20(1): 26-31, 1991 [Vol. I, Issue 2, May 17, 1991] Lung cancer rates in Xuanwei County are among the highest in China. Previous studies (not epidemiologic) have suggested that there may be an assodation between the indoor burning of "smoky" coal, as opposed to "smokeless" coal or wood, and lung cancer incidence. This case-control study included 110 cases (56 males, 54 females) and 426 controls matched for age, sex, occupa- tion (all were farmers), and village of residence (which controlled for type of fuel used). Only one of the female cases reported having ever smoked. Among men, the authors reported a statistically significant dose-response relationship with active smok- ing; however, of all the indices used to examine active smoking, a statistically significant odds ratio was reported ! ! i 11 I ! ! ! ! ! I ! i i i i

I VOLUME I, ISSUES 1-24 JULY 1992 ! i ! ! i ! ! I ! ! I I I I I I ! I for only one category. The authors claimed that their data supported the existence of a dose-response relationship in males between lung cancer and a smoking index (calcu- lated as years of smoking x amount of smoking). ETS exposure was assessed in women as whether there was at least one smoker (usuaLly the husband) living in the same household. For 45 cases and 176 controls, an adjusted OR of 0.77 (95% CI 0.30-1.96) was reported. (This is the only index of ETS exposure used; no information was presented on spousal smoking, workplace exposure or the like.) Statistically significant increases in risk were reported for a number of other factors. In females, these were chronic bronchitis, OR = 7.37 (95% CI 2.40-22.66) and family history of lung cancer, OR = 4.18 (95% CI 1.61-10.85). In males, the reported assodations were with chronic bronchi- tis, OR = 7.32 (95% CI 2.66-20.18); family history of lung cancer, OR = 3.79 (95% CI 1.70-8.42); and personal history of cooking food, OR = 3.36 (95% CI 1.27-8.88). (Note that some of the CIs are quite wide, indicating low statistical power.) Assodations with lung cancer were also suggested for duration of cooking food and for age at starting m cook (over coal-fired stoves) for women. [ 19] "Comparative Epidemiology of Cancer Between the United States and Japan," E.L. Wynder, Y. Fujita, R.E. Harris, T. Hirayama and T. Hiyama, Cancer 67: 746-763, 1991 [Vol. I, Issue 2, May 17, 1991] [No mention of ETS.] "It was postulated [in 1977] that as smo "king becomes more common among both men and women in Japan, tobacco-related cancer rates would also increase. As pre- dicted, rapid accelerations in mortality rates due to cancers of the lung and ompharynx have recendy been observed in the heavy smoking male population of Japan." "We also predicted that changes in Japanese patterns of dietary fat intake, fruit/vegetable consumption, and food preservation would result in modified rates of cancers associated with these factors. Accordingly, recent upward trends in Japanese mortality rates of colon and pancreatic cancer are evident.., whereas the rates of stomach cancer continue to decline.., the mortality rates for malignan- cies of the breast, ovary, corpus uteri, and prostate appear m be steadily rising in Japan, which may also be due to continued 'westernization' of the Japanese diet, although other life-style changes affecting the endocrine system cannot be ruled out as potentiating factors." "Still there are some discrepandes in the international trends and differences in cancer mortality that conflict with what is currendy known about the risk factors of specific malignandes. Lar~geal cancer rates in Japanese males are lower than in the US, and surprisingly are declin- ing, despite thdr reladvdy high levels of dgaretr.e smoking. Cancer of the urinary bladder, also a tobacco-rehted malignancy, does not show increased mortality in recent years among the heaw smoking male population of Japan. Esophageal cancer rates in Japan condnue m exceed those in the US, despite comparable levels of alcohol consumption in the two countries. These results in particular have not been satisfactorily explained and must be examined for valuable dues to the risk factors bearing on cancer etiology." [20] "Passive Smoking and Diet in the Etiology of Lung Cancer Among Non-Smokers," A. Kalandidi, tC Katsouy-anni, N. Vompoulou, G. Bastas, R. Saracci and D. Trichopoulos, Cancer Causes and Control 1: 15-21, 1990 [Vol. I, Issue 1, April 30, 1991] This case control study was undertaken in Athens to explore the role of ETS exposure and diet in the causatior; of lung cancer, by histologic type, in nonsmoking women. It reports that marriage ofa nonsmoking woman to a smoker was assodated with a relative risk for lung cancer of 2.1 (95% CI 1.1-4.1). It also reports that the number of dgarettes smoked daily by the husband and years of expo- sure to the husband's smoking were Positivdy, but not significantly, rdated to lung cancer risk. The authors indicated that there was no evidence of any assodation with exposure to smoking of other household members, and the assodation with exposure to ETS at work was small and not statistically significanr~ The authors report that ETS was assodated with an increase of the risk of all histologic types of cancer, although the devation was more modest for adenocardnoma. Dietary data collecraxt through a semi-quantitative food- frequency questionnaire indicated that high consumption of fruits was inversely rdated to the risk of lung cancer (the rdadve risk between extreme quanrdles was 0.27 (CA 0.10- 0.74)). Neither vegetables nor any other food group had an additional protective effect. The reported associations of lung cancer risk with ETS exposure and reduced fruit intake were independent and did not confound each other. Cam)[ovascut o, Issu [1] Letters to the Editor Regarding "An Estimate of Adult Mortality in the United States from Passive Smoking," A.J. Wells, Environment International 14: 249-265, 1988 [Issue 24, Item 36] Environment Inteenationa/recendy published two letters concerning this article, which was published prior to issue -11-

ETS AND IAQ_REFEKENCE JULY 1992 I 1 of this Report. The authors of the letters were (i) Peter N. Lee and (ii) A. Judson Wells, author of the original article. The letters appear at Environment International 18:315-317 and 321-325, 1992. In his original article, Welts calculated that "passive smoking" was responsible for more than 50,000 deaths armually in the U.S. The majority of the daimed deaths (32,000) were said to be due to cardiovascular disease, with the rest attributed to lung cancer and cancer of other sites. Lee's letter critidzed Wells' dalm that chemical and physical differences between ETS and mainstream smoke are related to purported health effects. Lee also com- mented on the "failure" of the American Cancer Society to publish data (which he believes have been collected) on ETS exposure and heart disease; Lee proposed that such data would be an important addition to the sdentific literature on this issue. Lee also called to Wells' attention epidemiologic literature on workplace or childhood exposure m ETS and lung cancer incidence which reported statistically non-significant relative risks. Finally, Lee proposed that sources of bias, namely confounding by dietary differences (e.g., beta-carotene intake), and misdassification of smokers as nonsmokers, could explain the reported association between spousal smoking and lung cancer. In his response to Lee's letter, Wells claimed that Lee's approach to the misclassification question was erroneous. Wells cited the Glantz and Parmley (1991) and Steenland (1992) papers as support for his conclusions. Wells further claimed that epidemiologic studies of ETS exposure and lung cancer have not demonstrated a confounding effect of diet. He also invoked a "latitude effect" in response to Lee's comments that workplace studies supported no assodation, claiming that "lung cancer signals were dearer from the southern studies than from the northern ones." Wells concluded that his estimate of a "passive smoking death toll in the 50,000 range" is still the best estimate, discounting Lee's claim that biasing factors were operating in the ETS studies. [2] ~Sertmn Lipopmteins in Nonsmokers Chronically Exposed to Tobacco Smoke in the Workplace," J.R. White, M. Criqui, Jak. Kulik, H.F. Froeb and P.J. Sinsheimer, Abstract No. 383, Eighth World Conference on Tobacco or Health: Building a Tobacco-Free World, March 30-April 3, 1992, Buenos Aires, Argentina [Vol. I, Issue 20, April 24, 1992] This abstract reported that male and female workers classified as exposed to ETS had statistically significant decreases in high density lipoprotein (HDL) cholesterol (i.e., "good" cholesterol) and elevated ratios of total cholesterol to HDL. In addition, females reportedly had significant elevations in low density lipoprotein (LDL) cholesterol ("bad" cholesterol). United Press International publicized this abstract in a newswire dated April 2, 1992. At the Eighth World Conference on Tobacco or Health in Buenos Aires, Argentina (March 30-April 3, 1992), a study was reported by James tL White, et al. (University of California-San Diego) which claimed that workplace ETS exposure adversely affected cholesterol profiles. This presentation also obtained publidty through a United Press International news'wire dated April 2, 1992. The conference abstract of the White, et al. study is the fourth publication, all since 1990, to address the general issue of a possible assodation of ETS exposure with changes in cholesterol levels. However, it is the first to report data on cholesterol levels and ETS exposure in an adult popula- tion and to specifically address workplace ETS exposure. White, et al. evaluated cholesterol and cholesterol fraction levels in nonsmoking workers. Carbon monoxide levels were used "as an index of dgaret~e smoke in the work place~" [Both the conference abstract and the UPI rdease refer, probably through publishing error, to measurement of carbon dioxide instead ofearbon monoxide.] Both male and female workers classified as exposed to ETS were reported to have statistically significant decreases in high density lipoprotein (HDL) cholesterol (i.e., the "good" cholesterol) and elevated ratios of total cholesterol to HDL. In addition, females exposed to ETS were reported to have significant elevations in low density lipoprotein (LDL) cholesterol (i.e., the "bad" cholesterol). The authors concluded: "Nonsmoking workers are at increased risk of developing coronary heart disease resulting from exposure to second-hand tobacco smoke." [3] "Passive Smoking and Your Heart," G.L. Huber, R.E. Brockie and V.K. Mahajan, Consumers' Research, Apri/1992, pp. 13-19, 32-33 [Vol. I, Issue 19, April 10, 1992] The authors review the biology of cardiovascular disease and discuss the concept of risk factors. They review the epidemiologic studies on ETS and heart disease, discussing in detail a number of factors (including lack of exposure assessment and failure to account for confounding variables) that could affect the reported results of these studies, and conclude that the studies should be "viewed with healthy sdentific skepfidsm." A copy of this article is attached as I i 1 i I I i I I I I I I -12- I

VOLUME I, ISSUES 1-24 JULY 1992 l ! ! ! ! l i i I II I ! I 1i I ! ! Appendix H. Another article by these authors, "Passive Smoking: How Great a Hazard?", which appeared in the July 1991 issue of the same magazine, was appended to Issue 4 of this Report, July 12, 1991. [4] "Passive Smoking and Carotid Artery Wall Thick- hess: The ARIC Study," G. Howard, M. Szido, G. Evans, G. Tell, J. Eckfeldt and G. Heiss, Cardiovas- cular Disease Epidemiology Abstracts, presented at Annual American Heart Association meeting, March 1992 [Vol. I, Issue 19,April 10, 1992] This abstract reported that increasing exposure to ETS was assodated with increases in carotid artery wall thickness. The authors concluded that exposure to ETS may contribute to atherogenesis, that is, narrowing of the armries that may increase the risk of clogging by cholesterol deposits and dots. In press cover-age of the abstract, researchers were described as cautioning that the study's findings were not yet conclu- sive. See Washington Posg March 20, 1992. Seelssue 13 of this Report, January 3, 1982, for a description of another, similar abstract by this group. "The assodation between passive and active cigarette smoking with carotid artery wall thickness was studied in 12,863 men and women ages 45 to 64 examined by the Atherosclerosis Risk in Communities (ARIC) Study. Of these, 3,509 were current smokers, 4,276 past smokers, 3,316 had never smoked but reported exposure to environmental tobacco smoke (ETS or 'passive smoke'), and 1,762 had never smoked and reported no exposure to ETS. Carotid artery wall thickness was measured by... ultrasound. Increasing exposure to dgarette smoke across the gradient from never smoking to current smoking was consistently associated with increases in carotid artery wall thickness within 5-year age groups." "The ETS group had thicker arterial walls than never smokers; these differences were statistically significant only at younger ages. Also, the ETS participants showed an increase in arterial wall thickness with an increasing number of hours per week of ETS exposure. Thus exposure to ETS may contribute to atherogenesis." [5] "Indoor Passive Smoking: Its Effects on Cardiac Performance," A. Leone, L. Moil, F. B.ertanelli, P. Fabiano and M. Filippelli, International Journal of Cardiology33(2): 247-252, 1991 [Vol. I, Issue 17, March 6, 1992] These researchers reported on cardiac performance during exercise testing in 19 male nonsmokers exposed to ETS in one test and tested once without the exposure. ETS exposure was reportedly associated with a decrease in peak exercise capacity in those subjects who had survived a previous myocardial infarction. In both groups of subjects, ETS exposure was reportedly associated with longer time to recovery of pre-exerdse heart rates. Cardiac performance during exercise testing was measured in 19 male nonsmokers, nine of whom were healthy and 10 of whom were myocardial infarction survivors. The subjects were tested twice, once while exposed to environmental tobacco smoke (leading to carbon monoxide concentrations of 30-35 ppm) and once without this exposure. The authors reported that ETS exposure was assodated with a decrease in peak exercise capacity in the myocardial infarction survivors, but not in the healthy subjects. For both groups of subjects, ETS exposure was associated with longer times to recovery of pre-exercise heart rates. ETS exposure during exerdse testing was also associated with increases in expired air and plasma carbon monoxide concentrations, although there were some irregularities in the data pertaining to these comparisons. The authors concluded, "Cardiac response to the exerdse is signifi- cantly worsened by passive smoke, espedally in those subjects with previous myocardial infarction." [6] "Passive Smoking and the Risk of Heart Disease," tC Steenland, JAMA 267(1): 94-99, 1992 [Vol. I, Issue 14, January 17, 1992] This article reviews epidemiological and experimental literature related to the claim that ETS causes heart disease. The author, Kyle Steenland, is affiliated with the National Institute for Occupational Safety and Health and "developed the epidemiology section" of NIOSH Current Intelligence Bulletin 54," Environmental Tobacco Smoke in the Workplace" (see Bulletin 54, Acknowledgement Section). He estimates that ETS exposure leads to 35,000-40,000 annual deaths in the U.S. and concludes that "heart disease mortality is contributing the bulk of the public health burden imposed by passive smoking." Steenland summarizes the data from nine epidemiologi- cal studies on ETS and heart disease. Although he does not perform a meta-analysis, he comments that the "[e]pidemiological evidence has been increasing that passive smoking at home is related to heart disease among never-smokers." Of the nine studies discussed, "Is]even are positive, while one is positive for women but not for -13-

ETg AND L&QR~FERENCE JULY 1992 I men," he says. Several weaknesses in the epidemiological studies are noted, particularly inadequate information on ETS exposure and the difficulties in controlling for multiple confounding variables. Steenland briefiy reviews several experimental and biochemical reports which he considers relevant to ETS and heart disease, insofar as they might suggest biological mechanisms underlying the claimed statistical relation- ship. He focuses especially on reports that low levels of carbon monoxide exposure lead to irregular heart rhythms in heart disease patients. He also rites recent research reporting that ETS exposure is associated with adverse effects on cholesterol and fibrinogen levels. He mentions, but does not discuss, specific studies dealing with a wider range of possible mechanisms, including platelet aggrega- tion, endothelial damage and the involvement of polycy- clic aromatic hydrocarbons. Much of the article's content is an effort to calculate an estimate of the annual number of U.S. heart disease deaths attributable to ETS. In general terms, this estimation process involves estimating an overall increase in heart disease risk assodated with ETS exposure; making adjustments for potential misdassification and for background exposure; estimating the extent of exposure to ETS; and estimating the fraction of never-smoker (and long-term ex-smoker) heart disease deaths attributable to ETS exposure. These estimates are incorporated into a formula using data on U.S. heart disease death rates and population estimates, from which is derived an esrdmated number ofarmual deaths attributed to ETS exposure. According to Steenland's calculations, "the overall estimate of ETS-at~ibutable heart disease deaths for never-smokers and former smokers is 35,000 to 40,000." He further comments that these increased risks of death "are higher than those ac~.epted in regulating environmental toxins. ~ [7] "The Association Between Carotid Arterial Wail Thickness and Active and Passive Cigarette Smok- ing," G. Howard, M. Szklo, G. Evans, G. Tell, J. Eckfeldt, G. Heiss and the ARIC Investigators, Arteriosclerosis and Thrombosis 11 (5): 1432a, 1991 [Vol. I, Issue 13, January 3, 1992] The authors report that ETS exposure is associated with carotid artery wall thickness, an indication of atherosclerosis. As this is only an abstract, only limited information is available. This abstract, from the Bowman Gray School of Medicine, Winston Salem, NC, reports that ETS exposure is assodated with thickness of the walls of the carotid arteries. The importance of carotid artery thickness is that it is an indica- tion of the severity of atherosderotic involvement. Athero- sclerosis of the carotid a.rveries is believed to underlie certain forms of stroke. These data stem from the Atherosderosis Risk in Commu- nities (ARIC) study. Details about the study are not given in the abstract. However, from other publications, it is known that the ARIC project is a longitudinal study of cardiovascu- lar and pulmonary disease sponsored by the National, Heart, Lung and Blood Institum The sample is drawn from communities in Minnesota, Maryland, North Carolina and Mississippi. The specific data in the abstract are based on a subsample of whites, who were categorized into four groups according to smoking history or ETS exposure: (1) current smokers, (2) exsmokers, (3) ETS-exposed never smokers, and (4) non-ETS-exposed never smokers. These groups were examined in terms of average carotid aa-tea-y thickness. Data were further divided into five-year age brackets, covering the range 45-65 years old. The authors reported that for each age group "there was a consistent gradient of wall thickness across the smoking exposure categories." Statistical tests indicated that carotid artery wails were significantly thicker in the ETS-exposed compared to the nonexposed group, but only for the younger age groups. Carotid artery wails were also signifi- cantiy tbdcker in the smokers compared to the exsmokers, but only for the older age groups. The authors concluded: "This graded relationship underscores the importance of smoking as a risk factor for atherosderosis." [8] "Urinary Cotinine Measurement in Patients With Buerger's Disease -- Effects of Active and Passive Smoking on the Disease Process," M. Matsushita, S. Shionoya and T. Matsumoto, Journal of Vascular Surgery 14(1): 53-58, 1991 [Vol. I, Issue 11, November 22, 1991] Buerger's disease, an inflammatory condition affecting the peripheral blood vessels, has been reported to be statistically assodated ~4th smoking. In this study, the authors reported no difference in the progression (i.e., worsening) of the disease between ETS-exposed and non- ETS-exposed nonsmokers. Buerger's disease is an inflammatory condition leading to arterial occlusion in the peripheral vascular system. It has been reported to be strongly associated statistically with cigarette smoking. Matsushita, et al. studied 40 Buerger's disease patients, all of whom had a smoking history. Using urinary cotinine levels as a marker, these l i I I ! I i I I I i I i i I I I I

VOLUME, l, l, ggUF-,S 1-24 JULY 1992 I - patients were classified either as smokers, as "passive smokers" (i.e., as nonsmokers exposed to ETS) or as nonsmokers not exposed to ETS. When the progression or "aggravation" of the disease was examined retrospectively, it was reported to have worsened in seven of 10 of the smokers, in none of the nine "passive smokers" and in four of the 21 non-ETS- exposed nonsmokers. Among this last group, three of the four admitted to "active" smoking and the fourth reported exposure to ETS in the workplace. Statistical tests revealed that the course of Buerger's disease had significantly worsened in the smokers, relative to the other two groups. However, there was no statisti- cally significant difference between the "passive smoking" and non-ETS-exposed group. Based on these data, the authors concluded that their results confirmed the relationship of"active" smoking with Buerger's disease, but that the "effects of passive smoking on the disease process were still inconclusive." [9] "Passive Smoking and the Risk of Heart Attack or Coronary Death," AJ. Dobson, H.M. Alexander, R.F. Heller and D.M. Lloyd, The Medical Journal of Australia 154: 793-797, 1991 [Vol. I, Issue 10, November 1, 1991] This AustraLian case-control study examined male~ and females who experienced a "farad or non-fatal definite or possible myocardial infarction or a corona-7 death." Odds ratios and 95% confidence intervals reported for heart disease risk assodated with ETS exposure at home were 0.97 (0.50-1.86) for men and 2.46 (1.47-4.13) for women. For ETS exposure at work, the odds ratios and confidence intervals were 0.95 (0.51-1.78) for men and 0.66 (0.17- 2.62) for women. This case-control study investigated the potential relationship between environmental tobacco smoke exposure and heart disease in residents of New South Wales, Australia. The cases were males or females who had experienced a "fatal or non-fatal definite or possible myocardial infarction or a coronary death." Controls were selected from partidpants in an ongoing WHO risk factor prevalence study. Data were collected on certain demo- graphic characteristics, medical history, dgarerte smoking and ETS exposure at home and work. Odds ratios and 95% confidence intervals reported for heart disease risk assodated with ETS exposure at home were 0.97 (0.50-1.86) for men and 2.46 (1.47-4.13) for women. For ETS exposure at work, the odds ratios and confidence intervals were 0.95 (0.51-1.78) for men and 0.66 (0.17- 2.62) for women. The authors concluded that their study "confirms previous findings of elevated risk of heart attack or coronary death associated with passive smoking at home." However, they observed that the "odds ratios for passive smoking at work did not suggest increased risk." Based on other aspects of their study, the authors claimed that the data confirmed increased heart disease risk in "active" smokers as well as increased ETS-related heart disease risk in exsmokers. Also, blood fibrinogen (a clotting factor) was measured in controls and correlated with levels of reported ETS exposure. The authors suggested that increased levels of fibrinogen were a marker of ETS-related heart disease risk. The authors commented on a variety of sources of bias in their study, including potential effects of confounding: "On balance, the effects of bias and confounding could have led to overestimation of risks due to passive and active smoking." [10] "Passive Smoking Alters Lipid Profiles in Adoles- cents," J. Feldman, I.R. Shenker, R.A. Etzel, F.W. Spierto, D.E. Lilienfleld, M. Nussbaum and M.S. Jacobson, Pediatrics88(2): 1-6, 1991 [Vol. I, Issue 6, August 27, 1991] This study, featured in the newspaper USA Today, compared cotinine levels (as an index of exposure) and blood lipid profiles in 391 New York adolescents. The authors reported that ETS exposure was associated with decreased high-density lipoprotein cholesterol (HDL-C) and with an increased ratio of total cholesterol to HDL- C. They propose this as a mechanism for the increased heart disease risk purportedly assodated with ETS exposure. [Note: This study was featured on the front page of the newspaper USA Today, on August 6, 1991, under the headline "Study: Parents' Smoke Hurts Teens."] "The TOTAL-C/HDL-C [total cholesterol/high-density lipoprotein cholesterol] ratio is a powerful predictor of the risk of atherosderotic cardiovascular disease and therefore its relationship to passive as well as active smoking has implications for pediatric atherosderosis prevention." "The present study investigated the relationship of passive smoking to lipid profiles in healthy adolescents. Cotinine, a major metabolite of nicotine, was used as a marker of passive exposure to tobacco smoke." Two hundred seven-four boys and 117 girls (mean age 14.8 years) from suburban New York high schools were -15-

ETS AND IAQREFERENCE JULY 1992 -included in the study. Elever~percent of the children had plasma cotinine concentrations _> 2.5 ng/mL, "the level considered indicative of exposure." Mean plasma cotinine concentrations were significantly higher among those who reported that one or both parents smoked. (Contrary to the U.S.A. Today report, the authors stated that sodoeconomic status and diet were not controlled.) "When other factors were adjusted, passive exposure to tobacco smoke was associated with an increased ratio of TOTAL-C/HDL-C and decreased HDL-C concentration of between 7% and 9%. However, compared with other factorssuch as BMI [body mass index] or triglyceride concentration, the impact of passive smoking on the ratio of TOTAL-C/HDL-C was relatively small." "The assodation of passive exposure to tobacco smoke with reduced HDL-C and elevated ratio of TOTAL-C/HDL-C is biologically plausible, inasmuch as several investigators have found that dgarerte smoking results in a lowering of HDL- "The effect of tobacco on lipid levels provides one plausible mechanism (among others such as platelet aggregation, vasoactivity, and compromised oxygen transport) for the wall-established elevation of coronary heart disease risk among smokers and suggests a mechanism for the possible increased coronary heart disease risk in passive smokers." [11] "Health Scare - the Misuse of Science in Public Health Policy," J.R. Johnstone and C. Ulyatt, Australian Institute for Public Policy, Critical Issues No. 14, 99 pp., 1991 [Vol. I, Issue 4, July 12, 1991] This commenrmy by Johnstone provides a critique of the failure of risk factor intervention trials to achieve the reductions in coronary heart disease incidence that might have been expected from the publidty generated by various health authorities based on the epidemiological data. ExceLlent, readable commentary by Johnstone on ETS and lung cancer and an especially devastating critique of the failure of risk factor intervention trials to achieve the reductions in coronary heart disease inddence that might have been expected from the publidty generated by various health authorities based on the epidemiological data. On ETS: "At face value the published studies do little to inspire confidence in anything stronger than a 'not proven' verdict, if not one of'not guilty.' Would such results lead to the banning of barbecues or the exclusion of the deodorised from public sodery? The notion would, I think, be dismissed as absurd and resting on quite inadequate evidence, even though both meat smoke and deodorants may contain cardnogens. But barbecues and the deodorised are not scapegoats. Not yet." On intervention trials: "Our ill-health and death are our own fault. If we lived properly, that is, if we lived as the wowsers and puritans think we should, then we would live forever, or at least a very long tLm~ That is the tadt morality which underpins much of modem medicine. It is scandal that so many people should die of cardiovascular disease and cancer. (What shau/dpeople die of?.)" [emphasis in original] "Medical and scientific vandals have hijacked the tools and results ofsdence and prostituted them to their own ends. Secure in the knowledge that the great majority of a deceived populace believe them, they have untrammeled freedom to persecute oppressed minorities. It is time for change. Let those with an interest in public health and a sense off-air play examine the facts for themselves and draw their own conclusions." Ulyatt's chapter is entitled: "Making the Nanny State Honest." "Two relatively unquestioned assumptions underpin Nanny's actions. First, it is assumed that the messages that Nanny delivers are soundly based on the best sdentific evidence. Second, it is seldom questioned whether it is proper for Nanny to obtrude into the lives of citizens in the way she does. Most of Dr. Johnstone's account challenges the first assumption." Ulyatt challenges the second particularly the tendency for "Nanny" to "exhort rather than inform." RESPIRATORY DISEASES AND CONDITIONS - ADULTS [1] "Does Environmental Tobacco Smoke (ETS) Cause Adverse Health Effects in Susceptible Individuals? A Critical Review of the Scientific Literature: I. Respiratory Disorders, Atopic Allergy and Related Conditions," P. Witorsch, Environ. raen~l Tect~ola~y 13: 323-340, 1992 [Issue 24, Item 37] This article cited 197 sources in a review of the available epidemiologic data on persons with underlying or pre- existing respiratory disorders (e.g., asthma or COPD), atopic allergy, claimed heightened sensitivity, and multiple chemical sensitivity. For these conditions, the author concluded that the available data were inconclu- sive and/or limited with regard to any claimed association with ETS exposure. EXCERPTS: "This paper and a subsequent paper to follow are

I JULY 1992 I i I I I I I I I I I I I i I I I intended to provide a comprehensive, critical review and analysis of the sdentific literature relative to the issue of whether certain groups of individuals are more susceptible than the general population to possible adverse health effects as a result of short- or long-term exposure to ETS. In the present report, Part I, the available data relative to individuals with underlying/pre-existing respiratory disorders, atopic allergy and rdated conditions will be addressed. In the subsequent paper that will follow in the near future, the literature relevant to individuals with cardiovascular disorders, pregnant women, persons taking various drugs and medications, and the elderly will be reviewed and analyzed." "[T]he epidemiological data with respect to long-term effects of ETS exposure on adult asthmatics are quite limited and inconclusive and do not support a finding of an asmdation between exposure and adverse effects. Similarly, the somewhat larger (although still reladvdy few) number of experimental studies of acute effects of ETS exposure in asthmatic adults have produced contradictory and equally inconclusive results. While it appears that a small subset of individuals with asthma may react adversely to ETS exposure with worsening of respiratoty flow rates and, in some cases, even clinical exacerbations of their disease, the reasons for these reactions remain to be determined. Likdy reasons include poor conr_rol of asth.n'm, with increased airway inflammation resulting in a lowered threshold for bronchial reactivity, not only to tobacco smoke but to a large variety of specific and non-specific stimuli as well, and a psychogenically-mediated reaction. These mechanisms axe not mutually exclusive. In any event, there is no evidence that such exacerbations, whatever thdr reasons, result in any long-term effecr.s or alter the course of the disease." "The data with respect to potential effects of both short- term and long-term exposure to ETS on Chronic Oh- • structive Pulmonary Disease (COPD) and other chronic respiratory disorders, for all practical purposes, are virtually non-existent." "It is... not surprising that it has been suggested that a number of complaints putatively attributable to ETS exposure, such as eye and upper respiratory 'irritation', as well as exacerbations of asthma, may be related to . . . allergic reactions to tobacco smoke in susceptible atopic individuals. Nevertheless, studies consistently have failed to find any correlation between subjective complaints of sensitivity to tobacco smoke and either skin or serologic tests of immunologic reactivity to tobacco-related antigens." "ETS can cause annoyance most likely related to odor perception, and eye and upper respiratory irritation, most likely on a non-specific, non-immunologic irritant basis. Such responses to ETS exposure occur in a dose (concen- tration) related fashion, dependent, in part, on levels of ventilation and relative humidity. Evidence also suggests that attitudinal emotional and psychogenic factors play a role in such reactions, at least in some individuals. It has not been convindngly demonstrated however, that certain individuals are particularly hypersensitive or susceptible to such effects on any physiological basis." "[E]xposure to tobacco smoke, including ETS, has been included among the various exposures alleged to predpitate symptoms or perpetuate chronic illness in individuals said to suffer from MCS [multiple chemical sensitivity]. However, the only support offered for this contention are subjective reports lacking objective verification; no acceptable sdendfic studies provide any foundation for such relationship. In short, the unscientific nature of these 'data', the highly questionable validity of the proposed etiologic bases and pathogenetic mechanisms for this condition, and the likdihood that many, if not most, of these individuals suffer from a psychiatric dimmer or adhere to a medical subculture or bellefsystem, coupled with the considerable doubt that exists regarding even the basic concept of MCS as a legiti- mate, defined disease, disorder, syndrome, or nosologic entity, make it very difficult to conclude that individuals said to have MCS are more susceptible to claimed health risks of exposure to ETS than the general population." [2] "Passive Cigarette Smoke-Challenge Studies: Increase in Bmnchlal HyperactMty," P. Menon, RJ. Rando, R.P. Stankus, J.E. Salvaggio and S.B. Lehrer, Journal of Allergy and Clinical Immunology 89: 560-566, 1992 [Vol. I, Issue 22, May 22, 1992] Self-reported "smoke-sensitive" subjects were exposed to ETS in a test chamber. Five of 31 asthmatic subjects reportedly "reacted" to ETS; none of 39 nonasthmadc subjects "reacted" to ETS. The authors reported that almost one-third of the asthmatic and one-fifth of the nonasthmatic subjects exhibited increased bronchial hyperactivity for periods of as long as eight weeks follow- ing the test exposure. "Degree and duration of bronchial hyperactivity (BHR) after environmental tobacco smoke (ETS) inhalation was assessed in 31 smoke-sensitive subjects with asthma who exhibited lower airway symptoms on ETS exposure (group I) and 39 smoke-sensitive subjects without asthma who manifested only upper airway symptoms on ciga- rette-smoke exposure (group II). Subjects were challenged with ETS for 4 hours in a static-test chamber." I -17-

ETS AND LaiQ REFERENCE JULY 1992 "Reactivity to ETS is uniquLg, occurs after 60 or more minutes of continuous cigarette-smoke exposure, and does not represent classic immediate or late pulmonary reactivity. The present study was designed to assess and characterize any increase in BHR after a single ETS challenge. This was achieved by serial MCs [methacholine challenges] administered before and after passive ciga- rette-smoke exposure." "All subjects were 'smoke sensitive' in that the subjects with asthma reported chest tighmess, shormess of breath, cough, or wheezing, and the subjects without asthma reported only upper respiratory symptoms on exposure to ETS." "The study subjects were exposed to mixed mainstream- sidestrearn smoke. During the 4-hour ETS challenge, airborne partioalate levds of 800 cpm were maintained that resulted in a TSP [total suspended particles] of 1266 +283 I~g/m3 and airborne nicotine level of 226 +49 gg/m3." "Five of the 31 SS [smoke sensitive] subjects with asthma reacted (> 20 percent fall from baseline FEV~ [a measure of pulmonary function]) on ETS challenge. These five subjects demonstrated persistent reactivity to ETS on repeated challenges for at least 3 consecutive years .... None of the 39 SS subjects without asthma demonstrated a sigrfificant decline in pulmonary function during or aleter 4-hour ETS challenge." "Our studies of passive dgarerte-smoke challenge in nonsmokers demonstrate that almost 1/3 of SS subjects with asthma and 1/5 of SS subjects without asthma have a marked increase in BHR 6 hours after ETS exposure. The increase in BHR varied. One subject with asthma had a 16-fold increase in BHR. Two other subjects, one subject with asthma and one without, each had increased BLIP, that persisted for up to 8 weeks after cigarette-smoke exposure .... The fact that all subjects were atopic may limit the applicability of these results to the general population." "It is of particular interest r_hat an increase in BHR was observed in asymptomatic subjects without asthma after ETS exposure. Since 'dose' and duration of passive dgaret~e- smoke exposure may have been important factors in their responses, every effort was made during the challenge to simulate ETS levds similar to levels encountered in real life. The dgarerte-smoke levels used in our bronchoprovocation studies have been reported in public places. However, as is the case with allergen challenges, the subjects may have been exposed to rdativdy higher doses of ETS in a shorter period of ti.me during challenge. Some of the subjects demonstrated increased BHR for prolonged periods of time (8 weeks). This finding could be due to ongoing dgaretxe-smoke -18- exposure at home, work, or elsewhere. Six of 11 subjects with asthma and two of seven without asthma who demon- strated increased BHR reported dgarerte-smoke exposure at home or work." "Clearly, passive dgarette-smoke challenge studies performed in the laboratory are limited by experimental conditions and thus cannot always be equated with ETS exposure in real life situation [sic]. However, we believe that our finding of increased mecholyl-induced BHR after ETS exposure, even in asymptomatic subjects, is relevant to the issue of dgarerte-smoke reactivity. Allergen avoidance has been demonstrated to decrease BHR and improve symptoms in a group of subjects with asthma, and it is possible that a similar beneficial effect could occur by avoidance of ETS." [3] "Multipollutant Exposures and Health Responses to Particulate Matter," M.D. Lebowitz, J.J. Q.uackenboss, M. ~owskl, M.tC O'Rourke and C. Hayes, Archives of Environmental Health 47(1): 71-75, 1992 [Vol. I, Issue 21, May8, 1992] This article reports on an ongoing study in Tucson, Arizona, which is investigating lung function and particulate matter exposures. The authors report that ETS is related to some respiratory symptoms. "A study of indoor-outdoor (total) exposures and their respiratory effecr.s is underway in Tucson, Arizona. The objective of the study is to evaluate responses of bronchial responsive subjects (those with greater than normal bron- chial lability who are likely to respond more to environmen- tal stimuli) and matched healthy subjects to total and combined air pollutant exposures. Ofspedal interest are the combinations of pollutants from several sources, including combustion emissions (gas, wood burrfingo ETS), HCHO [formaldehyde] (from materials), and outdoor sources. The quality assurance and analytical procedures are based on previous studies and are described elsewhere." "Each day individuals recorded in a diary the total time that they spent in five location categories: (1) home; (2) work/school; (3) outside, near roads; (4) outside, away from roads; (5) other indoors. Individuals spent more than 65 percent of their time at home, and the average time spent outdoors was 3 l-dd. Basic environmental inventory ques- tionnaires and weekly household ac~:ivity records were completed for each home so that presence and use o£source and removal factors could be identified." "As found previously, the indoor levels of [particulate matter, PM] were markedly higher in homes where I i i I I I I I i I I I I I i I I I

VOLUME I, ISSUES 1-24 JULY 1992 i II l !1 I I I I I -cigarettes were smoked at the time of the survey than in homes with no ETS, especially during the winter." "Estimates of personal exposure to [one measurement of PM] correlated well with indoor home estimates. This was expected because people spend more than 65 percent of their time at home. The measured indoor PM concen- tration was a predictor of morning PEFR [peak expiratory flow rate], as was the index of ETS in the homes. The interaction of the two was related to increased bronchial responsiveness. There was an increased prevalence of acute respiratory symptoms as indoor ETS and PM increased, espedally in lower SES [socioeconomic status] households; an increased inddence ofnonspecific symptoms was related to increased ETS." "Different forms of outdoor and indoor PM interact with each other and with certain gases. Surprisingly, other expected interactions were not seen. Formaldehyde did not interact with pollen or NO~ in relation to bronchial responsiveness went unaffected; nor did PM interact with NOu, as evidenced by the lack of symptoms. These and other interactions of indoor and outdoor PM, and interactions with other indoor or outdoor pollutants, need further exploration when the data set is larger." [4] "Respiratory Irritation from Environmental Tobacco Smoke," R.J. Shephard, Archives of Environmental Health 47 (2): 123-130, 1992 [Vol I, Issue 21, May 8, 1992] The author reviews sdentific literature on the alleged "irritant respiratory effects" of ETS. In his opinion, ETS exposure "induces" small changes in respiratory function, which can "progress to chronic reactions," particularly in children. "Approximatdy 81 percent of adults in North America believe that ETS is harrnfiA to health. It is also claimed that neither 'common courtesy' nor expensive ventilation systems will reduce ETS concentrations to levds that will afford adequate health protection to the nonsmoker. In light of the vocal lobbies that argue for and against legislation that would guarantee smoke-free indoor air, it is important to weigh concentrations of ETS, while being watchful for any possible biases in the publication of data." "This review is confined to the irritant respiratory effects of ETS .... Issues to be considered herein include (a) the potential for a noxious effect of ETS on community health, given our current knowledge of the likdy exposure dose at home and in the workplace and the proportion of nonsmok- ers who are so exposed; (b) the likely site of any effect (nose or bronchial airways); (c) the extent of physiological changes; (d) any discrepancy between reported symptoms and the magnitude of physiological responses; (e) possible differences in sensitivity between asthmatic and nonasthmatic subjects; and (f) the potential for progression from a physiological response to a pathological reaction, e.g., chronic obstructive lung disease (COLD)." "ETS exposure induces only small immediate changes of respiratory function. Weak psychogenic effects may contribute to, but cannot entirely explain, symptoms and lung function changes. There is mounting evidence (stronger in young children than in adults) that acute responses to ETS can progress to chronic reactions and can result in an increased prevalence of chronic respira- tory disease. ETS can also aggravate asthma, particularly in subjects who do not have a family history atopy. It is difficult to calculate dgarette equivalents because patterns of breathing differ between smokers and nonsmokers, and ETS differs in chemical composition from mainstream cigarette smoke. However, cotinine estimates suggest that in a smoking household, exposure averages the equivalent of 0.2-0.5 dgarettes [per day]. The increased risk of chronic respiratory lung disease and possible exacerbation of asthma are two additional reasons for reducing expo- sure of the nonsmoker to ETS." [5] =Environmental Tobacco Smoke: Causative Agent or White Hephant?" R. Rylander, Archives of Environmental Health 47(2): 102-103, 1992 [Vol. I, Issue 21, May8, 1992] In this editorial response to the Shephard paper, summarized above, Rylander advocates continued examination of the scientific data on ETS, with attention to such topics as biological plausibility and confounders. EXCERPTS: "Roy J. Shepard [sic] presents a review of respiratory irritation and ETS. He concludes, primarily on the basis of experimental challenge studies, that ETS causes only small, immediate changes in respiratory function. However, he finds that evidence, derived mainly from epidemiological studies, suggests that these acute changes may progress to c.ahronic reactions and an increased prevalence of chronic respiratory disease. The review is a logical interpretation of the literature." "When most sdentific data are considered, and perhaps particularly data from epidemiological studies, expert opinion may be divided. From a sdentific point of view, it is extremely important that such diverse opinions -- in this case, on ETS and respiratory disease -- continue to be discussed, regardless of whether they support the -19-

ETS AND IAQ REFERENCE JULY 1992 ,! Environmental Protection Agency, the tobacco industry, or the politician." "[A] wide range of disparate health consequences, including breast cancer and other diseases unrelated to active smoking, have been associated statistically with ETS exposure. The biological plausibility that ETS causes such a wide variety of effects.., must be rather low." "Could it be that we are committing a fundamental error by placing ETS in the category of a causative factor, when in reality it may be studying a confounder?" [6] "Asthmatic Responses to Passive Cigarette Smoke: Persistence of Reactivity and Effect of Medica- tions," P.K. Menon, R.P. Stankus, R.J. Rando, J.E. Salvaggio and S.B. Lehrer, Journal of Allergy and ClinicalImmunology 88: 861-869, 1991 [Vol. I, Issue 14, January 17, 1992] In this study, a group of 15 "smoke-sensitive" asthmatic subjec'cs and a group of 15 nonasthmatic subjects "with documented upper respiratory tract symptoms on exposure to ETS" were exposed for periods of two to six hours to ETS generar~xt in an inhalation chamber. The asthmatic subjects who had been described as "reactors" two years earlier reportedly remained reactive within one to two hours of continuous ETS exposure. The asthmatic subjects who had been described as "nonreactors" remained nonreactive after six hours of exposure` The authors also reported that pretreatment with certain drugs significantly diminished airway reactivity to ETS. "The present study assessed the persistence of cigarette- smoke reactivity and the effects of drug pretreatment on bronchial responsiveness to environmental tobacco smoke (ETS). Two groups of subjects were chosen for the study." "Group I: SSA consisted of 15 adult, smoke-sensitive, atopic subjects with asthma who claimed symptoms of asthma on exposure to ETS .... All subjects with asthma were 'smoke sensitive'; that is, they claimed symptoms of chest tighmess, shormess of breath, and/or wheezing on exposure to ETS .... Group II: SSNA consisted of 15 atopic volunteers without asthma .... All 15 subjects without asthma claimed upper respiratory tract symptoms (rhinorrhea, nasal obstruction, or sneeze) on exposure to ETS. Atopy was defined as consisting of symptoms of rhinitis and/or asthma plus immediate wheal-and-flare skin test reactivity to two or more common inhalant allergens from a skin test panel of 20 common seasonal environmental allergens." -20- "All subjects answered a questionnaire about their symp- toms, past medical history, current medications, smoking history, and whether they were exposed m passive ETS at home or work. Subjects underwent methacholine chaklenge and skin testing before challenge with ETS. The SSA (group I) had a 2- to 6-hour challenge with 800 cpm of ETS. All reactors to ETS were subjected to a sham challenge for 6 hours in the test chamber in the absence of ETS exposure. All nine SSA no~eacmrs were subjected to a 6-hour challenge with 800 cpm ofETS 4 weeks after the first challenge. All SSNA also underwent 6-hour ETSCs with 800 cpm of ETS." "Five/six subjects in group I, who were previously demonstrated as reactors 24 months earlier, remained reactive within 1 to 2 hours of continuous ETS exposure. Pretrearment with albuterol, cromolyn, and a combina- tion ofalbuterol and cromolyn 30 minutes before ETS exposure significantly diminished airway reactivity to ETS. All nine previous nonreactors in group I remained nonreactive despite rechallenge with ETS for up to 6 hours. Group II subjects challenged under identical conditions did not reveal a significant decline in FEV1 on challenge with ETS. These studies demonstrate the persistence of ETS reactivity during a 2-year period. Although cromolyn sodium and/or albuterol can protect against reactivity, mechanisms of ETS-induced airway reactivity remain unknown." "The findings of the present study confirm our previous observations of reactivity to ETS in SSA and demonstrate that this airway reactivity to ETS remains unchanged for relatively long (at least 2 years) periods of time in most smoke-sensitive subjects with asthma. Of equal impor- tance is the fact that nine previous nonreactors remained nonreactive on reexposure to ETS during this same time period." "Although the mechanism(s) of ETS-induced airway responses are as yet undefined, pretreatment with cromolyn sodium and albuterol can protect against ETS reac-dvity. Additional studies to define the role of airway inflammation, the change in airway reactivity, and the mediators involved in ETS-induced bronchospasm are dearly warranted, and such studies are now in progress." [7] "Respiratory Symptoms in Indian Women Using Domestic Cooking Fuels," D. Behera and SAC Jindal, Chest 100(2): 385-388, 1991 [Vol. I, Issue 12, December 10, 1991] This study examined respiratory symptoms in 3,701 Northern Indian women who used biomass, liquefied petroleum gas, kerosene, or mixed fuels for cooking. I i I ! ! ! ! l 1 ! ! ! ! I I ! I !

VOLUMF~ ~ [ggUF~ 1-24 JULY 1992 I I ! I I I I I I I I I ! I I II I J Symptoms were most prevalent among mixed fuel users, but regardless of the type of fuel used, at least 10 percent of users had respiratory symptoms. "The effect of domestic cooking fuels producing various respiratory symptoms was studied in 3,701 women. Of these, 3,608 were nonsmoking women who used four different types of cooking fuels: biomass [dried dung, crop residues, agricultural wastes], LPG [liquefied petroleum gas], kerosene, and mixed fuels." "Domestic cooking is one of the important functions for the average Indian housewife. The number of hours spent in the kitchen for domestic work and cooking is variable depending on the burden of extra work. On an average, an Indian housewife spends about six hours in the kitchen daily for cooking food and other purposes and because of sociocultural reasons, she is exposed to the fuel at an early age of about 15 years. Therefore, during her lifetime, she is exposed for 30 to 40 years, equivalent to 60,000 hours. The location of the kitchen, the type of ventilation, and the type of fuel used play a significant role on health." "Mixed fuei users experienced more respiratory symp- toms (16.7 percent), followed by biomass (12.6 percent), stove (11.4 percent), and LPG (9.9 percent). Chronic bronchitis in chulla [a cooking device burning biomass fuel] users was significantly higher than that in kerosene and LPG users (p<0.05). Dyspnea and posmasal drip were significantly higher in the women using mixed fuels. Smoking women who are also exposed to cooking fuels experienced respiratory symptoms more often than nonsmokers (33.3 percent vs 13 percent)." [8] "Domestic Pollution and Respiratory Illness in a Himalayan Village," T. Norboo, M. Yahya, N.G. • Bruce, J.A. Heady and K.P. Ball, International Journal of Epidemiology 20 (3): 749-757, 1991 [Vol. I, Issue 12, December 10, 1991] In a study of respiratory illness among villagers in northern India, pollution produced by indoor fires, burning wood and dried yak dung was implicated. "Summer and wLnter surveys of a village of Ladakh have been used to study respiratory illnesses and domestic pollution from fires in an arid high altitude region of northern India. The prevalence of villagers with a forced expiratory volume in one second/forced vital capadty ratio of less than 65% also rose with age. Lung function was significantly worse in those reporting chronic cough, independently of age and sex.... In non-smoking men and the women, levels of exhaled CO were very significantly higher in winter than in summer, as were the levels of CO measured in the houses .... During winter, fires without chimneys gave higher levels of house pollution and indi- vidual CO in exhaled air than those with chimneys." "The study set out to invesrdgate clinical impressions that the villagers of Chuchot experienced a considerable burden of chronic respiratory disease, and this appears to have been confirmed." "Pollution from domestic fires was suspected of being a cause. Heating and cooking in the home is provided by fires burning wood and dried yak dung, and many of these fires either have no chimney, or at best emit a lot of smoke and fumes into the living accommodation. During winter, when outdoor temperatures can fall to -30°C, families spend long periods in poorly ventilated and heavily polluted rooms." [9] "Respiratory Illness in Nonsmokers Chronically Exposed to Tobacco Smoke in the Work Place," J.R. White, H.F. Froeb and Jaei. Kulik, Chest 100(1): 39-43, 1991 [Vol. I, Issue 7, September 13, 1991] In a study based on 40 individuals exposed to tobacco smoke in the workplace and 40 control subjects, the authors used daily activity diaries, respiratory question- nalres, and carbon monoxide measurement as a marker for ETS exposure to investigate respiratory symptoms, eye irritation and chest colds. They reported statistically significant increases in the incidence of all three end- points in the exposed individuals, and concluded that workers and employers could undergo significant finan- cial loss due to illnesses resulting from ETS exposure in the workplace. [Two of the authors of this paper, White and Froeb, were responsible for a widely dted 1980 paper which asserted that nonsmokers exposed to tobacco smoke ar work for 20 or more years had reduced function of the small airways compared to nonsmokers who did not have such exposures.] "['T] he purpose of this study was to determine if tobacco smoke exposure in the work place put coworkers at greater risk of developing respiratory symptoms, eye irritation and chest colds than nonexposed coworkers." Final analyses were based on data from "40 passive smokers (nonsmokers chronically exposed [_> 12 months] to tobacco smoke in the work place) and 40 control subjects." The subjects were required to maintain a diary of daily activities for nine months and "responded to a -21-

ETS AND LaiQ REFERENCE JULY 1992 -respiratory questionnaire.. ; administered by a trained technidan during one of three visits to each work place." ETS exposure was measured by recording carbon monox- ide (CO) levels on each subject's desk continuously for 24 hours on three different occasions. "Carbon monoxide is an accurate tobacco smoke marker when no other sources of CO are present and when the ambient levels of CO are subtracted from the measured level." "Mean CO levels at the beginning of the workday are not significantly different. By mid-morning coffee break, however, passive smokers experience twice the level of CO exposure compared with nonsmokers. The CO level fails shortly after 12:00 noon when many leave the work area, then peaks immediately after lunch hour and remains high into the evening hours. Where smoking is permitted, ventilation [not measured in this study] is not sufficient to sustain CO concentrations below 9 ppm... "[P]assive smokers missed twice as many workdays due to chest illness during an average 12-month period than did nonsmokers .... p<0.001. Rated eye irritation was likewise significantly greater for the passive smokers than for the nonsmokers .... p<0.001." "Passive smokers reported more respiratory symptoms than nonsmokers. Specifically, passive smokers were significantly (p<0.001) more likely than nonsmokers to report chronic cough symptoms, chronic phlegm symp- toms, shormess of breath and chest illnesses." "Nonsmoking workers and their employers are likely to incur significant finandai loss because of missed workdays due to illnesses resulting from exposure to second-hand tobacco smoke." [10] "Effects of Air Pollution on Adult Pulmonary Function," X. Xu, D. Dockery and L. Wang, Archives of Environmental Healzh 46(4): 198-206, 1991 [Vol. I, Issue 7, September 13, 1991] In a study of 1,440 adults in Beijing, China, the authors reported that heating with a coal stove was a risk factor for impaired pulmonary function and was a confounding factor in analyses of possible effects of outdoor air pollution on lung function. SOu and total suspended particulate matter were reported to be assodated with decreases in lung function. This respiratory epidemiologic study was conducted in three representative areas of Beijing (residential, subur- -22- ban, and industrial) in August 1986. Lung function measurements were made of 1,440 never-smoking adults aged 40-69 years and compared to indoor and outdoor air pollution data gathered by interview and from WHO outdoor air monitoring stations, respectively. "IT]he presence ofcoai stove heating was used as a proxy for indoor air pollution... Passive exposures to active cigarette smoking at home or at the office were also accounted for." "The impact of dgarette smoking on pulmonary function and its interaction with age, sex, and air poilu- tion will be investigated in subsequent reports." "There were substantiai differences in passive smoking status (i.e., males were much more exposed than females), and exposures occurred most frequently in the industriai area compared with the suburban area, in which expo- sures were the least frequent." Levds of both SO2 and TSPM (total suspended particulate matter) were reported to be assodated with a decrease in FEV,a (forced expiratory volume in 1 second) and FVC (forced vital ¢~padty). "The estimated effect should not be interpreted as a unique contribution from a single pollutant because SO2 and TSPM are strongly correlated." "Use ofa coai stove for heating was found to be inde- pendently associated with a... reduction in FEV~.0 and. •. in FVC. These findings suggested that both indoor and outdoor pollutants had significant effects on pulmonary function in this population." "[U]se of coai stove heating was not only a risk factor for pulmonary function, but it was an important confound- ing factor to be controlled in the analysis of outdoor air pollution effects." [11] "The Environment and the Lung: Changing Perspectives," J.M. Samet and M. Utell, JAMA 266(5): 670-675, 1991 [-Vol. I, Issue 7, Septem- ber 13, 1991] This paper reviews asbestos, radon, ETS, acidic aerosols and sulfur dioxide, and oxidant pollutants with regard to available data on lung disease. The authors conclude that public policy on ETS should go forward despite the presence of unresolved scientific questions. [Dr. Samet was a member of the Science Advisory Board Committee that reviewed the EPA's risk assessment on ETS.] Asbestos, radon, ETS, acidic aerosols and sulfur dioxide, and oxidant pollutants, including ozone and nitrogen I I I I I I I I I I I I I

VC)t,~'ME 1, ISSxJES I-2& JULY 1992 l ! I i l ! ! ! ! ! ! l ! l ! ! ! ! dioxide, are reviewed. "This shift in emphasis from higher exposures producing clinical disease to lower levels projected to increase population risks has raised new questions and challenges for research on environmental lung disease. Providing assurance of safety, a level of risk judged to be acceptable, requires predse and confident characterization of risks, and the scope of research needs to extend beyond testing for an exposure-disease assodation to quantification of risk at various leeds of exposure and assessment of factors that modify the exposure-disease rdationship. Data from extensive animal studies and large-scale epidemiologic investigations are often required." "In contrast to the strong and causal assodations of active smoking with lung cancer, the risks of exposure to environ- mental tobacco smoke found in epidemiologic studies have been lower, and methodologic problems have been discussed as an alternative explanation to causality for the assodation. The extent of the lung cancer risk to neeer-smokers caused by exposure to environmental tobacco smoke has been particularly contentious, largely because nonsmokers in public buildings and workplaces involunra.rily inhale environmental tobacco smoke, and unacceptable risks for this exposure would provide a basis for limiting smoking in these locations. The lung cancer risk of exposure to environ- mental tobacco smoke has been primarily assessed by generalizing the exposure-response relationship From the studies of never-smokers exposed to the smoking of their spouses. Uncertainties are evident in this approach, includ- ing the lack of information on total exposure to environmen- tal tobacco smoke and the many assumptions inherent in deriving a general exposure-response relationship from studies of never-smokers exposed at home." "Because of the methodologic difficulties of assessing lifetime exposure to environmental tobacco smoke and predsdy describing risks that are not substantially devated, these uncertainties in assessing the lung cancer risk of environmental tobacco smoke may never be fully resolved, although they remain a subiect of research. Yet, full resolu- tion would seem unnecessary for the evolution of public policy on environmental tobacco smoke, a carcinogen with a readily controllable source. In the case of environmental tobacco smoke, it would be unfortunate if potentially irresolvable sdent[fic uncertainties thwarted control." [12] ``Adverse Health Effects Among Adults Exposed to Home Dampness and Molds," R.E. Dales, R. Burnett and H. Zwanenburg, American Review of Respiratory Disease 143: 505-509, 1991 [Vol. I, Issue 6, August 27, 1991] In a questionnaire study of 14,799 Canadian adults, home dampness and mold were reportedly associated with increased symptoms (both respiratory and nonrespirarory). No data were presented regarding smoking in the household. Data were collected via questionnaire from 14,799 adults aged 21 years or over in Canada. Symptoms considered included cough, wheeze, asthma, chronic respiratory disease, eye irritation and "other symptoms," e.g., headache, fever, nausea. Data were apparently collected on the number of smokers in the home, but there was no discussion of ETS. "We found that home dampness and mold were assodated with increased symptoms among adults. The odds ratios varied from 1.5 for other symptoms to 1.6 to lower respira- tory symptoms. Symptom prevalences increased with increased number of mold sites, supporting a dose-response relationship. Although those who reported asthma and respiratory allergies had more symptoms, the assodation between dampness and symptoms was present in those without these indicators of atopy." "Both respiratory health and home dampness may be indirectly related through socioeconomic status, indoor environmental factors, such as the presence of pets or hobbies that produce dust or fumes, and the type of home heating and cooking fuels used. In the present study the crude associations found between respiratory symptoms and dampness were not substantially altered when these many factors were taken into consideration." "The outdoor environment may aim affect respiratory symptoms and reported dampness indoors. In the present study, however, the assodation between symptoms and dampness persisted after adjusting for regional effects that represented differences in temperature, humidity, and ambient pollution across Canada." "In summary, increased respiratory (and other) com- plaints have been associated with the presence of home dampness and molds among both children and adults in several industrialized countries, including Canada. Our findings suggest that an immediate hypersensitivity allergic response is not responsible, and we recommend that future research concentrate on understanding the pathogenesis of the reported adverse health effects." [ 13] "Upper Respiratory Tract Environmental Tobacco Smoke Sensitivity," R. Bascom, T. Kulle, A. Kagey-Sobotka and D. Proud, American Review of Respiratory Disease 143:1304-1311, 1991 [Vol. I, Issue 5, August 9, 1991] This study investigated the frequency of rhinitis symp- toms among healthy nonsmokers and the upper respira- - 23-

ETS AND IAQKEFERENCE JULY 1992 I ..... tory response to controlled cl:mllenge with tobacco smoke in self-described ETS sensitive (ETS-S) and not ETS sensitive (ETS-NS) subjects. The authors reported statistically significant increases in symptom scores for eye irritation (both groups), nose and throat irritation and rhinorrhea immediately after exposure (both groups), and nasal congestion, headache, cough and chest tightness (ETS-S oniy). "The aims of this study were to determine the frequency of rhinitis symptoms among healthy nonsmokers and then to characterize the symptomatic, physiologic, and inflammatory upper respiratory response to controlled challenge with tobacco smoke in subjects with a history of" ETS sensitivity (ETS-S) compared with subjects with no history of ETS sensitivity (ETS-NS)." The ETS-S and ETS-NS classifications were determined by the authors using the subjects' responses to a questionnaire. Twenty-one subjects (ten ETS-S and eleven ETS-NS) were exposed to sidestream tobacco smoke in an environ- mental chamber. The authors reported statistically significant increases in symptom scores for eye irritation (both groups), nose and throat irritation and rhinorrhea immediately after exposure (both groups), and nasal congestion, headache, cough and chest tightness (ETS-S only). ETS-S subjects were also reported to have statisti- cally significant decrements in three measures of pulmo- nary function following exposure. "Our observation that ETS-S subjects are more likely to be atopic is consistent with others' clinical impression .... The increased responsiveness to ETS among atopic individuals may reflect an antigen-induced neural responsiveness." "In summary, we have demonstrated an upper respira- tory symptomatic and physiologic response to challenge .with tobacco smoke in a group of historically sensitive subjects." RESPIRATORY DISEASES AND CONDITIONS --CHILDREN [1] "The Effect of Maternal Smoking During Preg- nancy on Early Infant Lung Function," J.P. Hanrahan, I.B. Tager, M.R. Segal, T.D. Tosteson, R.G. Castile, H. Van Vunakls, S.T. Weiss and F.E. Speizer, American Review of Respiratory Disease 145: 1129-1135, 1992 [Issue 24, Item 38] This article reported on a study of pulmonary function in 82 four-week old infants. The authors reported that no differences in pulmonary function were found between infants reportedly exposed to ETS and those not so exposed. However, the authors assodated maternal smoking during pregnancy with a significant reduction in one measure of pulmonary function. "We studied the effect of prenatal maternal dgarette smoking on the pulmonary function (PF) of 80 healthy infants tested shortly after birth. Mothers' prenatal smoking was measured by: (1) questionnaire reports at each prenatal visit of the number of dgarettes smoked per day, and (2) urine cotinine concentrations (corrected for creatinMe) obtained at each visit. Infant PF was assessed by partial expiratory flow-volume curves and helium-dilution measure- merit of FRC. Forced expiratory flow rates were significantly lower in infants born to smoking mothers, both when unadjusted and after controlling for infant size, age, sex, and passive exposure to environmental tobacco smoke (ETS) between birth and the time of PF testing .... No differences in pulmonary function were evident among infants exposed and unexposed to ETS in the home after stratifying by prenatal exposure status. We conclude that maternal smoking during pregnancy is assodated with significant reductions in forced expiratory flow rates in young infants." "Multiple regression models did not demonstrate that postnatal ETS exposure of the infant before PF testing was assodated with a decline in any of the eight PF measures studied after controlling for infant weight, age, sex, and maternal prenatal smoking." "Previous investigations have shown that the lung function of 5- to 9-yr old children who live in households with smoking parents is lower than that of children who grow up in households with non-smoking parents, a difference seen mos~ consistently in children exposed to smoking mothers. Two explanations have been offered for the closer rdation- ship between reduced PF and maternal smoking: (1) young children spend more dme with their mothers than with other household smokers, so that a child's exposure most closely parallds maternal smoking behavior, and (2) PF defidts in ETS-exposed children are residual effects of maternal prenatal smoking that are present at birth. The current study suggests that prenatal mammal smoking exposure does exert a detrimental influence on forced expiratory flow levds that are observed in infants shordy after birth. The magnitude of the difference in flow levels between infants exposed and unexposed to maternal smoking during pregnancy is far in excess of the 3 to 5 percent disparity reported in studies of older ETS-exposed children." "Several lines of evidence suggest that the effect of smoking on infant pulmonary function observed in this study is due to prenatal mammal rather than to postnatal passive ETS i l I I I I I I I I I I I I I

JUI.Y 1992 I I I I I I I I I I I I I I I I • -exposure. First;the duration of ETS exposure between birth and PF testing was short, and the magrfitude of the PF difference was much larger than the 3 to 5 percent decre- ment reported in previous studies examining the effects of passive ETS exposure on children's PF. Second .... analysis that included separate terms for prenatal maternal smoking and posmatal infant ETS exposure did not identify any significant decline in infant PF independently attributable to ETS exposure .... Finally, infants of mothers who smoked after bixfla showed no significant differences in PF level when divided into two groups: those mothers who did and did not smoke in the same room as the infant. All these findings suggest that posmatal passive ETS exposure is not itself responsible for the lower flow levels seen in infants of smoking mothers. However, the dose concordance of prenatal and posmatal exposure among infants born to continuously smoking mothers does not exclude the possibility that ETS or another unexamined faczor may contribute to the observed decline~ Moreover, they do not preclude an independent deleterious effect of ETS on lung function if exposure were of longer duration." "A number of other maternal and infant factors that could have biased or modified the assodation between prenatal smoking and reduced infant lung function were evaluated. However, no consistent associations were found between mother's respiratory illness history, pregnancy or delivery complications, or mother's level of pulmonary function and reduced flow measures in young infants." "The implications of rd~is study are potentially far reach~g. Previous investigations have demonstrated that lower levels of PF in infancy are assodated with a subsequent increase of acute respiratory illnesses, espedally those associated with wheezing. Other studies in older children have demonstrated a clear association between passive exposure to maternal smoking and frequency of acute respiratory illness and chronic respiratory conditions such as whe2Mng or asthma. These data suggest that prenatal maternal smoking may be a contributing event to each of these adverse outcomes, possibly by causing a relative decrease in airway size or altering lung elastic properties. Any comprehensive under- standing of how prenatal smoking afieects the level o£1ung function attained by older children or adults requires longitudinal studies where maternal smoking during pregnancy, posmaral ETS exposure of the infant, and other potentially influential factors such as early respiratory illness morbidity are carefully chroMded." [2] "Asthma in Children," G.L Larsen, The New EnglandJournal of Medicine 326(23): 1540-1545, 1992 [Issue 24, Item 39] In a review of purported advances in current medical understanding, factors contributing to mortality, and recommendations regarding medications used to treat asthma, the author recommended that all children with asthma should avoid ETS. He extended that recommen- dation to all children, regardless of their asthmatic status, and advocated intensive efforts to prevent childhood exposure to ETS. "This review discusses the advances in our understand- ing of asthma, examines factors thought to contribute to asthma mortality, and summarizes current recommenda- tions for the five major classes of medications used to treat asthma. The importance of nonpharmacologic therapy is also summarized. Although the emphasis of this review is on childhood asthma, clinical investigations involving adults are dted when there are no comparable studies in children. Defidencies in our understanding of disease mechanisms and therapy relevant to children are also noted." "Airway responsiveness can be defined as the ease with which airways narrow in response to various nonallergic and nonsensitizing stimuli .... The level o£responsive- ness is commonly assessed by measuring lung function before and after the inhalation of increasing concentra- tions of an agent such as methacholine .... Subjects with asthma have a heightened response to methacholine and histamine, such that the amount needed to produce a given decrement in lung function is much smaller than in normal subjects. Thus, the airways of persons with asthma are said to be hyperresponsive." "IT]he diagnosis of asthma and classification of disease severity are usually based on signs and symptoms (wheeze, cough, and chest tighmess) as well as the response to therapy monitored in the clinic and with simple tests of lung function. Nevertheless, the concept that the level of airway responsiveness may correlate with disease severity has important clinical implications when one is consider- ing conditions that change responsiveness." "The level of airway responsiveness is not static; it may increase or decrease in response to various factors. Environmental factors can increase airway responsiveness and produce inflammation. For children, important stimufi include respiratory pathogens (especially viral agents), allergens, and air pollutants, including cigarette smoke. From clinical and mechanistic standpoints, the best-characterized stimuli are aeroallergens .... [A] vicious circ/e can develop in which continuous or re- peated exposure to allergen in sensitized persons insidi- ously increases airway responsiveness. Subsequent exposure to allergen and nonallergic stimufi may then lead more easily to airway obstruction. Thus, with - 25-

ETS AND IAQ_tLEFERENCE JULY 1992 increasing responsiveness, stimuli that normally do not produce symptoms (exercise or smoke) can produce noticeable problems." "Although the use of all classes of asthma medications is undergoing critical reevaluation, other types of therapy have emerged as safe, effective, and less controversial. The avoidance of environmental hazards (allergens, air pollution, and tobacco smoke) that increase airway responsiveness is recommended for all children with asthrm." "Exposure to tobacco smoke is an important environ- mental hazard for all children. Although maternal smoking increases the incidence of lower respiratory illness and diminishes pulmonary function in children, it is also associated with higher rates of asthma, an increased likelihood of the need for asthma medications, and an earlier onset of disease. Parental smoking may contribute to increases in airway responsiveness as early as the first 2 to 10 weeks of life. Passive exposure to smoke is a risk factor for the first episode ofbronchiolitis as well as for recurrent wheezing later in childhood and adolescence. Given the extent of passive exposure to cigarette smoke and its potential long-term consequences on the respira- tory health of children, more intensive efforts to prevent this exposure are increasingly justified." [3] "Passive Smoking Among Schoolchildren in Israel," A.I. Goren and S. Hellman, Environmental Health Perspectives 96:203-211, 1991 [Issue 24, Item 55] The parents of 8,259 Israeli school children filled out a health questionnaire for this survey, which evaluated the possible relationship of environmental and home expo- sures to children's respiratory conditions and pulmonary function. Statistically significant increased frequendes of wheezing or coughing without a cold were reported for children whose fathers or mothers smoked. However, much higher relative risks were reported for mothers' respiratory health. EXCERPTS: "In our study we tried to find out whether, in data sets collected in environmental surveys in Israel, there is a link between parental smoking and respiratory conditions among their children. We studied simultaneously the effects of passive smoking and of other home and envi- ronmental exposures, such as that of socioeconomic status (by crowding and by parental education), heating of homes, oriental origin, paternal and maternal respiratory diseases, and community air pollution, on the prevalence of respiratory symptoms and diseases as well as on pulmonary function tests." ,"By using multivariate analysis methods, we could show the effects of different background variables, especially the effecx of exposure to parental smoking on pulmonary conditions among their children. The study was carried out in different communities regarding home as well as environ- mental exposures, especially air pollution. This multicity study design enabled us to find out whether an effecx of exposure to environmental tobacco smoke characterizes communities with different home and environmental exposures or can be found only in certain set ups." "The frequency of reported respiratory symptoms as related to fathers' smoking habits was found to be higher among children whose fathers smoke than among children of nonsmoking fathers. The excess in respiratory symptoms among children exposed to smoking fathers is statistically significant for cough with cold, cough accompanied by sputum and wheezing with and without cold, and for wheezing accompanied by shormess of breath.~ "Respiratory diseases are also more common among children whose fathers smoke than among children of nonsmoking fathers; for bronchitis the difference in prevalence is statistically significant." "Most respiratory symptoms are more common among children of smoking mothers compared with children of nonsmoking mothers. For part of the symptoms, such as cough with cold, cough accompanied by sputum, and wheezing accompanied by shormess of breath, the excess in prevalence is statistically significant." "Respiratory diseases are also more common among children of smoking mothers; the higher prevalence of asthma and pneumonia among children whose mothers smoke is statistically significant." "The prevalence of most respiratory symptoms rises gradually according to the children's exposure to passive smoking, from those whose parents do not smoke to children with one smoking parent towards children whose parents are both smokers. Part of the differences in prevalence of symptoms, e.g., cough with cold, cough accompanied by sputum, wheezing with cold, and wheezing accompanied by shormess of breath, are statistically significant." "The models include, besides the effect of passive smoke exposure, the effects of other factors such as mothers' respiratory diseases, high home crowding, fathers' origin, heating, and community pollution. All the logistic regressions describing the prevalence of respiratory conditions among children include mothers' respiratory diseases as an important and highly significant factor in the models. The effect of community pollution on the i I I I I I I I I I I I I I I I

VOLUME I, ISSUES 1-24 JULY 1992 ! ! ! I ! ! ! I ! ! I ! ! ! ! I prevalence of respiratory conditions among children is highly significant in many modds; the magnitude ofrdative risks to suffer from respiratory conditions when exposed to commu- nity air poliution is srnalier to that connected with exposure to passive smoke and is much smaller than that connected with mothers' respkatory diseases." "Although lacking adequate exposure assessment, it should be stressed that our findings showing higher occurrence of respiratory conditions connected with passive smoking characterize children living in well- ventilated houses in a country with a warm climate and a relatively short winter. The magnitude of the effect of community air pollution on prevalence of respiratory conditions is similar to that observed for home exposure to passive smoke." "The striking effect of mothers' respiratory diseases on the frequency of respiratory conditions among their children is demonstrated by the magnitude of calculated relative risk values (1.51-6.23) compared with relative risk values found for passive smoking (1.13-1.41) and community pollution (1.23-1.54). It seems that relative risk values in this magnitude characterize home exposures to passive smoke as well as community exposure to air pollution typical to local conditions prevailing in Israel." [4] "Indoor Air Pollution and Acute Respiratory Infections in Children," The Lancet 339: 396-397, 1992 [Issue 24, Item 56] This short article commented on the decline of child- hood pneumonia in developed countries, which is in sharp contrast to the high incidence of pneumonia in developing countries. Indoor air pollution was implicated. as a possible cause, although parental smoking was also claimed to be a "recognised risk factor." "In developed countries pneumonia in childhood is now uncommon outside the neonatal period, and when it does occttr it is seldom fatal. By contrast, in many parts of the devdoping world acute respiratory infection (ARI), espedally pneumonia, remains a major cause of death in childhood. Over the past 60 years better medical care in developed countries has reduced the case-fatality rate from pneumonia, but the substantial fall in the inddence of pneumonia recorded over the same period must largely be due to improved living conditions. We do not know which environmental factors contributed most to this fail, but improved indoor air quality and reduced crowding are likely candidates .... Parental smoking is a recognized risk factor for childhood pneumonia in developed countries, but we should be caudous about extrapolating from findings based on tobacco smoke, which contains specific toxins, to predict the effects of other types of smoke." "Several studies of biomass-fuel-burning households have shown levels of respirable particulate matter, carbon monoxide, and toxic oxides of nitrogen and sulphur well above internationally accepted limits. Moreover, wood burning produces toxic hydrocarbons such as benzo[a]pyrene that are known to be carcinogenic." "The need to reduce exposure of young children to tobacco smoke in all countries is self-evident. In less developed countries control of indoor air pollution by methods such as the introduction of more effident stoves, use of alternative fuels, and improved ventilation can be expected to have many desirable effects. Apart from the ecological and economic advantages, adult health, espedally that of women, will be improved, and it is likely that in young children morbidity and mortality due to ARI will likewise to reduced." [5] "Effects of Passlve Smoking on Lung Growr_h Children," M.D. Lebowitz, D. Sherrill and C.J. Holberg, Pediatric Pulmonolog), 12: 37-42, 1992 [Vol. I, Issue 22, May 22, 1992] The authors reviewed data from an ongoing Tucson, Arizona, study on children's lung function and compared reported ETS extmsure (parental smoking) among groups with normal and low lung function measurements. Only male children with initially low lung function were reported to show any effect from parental smoking. EXCERPTS." "The current objective is to distinguish the different effects of passive smoking on lung development in the distinct groups previously defined as having significandy different lung function grovcth, using the same methodol- ogy. The primary alternate hypothesis is that those with passive smoking exposure in the low lung function group would respond adversely, with poorer lung growth compared to the nonexposed low function group. A secondary hypothesis is that passive smokers with normal lung function will respond adversely compared to their nonexposed counterparts." "The tests in the low lung function groups were not associated with more parental smoking exposure; in females, there were more tests in the high lung function group significantly assodated with more parental smoking exposure. Thus, exposure per se was not the critical determinant of lung growth and development and this association may tend to bias results towards the null hypothesis of no effect." - 27-

AND IAQ_ RY_.FKRENCE JULY 1992 I "Females in both the low and normal function groups showed no effects of passive smoking on lung growth; the growth curves for the exposed groups were well within the confidence intervals of the growth curves of the nonexposed. Exposed males in the normal function group did not differ from nonexposed males of that group, using the same criterion." "Males in the low function group with passive smoking exposure in either survey.., did show very different growth curves from males without such exposure in that low function group .... [M]ales exposed to passive smoke in the low function group had a different pattern of growth of FEV1 .... The FVC in the male low function exposed subgroup also differed from that of the low function nonexposed referent group." "IT]he symptomatology of the subjects was not related to parental smoking, though it was related to parental history of symptoms. Analyses of covariance, with sex, age, and income (SES) as covariables, showed no signifi- cant effect of passive smoking on the maximum PP FEV1 [percent predicted FEV,]. Interactions of parental smoking with disease or active smoking did not affect maximum PP FEV1 either. The lack of significant consistent effects of any of these potential modifiers, even with SES and sex as covariables, led to the conclusion that the probability was quite small, and that the passive smoking effect on lung function growth was modified or amplified by modifiers such as symptoms or diagnosis." [6] "The Extent and Impact of Environmental Tobacco Smoke (ETS) Exposure in Children with Asthma," L.M. Salmun, B.A. Chilmonczyk, K.N. Megathlin, L.M. Neveux, G.J. Knight, A.J. Pulkkinen and J.E. Haddow, Abstract No. 348, Journal of Allm~ and Ch'nical Immunology 89 ( 1): Part 2, 1992 [Vol. I, Issue 21, May 8, 1992] This abstract reports on a study of ETS exposure (assessed via questionnaire), ETS absorption (via urinary cotinine) and asthmatic episodes. The authors reported data suggesting an increased frequency of acute asthma episodes and decrements in lung function among children with higher levels of urinary cotinine. "This study examines ETS exposure (by questionnaire) and absorption (by urine cotinine) in children with documented asthma, and relates the extent of ETS exposure/absorption to frequency of acute asthma episodes and to pulmonary function. The 199 subjects, aged 8 months to 13 years, were recruited from a large asthma-allergy practice. ETS information was collected -28- from parents, urine samples were measured for cofinine, and medical reoords were reviewed to document pulmonary function and acute asthma episodes during the preceding year. Urine cotinine levels corrdated weal with reported ETS exposure, and household exposure was the dominant contributor. Subsequent analyses were based on the cotinine levds because: 1) they are an objective measure, and 2) they reflect actual ETS absorption. Urine cotinine measurements were > [greater than or equal to] 10 ng/rnL in 43 percent of the children, indicating significant ETS absorption. In those children, acute asthma episodes averaged 2.8 during the preceding year, as opposed to 2.3 in those with values < 10 ng/rnL. [This difference was not statistic~k!y significant.] Pulmonary fimction studies were available for 145 of the children. In those with cotinine values > 10 ng/mL, FEV1 averaged 7 percent lower [marginally statistically significant], FEF25.v5 12 percent lower [statistically significant], and FEVI/FVC 4 percent lower [statistically significant]. A trend toward diminished lung capadty was present with increasing cotinine concentration for all three parameters. This provides objective documentation of the need to persuade families of asthmatic children to avoid ETS exposure and identifies the family as the major source." [7] "Effect of Passive Smoking on Asthmatic Children Who Have and Who Have Not Had Atopic Dermatitis," A.B. Murray and B.J. Morrison, Chest 101: 16-18, 1992 [Vol. I, Issue 19, April 10, 1992] This study, of 240 children with asthma, reported that children whose mothers smoked had significantly more severe asthma than those whose mothers did not smoke. Atopic dermatitis (a chronic rash) and recent respiratory infection reportedly were not related to the severity of asthma. • "In our first study population .... we found that asthmatic symptoms were more severe, and that pulmo- nary function test results were lower in those [children] whose mothers were smokers than those whose mothers were nonsmokers." "Not only does passive smoking appear to exacerbate symptoms in those who are already asthmatic, there is also strong evidence that it causes asthma in certain children who would not otherwise have had this disease. •. We demonstrated that children with atopic dermatitis were much more likely to have asthma if the mother smoked than if she did not smoke." "This findinga that passive smoking is a risk factor for causing asthma only in the children who have a history of I I I I I l I l I I l I I I l l l l

VOLUME. I, ISSUES 1-24 JULY 1992 I I I I I I I I I I ! I ! I I f ! ! atopic dermatitis, raises the question as to whether it is also solely asthmatic children with atopic dermatitis whose asthma is aggravated by smoke pollution in r.he home." "In order to answer these questions, we reanalyzed the data from our first study." "We studied 240 children with asthma who were them- selves nonsmokers and had been referred consecutively to our clinic. They were aged 6 to 17 years. The severity of asthma was assessed by symptom score, by spkometry, and, in those who could perform the test reliably, by histamine bronchial challenge test. Those who reported having had a chronic or chronically rdapsing itchy rash in characteristic locations were recorded as having had atopic dermatitis." "[C]hildren whose mothers smoked were found to have significantly more severe asthma than those whose mothers did not smoke. In contrast, atopic dermatitis and recent respiratory infection had no apparent effect on the severity of asthma." "Our analysis.., indicated that this aggravation of their asthma was not confined to those who had had atopic dermatitis; atopic dermatitis was not a predictor of the severity of asthma, either on its own, or combined with maternal smoking." "Although passive smoking may cause asthma only in children who have had atopic dermatitis, we conclude that smoke aggravates asthma in those who have not had atopic dermatitis as well as in those who have. It is therefore appropriate to urge parents of all asthmatic children, even those who have not had atopic dermatitis, to refrain from smo "king when in the house or when in the car vdth the child." [8] =Childhood Asthn~ and Passive Smoking: Urinary Cotinine as a Biomarker of Exposure," tL Etutich, M. Kattan, J. Godbold, D.S. Saltzberg, tCT. Grimm, P.J. Landrigan and D.E. Lilienfeld, Amer/can of RerpiratoryDisease 148: 594-599, 1992 [Vol. I, Isstm 19, April 10, 1992] This study reportedly was designed to test the hypotheses that ETS exposure is a risk factor for asthma and that recent ETS exposure may trigger acute attacks of asthma. The authors reported that smoking by the "maternal caregiver" was the exposure variable most strongly associated with asthma. They reported no effect of ETS exposure on the predpitation of acute asthma attacks. "The aim of d~s study was to test ~wo hypotheses: first, passive smoking is a risk factor for the asthmatic state, and second, recent passive smoke exposure acts as a trigger of acute attacks of asthma. To provide a more objective determination of exposure to tobacco smoke, we measured cotinine in the urine of these children." "Three groups were recruited into a case-control study: 72 acute asthmatic children from the emergency room (ER), 35 nonacute asthmatic children from the asthma clinic, and 121 control children from the ET. Both questionnaire and urinary cotinine/creatinine ratio (CCR) were used to assess passive smoking. Levels of CCR > 30 ng/mg were used to identify children exposed at home." "The passive smoke exposure of the two groups [acute and nonacute] is compared. There was no significant difference in general household smoking. Smoking by the maternal caregiver was more common in the nonacute group. There was no difference in the proportions of children exposed at home as defined by CCR levels at or above 30 ng/mg. The mean CCR was nonsignificantly greater in the acute group (46.2 ng/mg) than among the nonacute children (38.5 ng/mgO." "Because the two asthma groups were similar with regard to demographic characteristics, smoking prevalences, and past ER use, they were combined into a single asthmatic group for comparison with the control group." "Comparing smoking variables, there was no significant difference in the proportions having any smokers at home or in daily dgarette consumption by all smokers. The maternal caregiver, however, was much more likely to smoke among the asthmatic group (OR = 2.0). This was confirmed by the differences in CCR." "CCR was most strongly associated with the maternal caregiver's smoking status [OR = 11.9 (6.3, 22.3)]. Associa- tion with smoking by household smokers other than the maternal caregiver was lower [OR = 3.4 (1.3, 8.7)]." "We found that passive smoking is associated with clinically significant childhood asthma in a sample of children drawn from an inner dty population of mainly Hispanic and African-American children using a hospital's ambulatory care services." "Smoking by the maternal caregiver was the exposure variable most strongly associated with the asthmatic state. • . . [W]e were unable to distinguish among current or past smoking by maternal caregiver or smoking in pregnancy by the biologic mother, as these measures were closely intercorrelated and all significantly assodated with the child's asthmatic status." "We were unable to show an effect of passive smoke exposure on the precipitation of acute asthmatic attacks.. -29-

ETfi AND IAQ_R.EFERASNCE JULY 19~2 I "; . We found no difference between the acute and nonacute groups when the CCR was used as a categorical measure (OR = 0.9), and self-reported smoking by the caregiver was actually more common among the nonacute group (OR = 060." "The clinical implications of this study are clear. Maternal smoking in the households of asthmatic children in this population is all too common. Reduction of this potentially important risk factor should be the target of clinicians and health educators working with the families of these asthmatic children." [9] ~Effects of Environment and Passive Smoking on the Respiratory Health of Children," F. Forastiere, G.M. Corbo, P. Michelozzi, R. Pistelli, N. Agabiti, G. Brancato, G. Ciappl and Cak. Perucci, Int.erna- tionalJournalofEpidemiology 21 (2): 66-73, 1992 [Vol. I, Issue 18, March 20, 1992] This study was designed to evaluate the possible effects of outdoor air pollution and parental smoking on the respira- tory health of children. A quesdormaire was administered to the parents of 2,929 children in an industrial town, the dry of Rome, and three rural communities in Italy. The authors reported that the frequency of most symptoms and illnesses was higher among children from the polluted areas (i.e., the industrial town or Rome). Exposure to parental smoking reportedly was assodated with an increased risk of night cough, snoring and respiratory infections during the first two years of life. '% cross-sectional survey was conducted to evaluate the possible effects of outdoor air pollution and of parental smoking on the respiratory health of children. A total of 3092 primary schoolchil&en living in two polluted areas (an industrial town, Civitavecchia, and the dty of Rome) and in a rural area, were chosen. A self-administered questionnaire was filled in by the parents of 2929 children. A broad spectrum of respiratory symptoms and illnesses were taken as outcome variables. The frequency of most outcome variables was higher among children from the polluted areas than among those growing up in the non-polluted area. Exposure to any passive smoking increased OR of having night cough (OR = 1.8), snoring (OR = 1.4), and respiratory infections during the first 2 years of life (OR = 1.3). A further increase in risk was observed in children whose mothers smoked or • both parents were smokers (asthma, OR = 1.5). When the separate and joint effects of the two exposures were studied, the patterns of OR did not suggest synergism between the two factors. The study indicates that both air pollution and passive smoking cause an increase in respiratory symptoms in -30- children, although there would seem to be no additional effects of the two exposures together." "IT]his study shares with other surveys on the effects of air pollution the problem of a small number of sampling units. It is also assumed that residence location serves as a surrogate for personal exposure to air pollution but we do not have any personal exposure data to provide validation. Even though we attempted to control for several con- founders, it is difficult to avoid the potential confounding effect of urbanization; other differences between urban and rural populations might be postulated." "Our findings on passive smoking agree with previous studies regarding night cough, asthma, early respiratory infections, snoring, and the greater importance of maternal versus paternal smoking." "The combined effect of environmental air pollution and passive smoking has been considered in several epidemiologi- cal studies, but mainly with an eye to detecting confounding rather for evaluating possible interactions." "We found that the combined effect of living in polluted areas and bring exposed to passive smoking was not higher than the sum of the separate effecm. In other words, it appeared that these effects were independent of each other and without additional effects of the two together." "In conclusion, both passive smoking and environmen- tal air pollution have deleterious effects on the respiratory health of children. They may increase the number of sensitive subjects or can act as trigger factors in some susceptible individuals. Their effects seem to be indepen- dent .... From the public health point of view, health benefits for children can be achieved by systematic and parallel efforts to reduce outdoor exposures to pollutants and to discourage parents from smoking." [10] "Lung Function In School-Age Children Who Had Mild Lower Respiratory Illnesses In Early Child- hood," G.L. Strope, P.W. Stewart, F.W. Henderson, S.S. Ivlns, H.C. Stedrnan and M.M. Henry, Amem'- can Review of Rwapiratory Disease 144: 655-662, 1991 [Vol. I, Issue 17, March 6, 1992] This study examined the relationship between patterns of mild lower respiratory illness during the first six years of life and later measures of lung function. There was a suggestion that boys with histories of two or more episodes of wheezing lower respiratory infection during the preschool years had lower average pulmonary func6on values. The authors examined ETS exposure as a potential confounding variable, and stated that the reported association between recurrent preschool wheezing lower respiratory infection and lower average lung function was not atxributable to ETS exposure. ! ! ! I ! I ! ! I ! I I I

VOLUME I, ISSUES 1-24 JULY 1992 ! i i I I II ! ! I I I ! | ! ! f ! "To investigate the relationship between mild LRI [lower respiratory illness] in early childhood and later respiratory health, we studied 159 children, 6 to 18 yr of age, who had prospective documentation of outpatient physician visits for LRI during the fist 6 yr ofllf~" [Questionnaire responses, medical records and lung function tests were used.] "Exposure to tobacco smoke, another possible confound- ing factor in the rdationship between LRI history and lung fi.mction, was assessed using estimates of exposure obtained by questiormaire. None of the children reported current active smoking. Exposure to environmental tobacco smoke was assessed as: (1) any maternal smoking during the first 2 yr of the child's life, (2) any paternal smoking during the first 2 yr of the child's life, (3) any smoking in the home during the first 2 yr of the child's life, and (4) any smoking in the home in the 2 yr prior to pulmonary function testing. In analyses designed to assess the potential confounding role of tobacco smoke exposure in the observed association between recurrent preschool wheezing LRI and later lung function, no evidence was found to suggest that differences in lung function asscrdated with preschool wheezing LRI were attributable to tobacco smoke exposure during early childhood or during the 2 yr prior to evaluation. There was an association between exposure to maternal dgare=e smoking during the first 2 yr of life and preschool wheezing LRI (odds ratio = 2.41 concerning maternal smoke exposure in children who had two or more versus children who had zero to one preschool wheezing LRI; 95% confidence interval = 1.11 to 5.25). Nonwheezing LRI experience was not rdated to maternal dgarette smoke exposure. Paternal dgarevce smoke exposure was not assodated with preschool wheezing or nonwheezing LRI histories." "In this study population, there was no evidence that a single mild preschool wheezing LRI was associated with any persistent differences in spirometry. However, boys with histories of two or more outpatient visits for wheez- ing LRI during the preschool years had lower average [measures of lung function] than did boys with histories of zero or one such illness .... No significant association between preschool wheezing LRI and later lung function was identified in girls." [11] "Effects of Asthma on Pulmonary Function in Children: A Longitudinal Population-Based Study," S.T. Weiss, T.D. Tosteson, M.R. Segal, I.B. Tager, S. Redline and F.E. Spelzer, American Review of RespiratoUl Disease 145: 58-64, 1992 [Vol. I, Issue 16, February 17, 1992] This study reported apparent sex differences in the rdadonship between asthma and lung function devdop- ment, with male children more likely to have asthma but female children experiencing a greater deficit in pulmonary function. Maternil and personal smoking were considered in the study, but the authors concluded that it was unlikely that tither acamunted for the different reported effects of asthma. EXCERPTS: "A population-based cohort of 602 white children [from East Boston, MA], initially aged 5 to 9 yr, was observed prospectively for 13 yr." "Standardized questionnaires were used to obtain a history of respiratory symptoms and illnesses, as well as smoking history and demographic data." "Asthma was defined as an affirmative answer to the question, 'Hasa doctor ever told you that you (your child) have (has) asthma?'" "Male ever-asthmatic subjects did not differ from nonasthmatic males in the percentage of personal ever- smokers, lifetime number of cigarettes smoked, or average number of years smoking. The data for females are r suggestive of a difference between ever-asthmatic and nonasthmatic subjects in the percentage of personal ever- smokers and the average lifetime number of'years smok- ing but not in the average number of cigarettes smoked." "It is unlikely that maternal and personal smoking account for the different effects of asthma on change in lung function in males and females observed in our study. Males and females with asthma did not differ in percent- age of smoking mothers from children without asthma. Asthmatic children were more likely to report personally ever smoking than children without asthma, but signifi- candy so only for females. In addition, the effect of" these variables was adjusted for in the analysis. Finalty, male and female asthmatic subjects did not differ from their nonasthmaric same-sex counterparts in terms of amount and duration of personal cigarette smoking." "In summary, this longitudinal analysis demonsu'ates male-female differences in the influence of asthma on change in lung function... Asthma was more prevalent in males but more severe as measured by level of function and hospitalization in females in this cohort. Even in mild asthma with initially normal pulmonary function, the growxh patterns lie outside the 95O/'o confidence limits of nonasthmatic children. Whether the effects of'asthma on change in pulmonary function observed in this study are linked to the airway abnormalities or are linked in some more fundamental way to hormonal, nutritional, or other influences on lung grov~h and maturation in unknown and unanswerable from the present analysis." -31-

ETg AND IACLREFERENCE JULY 1992 I [12] "Increased Incidence of~Asthma in Children of Smoking Mothers," F.D. Martinez, M. Cline and B. Burrows, Pediat','ics 89(1): 21-26, 1992 [Vol. I, Issue 15, January 31, 1992] [Press coverage of this study was summarized in Vol. I, Issue 14 (January 17, 1992) of this report.] In a prospec- tive study of more than 700 children in Arizona, an association between maternal smoking of 10 or more cigarettes per day and the development of asthma was reported for children whose mothers had a high school or less education. No assodation was repotted for children of less well-educated mothers who smoked less than 10 dgarettes per day or for children whose mothers had more than 12 years of education and smoked any amount. "The aim of this longitudinal study was to determine the rdationship of parental smoking at enrollment (before age 5) to both subsequent incidence of asthma and subsequent lung function in a random sample of children." "Information about dgarette smoking habits and respira- tory symptoms among parents was obtained at the t~ne of" enrollment of their chil&en. Questions about smoking focused on each parent separatdy." "Standardized questionnaires on respiratory symptoms were used to ask parents whether they had recurrent wheeze or chronic cough. Mothers also reported the number of years of formal education, and they were classified in two groups: with (> 12 years) or without more than a high school education (>12 years). We chose this variable as an index of sodoeconomic status because previous studies have shown that other more complex indices ofsodoeconomic level add little to or even obscxire the rdationships observed between levd of education and level of lung function or prevalence of respiratory disease-" "In this prospective study we showed that the risk of devdoping asthma before age 12 was two and a half times higher in children whose mothers smoked 10 or more dgarettes per day at enzoIlment and had 12 or fewer years of formal education than in children of mothers of the same levd of education who were nonsmokers or smoked fewer than 10 dgaretres per day. Maternal smoking had no significant effect on the inddence of asthma among children of more educated mothers. These results were independent of reported respiratory symptoms in parents. Likewise, children of less educated mothers who smoked more than 10 dgarettes per day at enrollment had 15% lower values for percent predicted FEF25~.75~, (a spirome~ic parameter that reflects intrathoradc airway function) during follow-up compared with children of mothers of similar level of education who smoked fewer than 10 dgarettes per day or did not smoke. Maternal smoking had no significant effect on percent predicted FEFa5~.~ among children of mothers with more than 12 years of education." "There are many possible factors which may explain r_he greater effect of maternal smoking on astlwna inddence in children of mothers with 12 or fewer years of formal education. Total dgarette consumption was not higher in less educated than in more educated mothers and, therefore, does not explain our findings. Crowding and worse housing conditions may increase exposure to sidestream cigarette smoke among children of less educated mothers. It could also be argued that knowledge of the possible ill effects of environmental tobacco smoke may have stimulated mothers with better education to avoid smoking when their children were present in the room." "Our results are therefore compatible with the hypoth- esis that the recent increases in prevalence and severity of childhood asthma may be at least in part attributable to an increase in the prevalence of smoking among less educated mothers." "In conclusion, our results suggest that many cases of childhood asthma may be prevented through a sustained effort to discourage smoking initiation and to encourage smoking cessation, particularly among less educated women of childbearing age." [13] "Impaired Pulmonary Function in Schoolchildren Exposed to Passive Smoking," R. Casale, D. Colantortio, M. Cialente, V. Colorizio, R. Barnabei and P. Pasqualetti, Respiration 58: 198-203, 1991 [Vol. I, Issue 15, January 31, 1992] In this study, pulmonary function was evaluated in t43 children aged 6-11 years. Parental smoking was ascertained via questionnaire; urinary cotinine levds were also measured. The authors reported that values for pulmonary function tests were significantly reduced among subjects whose parents smoked. The authors also suggest that children with "greater exposure" had poorer pulmonary function. "The purpose of this study was to examine the effects of passive smoking on pulmonary function in children. Passive smoking exposure was evaluated by questionnaire and by urinary cotinine levels." "The children included in the study were between 6 and 11 years old and attended primary school in L'Aqulla, Italy. We evaluated 152 children, 78 boys and 74 girls, all non-smokers. A questionnaire was distributed to all of them asking for the following: age, sex, smoking habits and number of cigarettes consumed in a day by their I I I l I l I l I I I l l -32- I

VOLUME I, ISSUES 1-24 ,JULY 1992 I I l I I I l I i I I i I l I l I I ! parents and/or people living with them, duration of exposure to passive smoking per day, prior respiration [sic] illnesses, acute respiratory episodes in the last 3 months and eventual consumption of drugs in the month preceding the study. The answers to the questionnaire were obtained from the schoolchildren's parents. The following parameters of respiratory function were evaluated: forced expiratory volume in 1 s (FEV,), forced vital capacity (FVC), peak expiratory flow (PEF), maxi- mal expiratory flow when 50% of the FVC remains in the lung (MEF~0~wc), maximal expiratory flow when 25% of the FVC remains in the lung (MEF25~ ~vc) and maximal mid-expiratory flow (MMEF) .... Urinary cotinine levels were determined by 'Tobacco Screen' [an assay] .... This method can measure urinary cotinine from 0 to 11.0 gg/ ml and is specific for cotinine since no substances have been found to interfere at levels of cotinine that might appear in the urine." "In our study non-exposed ckildren (according to the questionnaire) had mean levels of urinary cotinine less than 0.400 lag/M, a value below that indicating non- exposure. Levels of cotinine separated significantly groups of children at low and high exposure, determined the questionnaire results, even at values inferior to 1.0 gg/ml, which is the discriminant level for the identification of 'heavy smokers' according to the employed kit." "Our study indicates that both a questionnaire and measurement of urinary cotinine levels can identify various degrees of exposure to passive smoking, and that they seem of equal importance. Subdivision into groups resulted in homogeneous distribution regarding age, sex, height and weight." "The subdivision into groups at different degrees of exposure, both by the questionnaire results and by urinary cotinine levels, demonstrated a decrease of pulmonary function test values in exposed cases. The decrease was statistically significant for MEF~0~ v¢~ MEF=5~ wc and MMEF; no significant difference was found for FEV1. The lack of significant differences in the pulmonary function test values between the groups based on the questionnaire and the urinary cotinine levels supports the validity of the two employed methods." "Our study offers further evidences [sic] of a relation- ship between passive smoking and decreased respiratory function. The greater the exposure, the poorer the pulmonary function. Urinary cotinine is a useful marker for quantifying passive smoking, demonstrating that the exposed subjects are similar to light smokers. Measure- ment of urinary cotinine.., is simple and economical, and thus suitable for screening purposes." [14] "Childhood Asthma and the Indoor Environment," C. Dekker, IL Dales, S. Bartlett, B. Brunekreef and H. Zwanenburgo Chert 100(4): 922-926, 1991 [Vol. I, Issue 12, December 10, 1991] This study reports on a questionnaire-based study of 17,962 Canadian schoolchildren. The authors reported that diagnosed asthma was significantly associated with ETS exposure, home dampness, the use of gas for cooking and the use of a humidifier. Wheezing was reportedly associated with all of the a~ve factors except gas cooking. "To investigate the influence of indoor air quality on respiratory health, a questionnaire-based study of 17,962 Canadian schoolchildren ha kindergarten through grade 2 was carried out in 1988. The present report focuses on associations between several indoor environmentai factors and childhood asthma. Increased reports of physician- diagnosed asthma were sigrfificantly associated (p<0.001) with exposure to environmental tobacco smoke (OR= 1.4), living ha a damp home (OR= 1.5), the use of gas for cooking (OR=2.0) and the use of a humidifier (OR=l.7). Wheezing without a diagnosis of asthma also was associated (p<O.01) with environmental tobacco smoke (OR= 1.4), home dampness (OR= 1.6) and humidifier use (OR= 1.4), but not with gas oooking. [By statistical convention, "p<0.001" means that the probability that a certain result occurred by chance is less than one on one thousand; "p<0.01" means that the probability that a certain result occurred by chance is less than one ha one hundred.] Thus, several modifiable risk factors for respiratory illness may exist ha Canadian homes. Further research is required to determine the nature of these cross-sectional observations." "The present large cross-sectional study indicated that gas cooking, exposure to environmental tobacco smoke, home dampness and humidifier use were associated with the prevalence of current asthma. The latter three exposures were associated with wheezing. The OR were generally less than 2 for individual exposures, suggesting effects that were not very large. Although the OR were low, a relatively high proportion of subjects were exposed, resulting in important atr.ributable risks: approximately 20 percent for each of tobacco smoke, and home dampness and mold." [15] "Quantifying Health Aspects of Passive Smoking in British Children Aged 5-11 Years," S. Chinn an,l R.J. Rona, Jour,~ of EpW.,.mio~o~ a.d Community Health 45: 188-194, 1991 [Vol. I, Issue 11, November 22, 1991] In an analysis of existing studies of over 11,000 English and Scottish schoolchildren, the authors reported no -33-

ETS AND IAQ REFERENCE JULY 1992 assodafion between child's height and "passive smoking." They claimed, however, that ETS exposure was associated with persistent wheeze (RR = 1.35 for exposure to 20 cigarettes per day). EXCERPTS: "The aim [of this study] was to estimate the dose- response relations of height and respiratory symptoms to passive smoking in children aged 5-11 years .... The study was an analysis of existing observational studies, comprising three samples: English representative; English inner dry; Scottish representative .... The study popula- tion included 5002 English children from the representa- tive sample, 2903 English inner city children, and 3319 representative Scottish children." "The results of the analysis of height standard deviation scores were consistent with there being no association between a child's height and parental smoking, in contrast to the previous conclusions drawn from the 1982 data analysis." "Our results confirmed the finding ofMelia, Chinn and Rona of greater prevalence of respiratory symptoms in inner dry white children than in the representative sample, and also of greater prevalence of some symptoms of Afro-Caribbeans." "We cannot confirm an association between child's height and passive smoking for ages 5-11 years, but the relation of persistent wheeze to passive smoking consti- tutes a relative risk of around 1.35 on exposure to 20 cigarettes a day in the home, compared to no exposure, and the attributable risk of any respiratory symptom, particularly in inner city children, is not negligible." [16] "Parental Smoking and the Risk of Childhood Asthma," A. Bener, A.R. AI-Frayh Facharzt and T.Q. AI-Jawadi, Journal of Asthma 28(4): 281-286, 1991 [Vol. I, Issue 11, November 22, 1991] In a survey of 3,300 Saudi Arabian children, a positive correlation between parental smoking and asthma was reported. Other respiratory symptoms (wheeze, cough and family history of rhinitis) were also reportedly associated with parental smoking. "In order to explore the correlation between parents' smoking habits and bronchial asthma in children, we undertook a cross-sectional study of 3300 (54% males, 46% females) school children aged 7-12 years old. A survey of smoking habits and attitudes conducted in Saudi Arabia showed a positive correiation between -34- parental smoking and asthma. This study showed a significant llnk between parental smoking and chest wheeze or whistling, cough, and family history ofrhinitis. Evidence is accumulating that there is a relationship between parental smoking and respiratory symptoms in Saudi children. The present study results are dear evidence of a definite assodation between smoking in the home and bronchial asthma in young children, which not only may present immediate problems, but may also be a cause of illness in the future." [17] ~Children With Recurrent Respiratory Tract Infections Tend to Belong to Families With Health Problems," M. Soderstrom, B. Hovelius and tC Prellner, Acta Paeda'atr. &and. 80: 696- 703, 1991 [Vol. I, Issue 10, November 1, 1991] Medical and sodal factors in families with preschool children who had attended day care and had recurring respiratory tract infections [RTIs] were assessed by questionnaire, clinical examination of the children, and medical record review. Diseases, particularly cardiovascu- lar diseases, were reportedly more frequent in the families of children with recurrent RTIs. Fewer of the mothers of children with RTIs were smokers than were mothers of controls, so questions about ETS were not investigated. EXCERPTS~ "The aim of the study has been to assess some medical and sodal factors in families with children who as preschoolers attended day-care centres and had recurrent episodes of antibiotic-treated RTI [respiratory react infections], and compare these wi~h a control group of families with children who had no such RTI or only isolated episodes. The assessment was done by means of a questionnaire answered by the parents, clinical examination of the children, and scrutiny of their medical records." Diseases, particularly cardiovascular diseases, were reportedly more frequent in the families of children with recurrent RTIs (p < 0.01). Similarly, parents of children with RTIs were less satisfied with their own health than were parents of controls. "The present results suggest that children who have suffered from recurrent bacterial RTIs as preschoolers belong to families in which morbidity is high and health problems generally prevalent. This might be a reflection of environmental factors and/or genetic predisposition, involving increased susceptibility both to RTI and to other diseases." "In the present study, however, fewer of the mothers of RTI-afflic~ed children were still smokers than were I I I I ! ! i I I ! i I I ! ! I I !

JULY 1992 l I I I I I I I I I I I I I I I I I mothers of controls .... Accordingly, the effect of passive smoking on the frequency of RTI during the day-care period of the children could not be evaluated." [18] "Passive Smoking and Childhood Asthma," S. Willers, F_. Svenonius and G. Skarping, A//ergy 46: 330-334, 1991 [Vol. I, Issue 10, November 1, 1991] Urine cotinine measurements were used to investigate the relationship between parental smoking and asthma in children. No difference was reported for the prevalence of parental smoking in asthmatics and controls; however, maternal smoking was reported to be statistically signifi- cantly more prevalent among asthmatics [RR = 2.56, 95% CI 1.23-5.32]. Urinary cotinine levels were also reported to be significantly higher in the asthmatic children. EXCERPTS: "Cotinine (the major nicotine metabollte) has been shown to be the biological marker of choice for passive smoking. The aim of this study was to investigate the relationship between passive smoking and asthma in children by using this objective measurement of passive exposure to tobacco smoke." Forty-nine children aged 3-15 diagnosed with asthma were included in the study. One parent answered ques- tions about parental smoking habir.s. Cotinine levels were assayed from urine samples. A referent group of 77 children was used, with parents answering questions about smoking and filling out a questionnaire on the child's respiratory health. The absence of asthma in the controls was not confirmed by examination. "In the asthmatic children, the prevalence of smoking among parents (mother and father) was not statistically significantly higher than among the referents [RR = 1.97, 95% CI 0.90-4.35]. However, maternal smoking was more prevalent among asthmatics [tLR = 2.56, 95% CI 1.23-5.32]." "The cotinine levels in the urine of asthmatic children were significantly higher than in those of the referents." "[P]assive smoking may be a predisposing and/or aggravating factor in asthma." [19] "Indoor Air Pollution Exposure and Lower Respiratory Infections in Young Gambian Chil- dren," J.R.M. Armstrong and H. Campbell, International Journal of Epidemiology 20: 424-429, 1991 [Vol. I, Issue 10, November 1, 1991] In this rural population-based cohort study of approxi- mately 500 Gambian children, the incidence of acute lower respiratory infections (ALPd) was reportedly related to father's smoking habits [OR = 1.8, 95% CA 1.0-3.4]. (Only 3% of mothers reported regular smoking, so it was not investigated.) A significant association between exposure to smoke from cooking fires and the incidence of ALRI was also reported for girls only [OR = 6.0, 95% CI 1.1-34.2]. "In a rural population-based cohort study of approxi- mately 500 Gambian children under five years old followed for one year, incidence of acute lower respiratory infections (ALP, I) was related to various risk factors including parental smoking and regular carriage on the mother's back while cooking, a proxy measure for exposure to smoke from cooking fires." "We found that the incidence of lower respiratory infections in Gambian children is related to their father's smoking habits [OR = 1.8, 95% CI 1.0-3.4]. Due to the much greater contact between mothers and their child#en it would be expected that maternal smoking would have a larger effect. However, only 3% of mothers in our study reported regular smoking, so we were unable to investi- gate this. In this community maternal smoking is unlikely to be an important risk factor." "We were able to show a significant association between smoke exposure and incidence of AH~, but only in girls [boys' OR = 0.5, 95% CI 0.2-1.3; girls' OR = 6.0, 95% CI 1.1-34.2]. Since combustion of biofuels is not generally required for space heating in rural Gambian populations, smoke exposure is predominantly from cooking fires." [20] "Children's Exposure to Environmental Cigarette Smoke Before and After Birth," M.D. Overpeck and A.J. Moss, Advance Data, No. 202, 11 pp., June 18, 1991, National Center for Health Statistics [Vol. I, Issue 7, September 13, 1991] The authors reviewed data from the 1988 National Health Interview Survey on Child Health and concluded that children whose health was reported as "fair" or "poor" were more likely to be exposed to ETS. Neverthe- less, the differences reported were not statistically signifi- Data from the 1988 National Health Interview Survey on Child Health were reviewed to investigate demo- graphic and socioeconomic factors and ETS exposure in U.S. children. "The findings show that a large proportion of children at disadvantage from low income and educational levels -35-

ETS AND IAQREFEI~NCE JULY 1992 ' - in the household are a/so at increased risk of exposure to maternal and sidest~eam smoke, potentially adding to differentials in their health risks." "[A]bout 3 percent of all U.S. children 5 years of age and under are said [by interview respondent, usually a parent] to be in fair or poor health. The relative risk of fair or poor health was almost twice as great for children who lived in households with current smokers as it ,vas for children who were never exposed w 4.1 and 2.4 percent, respectively. This approximate ratio is a/so observed between the estimates for most of the smoking categories .... However, the differences seen are not statistically significant at the 0.05 level when the standard errors associated with these estimates are considered." "Nevertheless, the estimates.., are included in this report to show an apparent pattern suggesting that, for most children, fair or poor health appears to be assodated with various exposures to dgarette smoke .... The estimates.., should be interpreted with caution, how- ever, because sampling variability may account for the differences that are observed and other determinants of perceived health status have not been taken into account." [21] "Environmental Tobacco Smoke Exposure and Respiratory Health in Children: An Updated Critical Review and Analysis of the Epidemlologl- cal Literature," R.D. Hood, J.M. Wu, R.J. Witorsch and P. Witorsch, Indoor Environment 1 : 19-35, 1991 [Vol. I, Issue 7, September 13, 1991] This paper attempts to be a "independent, comprehensive and updated review and objective analysis of the relevant epidemiologic literature" on ETS exposure and the respira- tory health of children. The authors conclude that epidemiologic reports suggest an association between parental smoking and an increased inddence of respiratory symptoms and illnesses in infants and preschool children, but that the reports dealing with older children are inconsis- tent. Several possible explanations are presented. "The purpose of the current study was to conduct an independent, comprehensive and updated review and objective analysis of the relevant epidemiologic literature, in order to determine whether there is consistency among studies with regard to associations between ETS exposure and the respiratory health of children. In addition, we have attempted to provide explanations and/or mecha- nisms for observed consistendes and/or inconsistendes among the reported studles." "IT]he 44 epidemiologic reports dealing with the assodation between parental smoking and respiratory symptoms and diseases in preschool-aged children... generally reported an assodation between parental, usually maternal, smoking and an increased inddence of respiratory symptoms or certain illnesses in infants and preschool children." "The 46 epidemiologic reports dealing with the associa- tion between parental smoking and respiratory symptoms and diseases in predominantly school-aged children are prmented.., although the majority of the studies retx~rted one or more significant relationships between parental smoking and respiratory symptoms or certain diseases, the specifidty of these apparent as~odations varied considerably." "The 17 reports addressing whether middle ear effusion in preschool and school-aged children, as a sign of chronic middle ear disease, is rdated to paternal smoking are listed.. • Less than half of the studies reported statistically significant assodations between rniddie ear disease and parental smoking." "Thirty-eight epidemiologic reports have addressed the issue of parental smoking and pulmonary flmction in school- age or over children .... These data indicate that changes in most of the parameters evaluated were not consistently assodated with parental smoking." "[l]nvestigators addressing the question have noted an apparent age dependency of the assodation of respiratory symptoms/disease with parental smoking .... If such an age dependency is confirmed, it could be due to a number of possible causes. One possibility is that there is an actual cause-and-effect relationship with ETS, as exposure may be greatest during the first year or two of lhre. At this time, the mother-child relationship is the most intimate, and ETS exposure may be more intense than it is later. It is also possible that the very young child, with its immature lungs, could be more sensitive to a number of environmental factors, such as ETS." "Several factors other than ETS (i.e., airborne constituents of tobacco smoke), tither biological or nonbiological, alone or in combination, may account for the reported assodation between parental smoking and increased frequency of respiratory symptoms or certain illnesses in younger chil- &en. Among the possible biological mechanisms.., are pre- or perinatal effects (e.g., effects of maternal active smoking during pregnancy or lactation)." "Among the predominantly nonbiological factors contrib- uting to the association between parental smoking and increased incidence of respiratory symptoms and disease in young children are SES [sodoeconomic status] and related variables .... Other findings also suggest that SES-rdated i l I ! ! l ! I ! I ! ! ! ! ! i I !

VOLUME I~ ISSUF.g 1-24 JULY 1992 I I ! I l I I l I i ! I ! ! I ! ! factors may confound effects presumed to be caused by children's exposure to paren~ smoking." "Active smoking by the child c~uld be a source of bias in studies of children, in that such active smoking is likdy to be a much more important respiratory risk factor than is ETS "In a number of the epidemiologic studies, the authors ignore statistical criteria and emphasize trends that are not staristica.Uy significant as being meaningfiA. This tendency is biased and perpetuates misunderstanding. Lack of statistical significance, while it does not eliminate the possibility of a real assodation, should be considered to be consistent with the null hypothesis, unless proven otherwise." [22] "Acute Effect of Passive Smoking on Lung Function and Airway Responsiveness in Asthmatic Children," M. Oldigs, R. Jorres and H. Magnussen, Pubrmno/ogy 10: 123-131, 1991 [Vol. I, Issue 7, September 13, 1991] Eleven asthmatic children were exposed to machine- smoked cigarette smoke in an exposure chamber. One hour of "passive smoking" produced eye irritation but no consistent changes in lung function and bronchial responsiveness. "In contrast to chronic exposure, little is known on the acute effect of passive smoking in children. We therefore studied symptoms, lung function, and airway responsive- ness of children with bronchial asthma before and after 1 hour exposure to cigarette smoke as compared to control conditions." The subjects were ten boys and one girl with allergic bronchial asthma (aged 8-1.3 years). They were "not selected on symptoms induced by previous exposure to cigarette smoke." A 24 m~ exposure chamber was used. Cigarettes were machine-smoked to generate a target concentration of about 20 ppm CO. "Our observations demonstrate that in children with mild bronchial asthma 1 hour of passive smoking pro- duced mainly eye irritation but no consistent changes in lung function and bronchial responsiveness to inhaled histamine." "[W]e do not believe that our inability to demonstrate an adverse acute effect of passive cigarette smoking on lung function was due to an insufficient reproducibility of lung function data." "lilt is unlikely that our [negative] findings [on airway hyperresponsiveness to inhaled histamine] were due to a poor reprodudbility of bronchial responsiveness measure- ment." "[W]e found elevated levds ofcotinine in five of six children with reported smoke exposure at home. In contrast, two of five children without a positive history showed increased codnine levds." "Since the purpose of our study was to investigate the acute effects of passive smoking and since we did not find an effect.., it is difficult to compare our data with those of chronic exposure studies. Chronic exposure has been demonstrated to increase bronchial responsiveness and to impair lung function .... [I]t would be of interest to study the acute airway response of asthmatic children with and without chronic smoke exposure." [23] "Association of Indoor Nitrogen Dioxide With Respiratory Symptoms and Pulmonary Function in Children," L.M. Neas, D.W. Dockery, J.H. Ware, J.D. Spengler, F.E. Speizer and B.G. Ferris, Jr., American Journal qIVEpidemiology 134(2): 204- 219, 1991 [Vol. I, Issue 7, September 13, 1991] Residential exposure to nitrogen dioxide (NO1) was measured by indoor monitoring for a group of children in the Harvard Six Cities Study. Lower respiratory symp- toms were reported to be assodated with indoor NO2 , levels equivalent to those from gas stove use. The magni- tude of the reported assodation was similar to that of an association reported elsewhere for maternal smoking. "Previously published results from the Harvard Six Cities Study have described the assodafions between respiratory symptoms and pulmonary function in two cohorts of pre- adolescent children with indicators of indoor pollution sources de~ermined from questiormaires: p~ren~ smoking, gas moves, and kerosene heaters. WNle evidence of increased respiratory symptoms and lower lung function has been reported for passive smoke exposure, the assodation with nitrogen dioxide sources has been less oonsistent .... In this paper, we present results from a study of a subset of the second cohort of children.., in which each child's residen- tial exposure to nitrogen dioxide was directly measured by indoor monitoring." "In this study, lower respiratory symptoms were linearly associated with indoor nitrogen dioxide with an odds ratio of 1.40 [95% CI 1.14-1.72] for an increase in nitrogen dioxide equivalent to that for a gas stove. This is similar in magnitude to the previously reported effect of passive smoke exposure. For example, Ware et al. report a relative odds of 1.23 for the association between maternal smoking and an index of lower respiratory illness in an 37-

ETS A~ IAQ ~EFE~E~CE JI.YLY 1992 earlier cohort in these cities. The association is stronger among girls and among children living in smoking homes, but the effect is still present among boys and among children living in nonsmoking homes." "IS]elective inclusion or exclusion of participants is a potential source of bias due to the loss of children over the 3 years of the study." "A second alternative explanation for these findings is that other variables associated with both nitrogen dioxide exposure and respiratory symptom reporting, i.e., confound- ers, may be producing a spurious association. The strongest confounder in this. study was socioeconomic status, and controlling for this and other potential confounders in the model actually increases the association .... Family size, the presence of younger siblings, and the number of persons per room were found to have no significant effect.... Neither excluding parental i]/ness from the model nor including an indicator for maternal smoking during pregnancy altered the odds ratio." "The monotonic increase in the reporting of lower respiratory symptoms with the ordered nitrogen dioxide categories implies that nitrogen dioxide has adverse health effects at levels below the current ambient outdoor standard of 53-ppb annual mean. Since the specific toxic agent may be a product of subsequent reaction of nitro- gen dioxide on indoor surfaces to produce acid gases, future investigations of the indoor chemistry of nitrogen dioxide may suggest alternative mitigation techniques. However .... a direc~ reduction of indoor nitrogen dioxide exposures would have health benefits .... Such exposure reductions could come through the control of indoor nitrogen dioxide sources, through the removal of nitrogen dioxide from outdoor air infiltrating into the home, and through the reduction of ambient nitrogen dioxide concentrations." [24] "The Effects of Indoor Environmental Factors on Respiratory Illness in Primary School Children in Kuala Lumpur," B.H.O. Azizl and R.L. Henry, International Journal of Epidemiology 20 (1): 144- 150, 1991 [Vol. I, Issue 7, September 13, 1991] Primary school children (Malayan, Chinese and Indian) were studied via collection of pulmonary function data and parental questionnaire responses. Several indices of indoor air pollution, namely, tobacco smoking, use of mosquito coils, and cooking, were examined with regard to respiratory symptoms and illnesses. Significantly elevated odds ratios were reported for exposure to mosquito coil smoke and both asthma and persistent wheeze as well as for "passive smoking" and chest illness. EXCERPTS: This c~oss-sectional study of 7-12 year old primary school children in the Kuala Lumpur city area was conducted during July-Oczober 1987. Results are based on analyses of parental questionnaire responses and pulmonary function data from 1501 children (Malayan, Chinese and Indian). "Too few studies have been done to examine these potential [indoor] pollutants in developing tropical countries where the morbidity and mortality due to respiratory illness are of enormous proportions.* "Risk factors examined in this study were indoor factors commonly encountered in Malaysian households. Indoor sources of air pollution belonged to three groups of activity, namely, tobacco smoking, avoiding mosquito bites, and cooking." "Exposure to mosquito coil smoke was confirmed to be independently associated with asthma [defined as persis= tent wheeze and/or doctor diagnosed asthma] (odds ratio = 1.4, p = 0.001) and persistent wheeze (odds ratio = 1.4, p = 0.005), while passive smoking was confirmed to be independently associated with chest illness (odds ratio = 1.7, p = 0.003). Other indoor environmental factors were not significantly associated with any symptom or illness. Interactions were not found between mosquito coil smoke or passive smoking and other environmental and host factors." "The logistic regression analysis showed that sharing a room with a smoker was independently associated with chest illness [defined as a positive reply to the question: 'During the past one year has this child had any chest illness that has kept him/her from his/her usual activities as much as 3 days?']" "This study could not find any relationship between stove type and respiratory symptoms or illnesses." "This study was the first to examine respiratory risk factors in Malaysian children and the results suggested that mosquito coil smoke and passive smoking were avoidable risk factors." [25] "Respiratory Health Effects of Home Dampness and Molds Among Canadian Children," R.E. Dales, H. Zwanenburg, R. Burnett and Caet. Frankl'm, American Journal of Epidemiology 134(2): 196-203, 1991 [Vol. I, Issue 6, August 27, 1991] Home dampness and molds and respiratory health were examined in 13,495 Canadian children. All respiratory I I I I l I I I I I I I I I I I -38-

I JULY 1992 I I I I I I I I I I I I I I I I I I symptoms were reportedly higher in homes with reported mold or dampness. No data were presented concerning parental smoking. This questionnaire-based study included 13,495 children between the ages of five and eight. "Primary exposure variables" used to indicate the presence of dampness or molds included: the number of mold sites in the home, the appearance of damp or wet spots, and basement flooding. Symptoms assessed included cough, wheeze, asthma, chest illness, upper respiratory symp- toms, eye irritation and other nonrespiratory symptoms. Although the number of smokers in the household was assessed, there was little reference to smoking, except for the comment that smoking was more prevalent in damp homes (55o/6 vs. 51%). "All health indicators were associated (p < 0.05) with ~ indicators of home dampness or molds; [crude] odds ratios ranged from 1.14 for flood and bronchitis to 2.26 for cough and two mold sites]." "Odds ratios were adjusted for age, sex, race, education of parent/guardian, gas cooking, number of household smokers, hobbies, sex of respondent, and region of residence. Overall, adjusted odds ratios ranged from 1.08 for wheeze with dyspnea and flooding to 2.55 for the association between cough and number of mold sites." "We found that the reported presence of indoor molds and dampness may cause adverse health effects in Cana- dian children .... 1) A dose-response gradient was demonstrated between the number of mold sites and health outcomes with a maximum odds ratio reaching 2.55 for cough and the presence of two mold sites. 2) The observed relation was independent of the age and sex of the child, the number of household smokers, the presence of gas stoves, and the region of residence in Canada~ 3) Respondents were essentially 'blinded' to the hypothesis tested .... 4) It is unlikely that general over- or underreporting biased the study because stratifying by the question used to indicate reporting bias did not erase the observed relation .... 5) Parents of 'symptomatic' children, searching for etiologic factors, may be more aware of indoor air problems than parents of asymptomatic children." "Thus, although the associations between respiratory symptoms and home dampness or molds are probably causal, doubt remains because of the subjective nature of questionnaire reporting and because there is insuffident knowledge to accurately identify and measure the pathogenic molds." "Upper respiratory symptoms and eye irritation were also associated with home dampness/mold in the present study, suggesting that an airborne irritant or allergen is involved in the pathogenesis." [26] "The Effect of Indoor Air Pollutants on Otitis Media and Asthma in Children," G.E. Daigler, S.J. Markello and [CM. Cummings, Laryngoscope 101: 293-296, 1991 [Vol. I, Issue 5, August 9, 1991] This case-control study of 125 children with a history of otitis media and 137 children with a history of asthma reported that maternal smoking was associated with childhood asthma (OR = 1.96), but that there was no association with otitis media. Associations between asthma and pets, asthma and humidifier use, and otitis media and use of wood-burning stoves were also reported. This case-control study investigated "the possible association between home environmental air pollutants and their effect on otitis media and asthma in children, in upstate NY in 1986-1987. Cases included 125 children with two or more doctor's visits for otitis media and 137 children either hospitalized or with two or more office visits for asthma. Controls (n = 237) were randomly selected from children who did not have an acute respira- tory illness during the study period. Data on exposure, family history, etc., were gathered by a questionnaire mailed to the parents. "Findings from this study confirm the previously reported association between childhood asthma and exposure to pets and maternal smoking." [maternal smoking, OR = 1.96, 95%0 CI = 1.10-3.47; pets, OR = 1.82, 95% CI = 1.05-3.15] "We were unable to confirm the assodation between smoking and otitis media... One explanation is that our study dealt with acute episodes of otitis media and the other studies addressed persistent middle ear effusions." "Study results suggest an association between the use of woodburning stoves and otitis media." [OR = 1.73, 95% CI = 1.03-2.89] "One possible reason wood stoves were more prevalent among the otitis group and not the asthmatics was that families with asthmatic children were counseled by the pediatric group to avoid using woodburning heat." "The use of hutrfidifiers in the asthmatic group is consistent with physidan advice to this group and probably not a predictor of disease [OR = 2.22, 95% CI = 1.29-3.82]." -39-

ETS AND IAQREFFRENCE JULY 1992 "Study findings revealed a greater likelihood of living in a new home (under 10 years old) among both case groups relative to controls, but this was not stafisticalJy signifi- cant .... Neither formaldehyde exposure, renovations (a surrogate measure for dust exposure), or gas cooking (a surrogate for NO2 exposure) were significant predictors of disease in this study." "Future studies should attempt to directly measure emissions to more accurately evaluate the impact of exposure on respiratory outcomes." [27] "The Influence of a Family History of Asthma and Parental Smoking on Airway Responsiveness in Early Inf-a.ncy," S. Young, P.N. Lesouef, G.C. Geelhoed, S.M. Stick, K.J. Turner and L.I. Landau, The New 2England Joumzal of Medicine 324:1168-1173, 1991 [Vol. I, Issue 5, August 9, 1991] In a prospective, longitudinal study, the presence and level of airway responsiveness and its relationship to a family history of asthrna or maternal smoking were investigated in 63 norms/inffmts. The authors reported that the level of responsiveness to histamMe in infants was related to the presence or absence of a family history of asthma and that airway responsiveness was increased in infants whose parents reported smoking during pregnancy. This is a prospective, longitudinal study "to determine the presence and level of airway responsiveness and its relationship to a family history of asthma or maternal smoking in 63 normal infants." "We wished to investigate whether airway responsive- hess could be detected in very early infancy. Therefore, in this study we assessed infants at a mean age of 4_ weeks, with some only 2 weeks old. A response to histamine was observed in all but 5 of the 63 infants. This finding indicates dearly that airway responsiveness is present very early in life, and it is not unreasonable to suggest that it may be present from birth." "We found that the level of responsiveness to histamine in infants was related to the presence or absence of a family history of asthma. This finding suggests that the initial level of airway responsiveness may be genetically determined." "We also found that airway responsiveness was increased in infants whose parents reported smoking during pregnancy .... Our study also demonstrates an assoda- tion between parental smoking and the level of airway responsiveness in early infancy, although we were unable to separate the effects of prenatal and postnatal exposure to dgarette smoke. The effect of continued postnatal exposure on the base-line level of responsiveness and on the subsequent development of the symptoms of* asthma is unknown. Moreover, we have not reported the amount of smoking, since it is widely recognized that the relation between the level of smoking reported by parents and the actual level of passive smoking by the fetus or infant is poor because of underreporting by parents, variations in ventilation in rooms and houses, and differences in the distance between the smoker and the infant." [28] "Risk Factors for Respiratory Syncytial Virus- Associated Lower Respiratory Illnesses in the First Year of Life," C.J. Holberg, A.L. Wright, F.D. Martinez, C.G. Ray, L.M. Taussig, M.D. Lebowitz and Group Health Medical Associates, Amer/can Jourmd of Ep idemiology 133(11): 1135-1151, 1991 [Vol. I, Issue 5, August 9, 1991] This paper presents conclusions from a prospective study of 1179 healthy infants in Tucson, AZ. The authors reported no association for maternal smoking and the inddenoe of respiratory syncytial virus-associated (RSV) lower respiratory illnesses (LRIs) at any age in the first year of life, although they did report an association between "likelihood of exposure to infection" and susceptibility to infection and RSV-LRIs. However, they also reported that their data supported a relation between maternal smoking, particularly > 20 cigarettes per day, and all LR/s. This paper presents conclusions from a study of 1179 healthy infants, enrolled at birth into the Tucson Children's Respiratory Study -- a prospective study of respiratory illness in children -- benveen May 1980 and January 1984, including illnesses observed until December 1985. "Respiratory syncy~ial virus (RSV) is a major cause of serious life-threatening illness of the lower respiratory tract in infancy, h is responsible for yearly epidemics, of ~pproxirnately 5-month duration, which are closely associated with an increase in the numbers of infants and young children hospitalized with lower respiratory tract illness (LRI)." "Maternal smoking was not related to the incidence of RSV-LRIs at any age in the first year of life." "A relation between maternal smoking, particularly > 20 cigarettes per day, and all LRIs (wheezing and nonwheezing) has been demonstrated in our study population for the first year of life. We have not exam- I I I I I I I I I I I I I I I I I

VOLUME I, ISSUES 1-24 JULY 1992 I I I I I I I I l I I I l I I I l ! • ined quantity of dgarettes smoked in this month-by- month analysis because of smaller numbers." "In summary.., a variety of factors appear to increase an infant's risk for... RSV-LRI .... : 1) likelihood of exposure to infection, including crowding and low socioeconomic status, day care, and birth month; and 2) host susceptibility to infection, including Imver cord serum RSV antibody levels, being male, and being mi_rfimalJy breast-fed .... Combinations of factors greatly increase the risk of having a RSV-LR.I. In addition, breast feeding appears to modify the risk of having a RSV-LRI and provides protection in conditions of lower sodoeconomic status where exposure may be greater." [29] "Relationship of Parental Smoking to Wheezing and Nonwheezing Lower Respiratory Tract Illnesses in Infancy," A.L. Wright, C. Holberg, F.D. Martinez, L.M. Taussig and Group Health Medical Associa- tion, The Journal of Pedia..,rics 118(2): 207-214, 1991 [Vol. I, Issue 2, May 17, 1991] Data from the Tuscon Children's Respiratory Study - a longitudinal study to investigate a variety of risk factors for acute and chronic respiratory illnesses in childhood. A tora2 of 1246 healthy infants and their families were initially enrolled at birth (May 1980-Oct 1984). This paper is restricted to those children who had at least one white, non-Hispanic parent and who remained under the care of a pediatrician throughout the first year of life (n = 847). Only acute lower respiratory tract illnesses (LRI) that were diagnosed by the pediatridans were included. Ascertainment of smoking status was done prospectively at time of enrollment in the study and on a questionnaire completed when the child was 15-20 months of age. Verification of smoking habits was also conducted by measuring cotinine in umbilical cord serum of 133 infants who underwent infant pulmonary function tests. "The LRI rate.., was significantly higher in infants whose mother smoked (OR = 1.52; 95% CA = 1.07- 2.15). The odds were higher if the mother smoked a pack of dgarettes or more per day and if the child stayed home rather than attending day care (OR = 2.8; 95% CI = 1.4- 5.5). Logistic regression indicated that the LRI rate was significantly elevated both in children exposed to heaW maternal smoke in the absence of day care, and in those who use day care but were not exposed to maternal smoking of a pack or more per day. These findings could not be attributed to other confounding variables. Neither paternal smoking nor smoking by other household members was consistently related to the LRI rate." "CotiMne was detectable in the umbilical cord sera of all infants whose mothers reported smoking during preg- nancy... Cotinine was also detected in 7 of 100 cord spedmens of infants whose mothers said they had not smoked during pregnancy." "We were unable to differentiate potential relationships between smoking during pregnancy from those of smoking after birth, because women who smoked during pregnancy tended to be the heaviest smokers and few stopped smoking when their child was born." Authors speculate that the reason why most of the relation- ships they observed for maternal smoking and LRIs were accounted for by women who were heavy smokers (more than a pack per day) may be because Tucson is a city ".. characterized by an outdoor life-style in both summer and winter and by wall-ventilated homes." Thus, ".. studies conducted in Tucson... have usually demonstrated more subtle rdationships daan has research conducted in areas where homes are better insulated and where people spend more time indoors." "Although some have reported that day care use is assod.- ated with increased LRI rates, this practice may be protective he the mother is a heavy smoker, perhaps because it reduces exposure to environmental tobacco smoke." Ovl-mt [ 1] =Breast Cancer, Cigarette Smoking, and Passive Smoking," A.J. Wells, AmetqcanJournal of Epidemiolo$O, 133(2): 208-210, 1991 and Letters to the Editor Regarding Same [Vol. I, Issue 22, May 22, 1992] This article proposed that data from studies by Hirayavna and Sandier, et al., support an elevated risk of breast cancer among women whose husbands smoke. Wells suggested that an upward adjustment for this presumed assodation is warranted in studies of active smoking and breast cancer incidence. The American Journal of Epidemiology recently published two letters concerning this article, which appeared in the journal prior to the first issue of this Report. The authors of the letters were Nathan Mantel and Wells, the author of the original article. The letters appear at American Journal of Epidemiology 135(6): 710-712, 1992. Mantel's letter recommended a downward adjustment in studies of active smoking and breast cancer. He proposed a hypothesis that "passive smoking cannot be more injurious than active smoking," which would lead to rejection of the claimed elevated breast cancer risk due to ETS. In his reply letter, Wells defended his original conclu- sion. -41-

ETS AND IAQREFEKENCE JULY 1992 "In recent papers on dgarette smoking and breast cancer.. • the reference category was women who had never smoked. Yet there is evidence, some published, some unpublished, that passive smoking among never smokers oould increase breast cancer risk. Thus, a better reference category for these studies may be women who never smoked and who have no reported exposure to environmental tobacco smoke." "Hirayarna has mentioned breast cancer as possibly assodated with passive smoking. For this communication, Dr. Hirayarna has provided detailed mortality data for never smoking women according to husband's smoking habit. The data show an devated risk of breast cancer in never smokers whose husbands smoked. Sandier et al. report an assodation between passive smoking and breast cancer. Data provided by Dr. Sandler also show an increased risk of breast cancer in never smoking women whose husbands smoked." "A reasonable estimate for the rdative risk of passive smoking breast cancer based on both the Hirayama and Sandler data sets would be about 1.4 for all ages combined.. •. The estimated relative risk of breast cancer for all never smokers compared with nonexposed never smokers would then be 1.3, and the inddence risk for those under age 50 would be about 1.8." "The results in the various papers on active smoking and breast cancer are at best comeusing. However, adjusting for a presumed assodation of passive smoking with breast cancer would devate probably all of the smoking relative risks m a levd greater than unity." "The intent of this letter is not to 'prove' an assodation between breast cancer and passive or direct smoking. Rather, it is to alert researchers in the breast cancer field to consider seriously, as a reference category in future studies, the use of never smokers who have had lit-de or no exposure to environmental tobacco smoke. Considering the long induction periods involved in breast cancer, it vcL!J be necessary to mace histories of passive exposure over long periods of time in order to get meaningfM data. Only by performing such investigations can we determine the real risk, if any, of breast cancer from various degrees of passive smoking and thus determine the real risk, if any, from direct smoking." [2] Letters to the Editor Regarding "Prenatal Exposure to Parents' Smoking and Childhood Cancer," F_,M. John, D.A. Savitz and D.P. Sandier, Ame~an Journal of Epiderniology 133: 123-132, 1991 [Vol. I, Issue 22, May 22, 1992] The American fournal of Epidemiology recendy published four letters concerning this artide, which appeared in the -42- .- journal prior to the first issue of this Report. The authors of the letters were: S. James Kilpatrick; Phi.kip Witorsch, Joseph M. Wu and Maurice E. LeVois; Peter N. Lee; and John, et al., the authors of the original arfide. The letters appear at American Journal afEpidemiok~ 135 (6): 712-715, 1992. • The article claimed to find elevated cancer risks for children whose fathers smoked during their mothers' pregnancies. None of the reported odds ratios (for cancers combined or for several spedfic cancers) was statistically significant. S. James Kilpatrick's letter stated that this study was a reanalysis of data from a study investigating electromag- netic fields. Moreover, he recommended the use of "modern" statistical modeling. Philip Witorsch, Joseph M. Wu and Maurice E. LeVois criticized the study's post hoc data analysis. They also commented on case-control differences and on the authors' interpretations of their data. Witorsch, et al., concluded by noting that the study failed to demonstrate "compelling" sratisticadly significant results. Peter N. Lee's letter focused on the authors' speculations concerning their reported results, given "the extreme weakness of the authors' case." He, too, commented on the lack of statistical significance, on case-control differ- ences and on inadequate control for possible confounders. In their reply letter, John, et al., called attention to the caveats in their paper and called for further research on this issue. [3] "Epidemiologic Evidence for the Role of Indoor Tobacco Smoke as an Initiator of Human Breast Carcinogenesis," A.W. Horton, Cancer Detection and Prevention 16(2): 119-127, 1992 [Vol. I, Issue 21, May 8, 1992] In this paper, the author statistically analyzes data on either cigarette consumption or male lung cancer mortal- ity and female breast cancer mortality data. He concludes that this analysis shows a correlation between trends in breast cancer inddence and trends in cigarette consump- lion or male lung cancer mortality (used as a surrogate for cigarette consumption). He concludes that these data show a role of exposure to "indoor tobacco smoke~ in a very early stage of breast cardnogenesis. "Extensive tabulations are now available of site-, sex-, and race-spedfic cancer mortality data for each U.S. state and country for 3 decades from 1950 to 1979. These data are analyzed for further evidence of the time relationship I I I I I I I I I I I I I °1 I I

I VOLUME I, ISSUES 1-24 JULY 1992 I I I I I I I I I I I I I I I ! ! I for highest correladori between dgarette consumption and the time of fatal outcome of the disease." "Since data on cigarette sales are not available on the county level, male lung cancer mortality rates provide a useful surrogate. It appears that in states having the highest breast cancer mortality in the 1970s, the highest correlation is found between 1970s county breast cancer and 1950s county male lung cancer rates. Data from the Northwest even permit us to demonstrate in the coundes having the highest educational levels the differing effects of smoking cessation on the risks of lung and breast cancers." "We utilize data on the incidence of breast cancer from the Connecticut Tumor Registry for the period 1934 to 1989 .... As noted above, the first major dip in cigarette sales after the 1930s depression occurred in the mid- 1950s. However, by the mid-1960s, sales both in Con- necticut and nationally (per capita) had exceeded the 1953 maxima." "The history of breast cancer inddence in Connecdcut from 1935 to 1989 follows almost in lock-step with cigarette consumption after a latent period of about 22 to 25 years. The 1930 peak in consumption is reflected by the 1953 peak in breast cancer inddence and the tripling in consumption from I933 to 1953 by the 50 percent increase in the incidence from 1955 to 1975, with a_n inflection between 1959 and 1963 corresponding to the 1938 recession. Even each brief interruption in dgarette consumption at the end of the two World Wars has its counterpart in brief drops in breast cancer inddence about 23 years later (1943 to 1946 and 1967)." "An alternate procedure.., for estimating the latent period involves the use of the temporal mid-points of the two steep rises in each curve. This option may be more satisfying from a theoretical standpoin~ dgarette consumption, 1915 to 1929, midpoint 1922; for 1933 to 1953, midpoint 1943; rise in cancer inddence, t945 to 1953, midpoint 1949; for 1956 to 1975, midpoint 1966. Estimated latent period for first rise, 27 years; for second rise, 23 years. Since the production of indoor tobacco smoke was almost three times as high (per capita) in 1943 as it was in 1922, it is reasonable that the average latent period for breast cardnogenesis in the later period would be shorter than in the first steep rise" "The relationship between average annual cigarette sales in 41 states and the District of Columbia in I950 to 1954 and mortality rates from breast cancer in 1979 to 1981 among white women, 35 to 74 years of age, is shown .... The R value is seen to have its highest value, 0.64, for sales in 1950 to 1954 and to drop in each subsequent interval. The correlation remains highly significant until the 1970 sales. Thus this gives us an estimate of the average latent period between initiation by indoor smoke and fatal outcome from cancer of at least 25 years, with a range of 10 to 30 years or more." "This research has suggested strongly that exposure to indoor tobacco smoke plays a role in a very early stage of the mechanism of this disease. Thus, a history of spousal smoking may not be relevant unless the marriage (and the smoking) has lasted more than a decade. Hence question- naires for case-control studies must glean information about exposure history dating back at least to puberty. Before the development of a significant mammary fat pad, localization of fat-soluble carcinogens in the breast area is unlikely. For the younger patients, exposure as teenagers may well be the most significant. Thus, smok- ing by parents or roommates must be documented." "Women in continental European countries have a much higher risk of breast cancer if male tobacco consumption and lung cancer mortality is high. This is the case even in those cudtures (e.g., the Netherlands and Switzerland) where women generally did not smoke until recently and female lung cancer mortality was still low in the 1970s. It could be logically concluded that breast tissue is more sensitive than lung tissue to indoor tobacco smoke." "Causal significance is a matter ofjudgrnent which goes beyond any statement of statistical probability. The criteria include the mnsistency, strength, specifidty, temporal relationship, and coherence of the assodation. In particular in this paper, the temporal relationships have been examined in the light of an extensive background in experimental chemical cardnogenesis .... [A]ll the data fit the sequence: increasing exposure to indoor tobacco smoke, then within 1 decade increasing rates of male lung cancer and finally, 2 to 3 decades after the spurts in dgarette sales (per capita), soaring incidence rates of breast cancer." [4] "Exposure-Response Relationships Between Woodworking, Smoking or Passive Smoking, Squamous Cell Neoplasms of the Maxillary Sinus," IC Fukuda and A. Shibata, Cancer Causes and Controll: 165-168, 1990 [Vol. I, Issue 15, January 31, 1992] This paper reports on a case-control study conducted in Japan. The authors reported that in men, squamous cell neoplasms of the maxillary sinus were related to a history of nasal disease other than trauma, occupational history of woodworking and past or current smoking habits. Among women, the authors reported that history of nasal disease and domestic "passive" smoking were independent risk factors for sinus cancer. In 1984, Hirayama reported a similar result from his large cohort study in Japan. - 43-

JULY 1992 '% case-control study of neoplasms of the maxillary sinuses (NMS) was performed in Hokkaido, Japan between 1982 and 1986. Preliminary findings based on 106 squamous cell (SC) cases of NMS and 212 community controls collected from 1982 to 1984 have been reported previously. They suggested that past histories of chronic sinusitis or nasal polyps, occupational history of woodworking, and smoking might be risk factors for SC-NMS in Hokkaldo, at least for males. By the end of 1986, additional new cases and controls had been collected by the same methods and the data as a whole were re-analyzed." "There were 258 cases: 180 men and 78 women. Two hundred and twenty-three cases (155 men and 68 women) had NMS, and 197 (142 men and 55 women) were classified histologically as SC-NMS. For this study, cases were limited to SC-NMS and to patients aged 40 to 79 years old, of which there were 175:129 men and 46 women. Information on exposures was obtained by mail questionnaires." "Stepwise regression [a statistical technique] of SC- NMS risk on potential risk factors suggested that history of nasal disease other than trauma, occupational history of woodworking, and past or current smoking habits, were independent risk factors for SC-NMS in men, and that history of nasal disease and domestic passive smoking were independent risk factors in women." "The number of smokers in the household also had a significant exposure-response relationship to the neo- plasms among women. The result was not changed when the analysis was restricted to female nonsmokers." [For one smoker in the household, an OR. of 1.40 was re- ported; for more than one smoker, an OR of 5.73 was reported. Statistical significance was not indicated.] "None of the previously published papers reported an exposure-response relationship between smoking and NMS except for one large cohort study which showed a significant risk elevation for neoplasms of the paranasal sinuses in nonsmoking married women according to the amount smoked by their husbands [Hirayama, 1984]. The exposure-response relationship observed in the present study between the number of smokers in the household and SC-NMS in nonsmoking women might be regarded as consistent with that finding and both seem to suggest an effect of passive smoking on this tumor. The trend of the ORs by level of active smoking for men strongly suggests an effect of active smoking on develop- ment of SC-NMS, at least among men." [5] "Parental Occupation and Other Environmental Factors in the Etiology of Leukemias and Non- Hodgkin's Lymphomas in Childhood: A Case- Control Study," C. Magnani, G. Pastore, L. L~_~tto and B. Terracini, Tumori76: 413-419, 1990 [Vol. I, Isstm 7, September 13, 1991] This case-control study examined parental smoking in children with two forms of leukemia and with non- Hodgkin's lymphoma. No association was reported for parental smoking and acute lymphocytic leukemia; however, elevated odds ratios were reported for non- Hodgkin's lymphoma in children whose parents smoked. This hospita/-based case-control study on acute lympho- cytic leukerrfia (ALL), acute non-lymphocytic leukemia (AnLL), and non-Hodgkin's lymphoma (NHL) in childhood was conducted in Turin, Italy from 1981- 1984. Cases included 142 children withALL, 22 with AnLL, and t 9 with NHL matched with 307 controls randomly selected from among children hospitalized at the same hospital. "Parental smoking habits up to the child's birth are described... For ALL, no association was observed, and the ORs were not dose-related. Findings did not differ when smoking during the interval between birth and diagnosis or during the year before birth were considered, nor when other indicators were considered, like years of smoking, age at start or cumulative dgarette consumption (data not presented). The exclusion of 40 control children affected by respiratory diseases did not modify the ORs." "ORs for NHL were high for both maternal and paternal smoking. Before the birth of the index child, 4 case mothers and 13 control mothers smoked more than 16 cigarettes per day (crude OR, 5.8; 95% CI, 1.9-17.5; not modified by excluding subjects whose mother was not present at the interview). Among fathers, the ORs for smoking 1-15 and 16+ cigarettes per day were respectively 6.4 (95% CI, 1.0-45.5, based on 6 cases and 84 controls) and 5.6 (95% CI, 0.9-35.7, 11 cases and 136 controls)." "The role of passive exposure to smoking in childhood leukemias or lymphomas has been investigated by a limited number of studies, which mainly reported a negative or indicated a weak assodation. Stjernfeld et al. estimated ORS between 1.5 and 2 for maternal smoking and childhood leukemias and NHL, in a study the design of which has been criticized because of the choice of the control group." I I I I I I I ! I I i

I VOLUME [. [~UF_q 1-24 .JULY 1992 I I I I I I I I I I I I I I I I I I OTHER HFALTH ISSUES [ 1 ] "Childhood Exposure to Environmental Tobacco Smoke and the Risk of Ulcerative Colitis," R.S. Sandier, D.P. Sandier, C.W. McDonnell and J.I. Wurzelmann, American Journal of Epidemiology 135(6): 603-608, 1992 [Vol. I, Issue 22, May 22, 1992] A decreased risk of ulcerative colitis, a chronic inflm-n- matory bowel disorder of unknown cause, has been reported among dgarette smokers. This paper reported on a case-control study of ulcerative colitis patients that included questions about parental smoking during childhood. Reportedly, individuals who responded that their parents smoked had a decreased risk of ulcerative colitis in adulthood. "Ulcerative colitis is a chronic inflammatory bowel disorder of unknown cause. There are striking geographic differences in incidence .... Variations in incidence with time, place, and migration have suggested an important environmental etiology." "The environmental exposure that has received the most attention is dgarette smoke, based on the unexpected finding.., that smoking seemed to protect against ulcerative colitis." "The health consequences of 'passive smoking,' or inhaling environmental tobacco smoke, have received considerable attention .... Recent reports suggest that passive smoking in childhood increases the risk of cancer in adulthood .... If dgarette smoking protects against ulcerative colitis, and if childhood exposure to environmental tobacco smoke can influence adult disease risk, then childhood passive smoking might decrease the adult risk of ulcerative colitis. The authors examined this hypothesis as part of a larger study of risk factors for inflammatory bowel disease." "Names of potential study cases were obtained from rosters of the three North Carolina chapters of the Crohn's & Colitis Foundation of America... a self-help organization dedicated to finding the cause of and cure for inflammatory bowel disease." "To identify controls, we asked cases to provide the name of their geographically closest neighbor who was of similar sex, race, and age (within 5 years). If no neighbor met the matching criteria, a colleague or friend could be substituted." "To assess childhood exposure to environmental tobacco smoke, we asked each subject whether his or her mother or father had smoked in the home on a regular basis when the subject was younger than age 15 years. Subjects who had had one or more parents who smoked were consid- ered passive smokers." "The risk of ulcerative colitis among active smokers with nonsmoking parents was significantly decreased (odds ratio (OR) = 0.34, 95 percent confidence interval (CI) 0.13-0.92). The risk was somewhat higher for active smokers with parents who had smoked (OR = 0.72, 95 percent CI 0.32-1.63). The risk was significantly de- creased for passive smokers, i.e., nonsmokers whose parents had smoked (OR = 0.50, 95 percent CI 0.25- 1.00)." "The present study supports the previously reported deficit of smokers among patients with ulcerative colitis. Moreover, childhood exposure to environmental tobacco smoke significantly decreased the adult risk of ulcerative colitis. The influence of early life factors on adult risk may have important implications in understanding the etiology of ulcerative colitis." "The observed decreased risk could be assodated with environmental tobacco smoke during childhood but coudd also be associated with genetic or transpiacental effects of dgarerte smoke exposure. Similarly, the study questionnaire did not provide sufficient detail with regard to marital history" to allow accurate assessment of exposure to a spouse's smoking. While the results seem to suggest a role for early llfe exposure to passive smoking, the possibility remains that effects are due to passive smoking in general, regardless of age at exposure. The relevant biologic mechanisms are also unknown." "The important conclusion from rafts study is not so much that passive smoking in chi/dhood protects against ulcerative colitis as that childhood factors can influence adult suscepti- bility. Future research should explore other early life factors that might be related to ulcerative cofitis in adulthood." [2] "Demographic and Socioeconomic Differences in Beliefi~ about the Health Effects of Smoking," tLC~ Brownson, J. Jackson-Thompson, J.C. Wilkerson, J.R. Davis, N.W. Owens and E.B. Fisher, Amer/ca~ Journal of Public Health 82(1): 99-103, 1992 [Vol. I, Issue 22, May 22, 1992] This study reported on a telephone survey of 2,092 adults in inner-dry areas of St. Louis and Kansas City, Missouri. Reportedly, 78 percent of those surveyed believed "passive smoke exposure" was "harmKfl" to nonsmokers. "The harmful health effects of dgarette smoking are well documented." -45-

ETS AND IAQREFEP,_ENCE. JULY 1992 "The health hazards of passive smoke exposure have only recently been established. Passive smoking is now regarded as a cause of lung cancer in healthy nonsmokers, accounting for approximately 3800 lung cancer deaths per year in the United States. In addition, the children of parents who smoke have a higher frequency of respiratory infec-tions and decreased lung function as the lungs mature." "Public know/edge and beliefs about the health risks of smoking and passive smoke exposure have increased subsra.ntially over the past several decades .... A nation- wide study in 1986 found that 81 percent of respondents knew that passive smoke exposure was hazardous to nonsmokers' health." "This paper presents information on knowledge and beliefs about dgarette smoking and passive smoke exposure that was collected in a recent survey of an urban adult population." "The present study was conducted to provide planning and evaluation data for a smoking prevention and cessation intervention in a high-risk, inner-dEE popula- tion .... [T~elephone interviews were conducted with 2092 adults 18 years of age and older. Study subjects were randomly selected from among residents of approximately 60 census tracts in St. Louis and Kansas City, Mo." "The vast majority of respondents (88.4 percent) believed that smoking is harmful to headth. Belief in harmful health effects was inversely related to age and positively correlated with education level. Current smokers were significantly less likely than never smokers to acknowledge the health effects of smoking." "A majority of respondents believed passive smoke exposure to be harmful to nonsmokers' health (78.2 percent), harmful to young children (87.2 percent), and annoying (65.5 percent). Results regarding health effects and annoyance of passive smoking were inversely corre- lated with age. Blacks were more likely than Whites to agree that passive smoke exposure is harmful to young children's health (OR = 1.3) and to find passive smoke annoying (OR = 1.2). Current smokers were significantly less likely than never smokers to acknowledge the effects of passive smoking on nonsmokers' health (OR = 0.3) and children's health (OR = 0.2) and were less often annoyed by passive smoke (OR = 0.1)." "[W]e found knowledge about the health effects of smoking and passive smoking to be lower among older age groups, women, less-educated respondents, and current smokers. Racial differences were interesting in that Blacks were generally less likely to appredate the -46- health effects of active smoking but were more likely to acknowledge passive smoking's health effects." "Understanding of the health effects of passive smoking was relatively high in our study. Overall, 78 percent of those surveyed indicated that passive smoke exposure was harmful to nonsmokers' health. This compares with 81 percent of those surveyed in two recent national studies. The majority of respondents acknowledged the detrimen- tad effects of passive smoke exposure on children. Among subgroups, knowledge was lower in older age groups, women, Whites, and former or current smokers. Most of those surveyed were annoyed by passive smoke exposure, as reported earlier. Data on passive smoking annoyance affirm the declining sodal acceptability of smoking in public places. Nearly 40 percent of the current smokers we surveyed were annoyed by others' dgarette smoke." "Our findings may not be entirely representative of the general population, since we surveyed a low-income population. Thirty percent of the respondents in our study had household incomes of less than $10 000 per year, which compares vdth 22 percent of the overall state population." "The patterns that we identified should be considered when targeting public education and advocacy efforts to the segments of the population at highest risk for smok- ing and passive smoke exposure." [3] Letters to the Editor Regarding "Group Day Care and the Risk of Serious Infectious Illnesses," A.T. Berg, E.D. Shapiro and L.A. Capobianco, Amer/- can Journal of Epldemlology 133(2): 154-163, 1991 [Vol. I, Issue 22, May 22, 1992] The American Journal of Epidemiology recently published two letters concerning this article, which appeared in the journal prior to the first issue of this Report. The authors of the letters were: Philip Witorsch, Joseph M. Wu and Maurice E. LeVois; and Berg~ et al., the authors of the original article. The letters appear at American Journal of Epidemio/og7 135(6): 715-717, 1992. The article reported on a study of 193 case-control pairs Of children hospitalized for serious infectious illness. A statistically nonsignificant risk for group day care atten- dance was reported. The authors also claimed that "passive smoking" was associated with a risk of 3.96 (95 percent CI 2.16, 7.24) for serious infectious illness. In their letter, Phitip Witorsch, Joseph M. Wu and Maurice E. LeVois discussed several problems they atxribute to the study. In particular, they noted that the study was not designed to investigate parental smokin~ I I I I I I I I I I I I I I I I I l

VOLUME I, ISSUES 1-24 JULY 1992 I I I I I I I I I I I I I I I I I I I " such an "after-the-fact" analysis as was conducted could have potentially biased the conclusions. Moreover, the letter commented on inadequate attention to confound- ers, case-control differences and lack of "dose-response" information. Berg, et al., the authors of the study, replied, stating that their parental smoking analysis was not post hoc, but rather, had been "planned from the start." They also commented on the diffculties of controlling for numer- ous potential confounders and stated that case-control differences were accounted for via a statistical model. Finally, Berg, et al., noted that this issue is controversial because of "the large financial interests involved," saying that it "must be addressed dispassionately." [4] "Maternal Smoking and Neuropsychological Development in Childhood: A Review of the Evidence," S. Tong and A.J. McMichael, Develop- mental Medicine and ChiM Neurology 34: 191- 197, 1992 [Vol. I, Issue 21, May 8, 1992] This article reviews the available sdentific literature concerning maternal smoking and children's neurological and psychological devdopment. The authors comment on several methodological issues related to the studies they review and conclude that the existing data are "inconclusive." "Recently there has also been interest in the neuropsychological consequences of maternal smoking. Most studies have sought empirical evidence for an association between exposure to maternal smoking and neuropsychological development in childhood. Since the possible long-term effects of maternal smoking on children's neuropsychological development are of both theoretical and practical importance, the question arises as to whether the observed impairment of neuropsychological functioning should be attributed to exposure to maternal smoking or to co-existent genetic factors and other sodo-environmental influences." "This paper reviews the contemporary epidemiological and toxicological evidence on the effects of exposure to maternal smoking and neumpsychological development in childhood, and discusses some strategies which need to be considered in future studies." "The epidemiological research on the neuropsychological effects of maternal smoking consists mainly of six cohort studies." "In the above-mentioned studies, exposure was most frequently estimated by questionnaires. Typically, the exposure measure has been the number of parents who smoked and/or the number of dgarettes smoked by the parents." "Moreover, the questionnaire approach has several potential limitations, although it may be an effective and simple means of obtaining exposure information. First, the questionnaire-based measures of ETS exposure generally have not been standardized, and may have limited validity and reliability; random imprecision or misdassification will bias the estimated effect (relation- ship) toward the null value. Second, recall bias or report- ing bias may influence the validity of the information. Third, an accurate reconstruction of integrated exposure is difficult to achieve from the reporting, because factors other than maternal smoking may influence in utero and posmatal exposure to ETS... Finally, the questionnaire approach to assess exposure to ETS provides only a surrogate of dose." "Some of the previous studies did not use standard neuropsychological rests to estimate developmental : outcome. Furthermore, only a few studies have taken maternal intelligence, quality of home environment and posmatal exposure to ETS into account .... The ever- present problems of bias and confounding are a recurrent threat to the validity of causal inference in observational (e.g. cohort) epidemiological studies; they are of particu- lar concern when the actual effect being studied is of small magnitude." "[S]ome methodological strategies need to be considered in future studies." "First, a suffciently valid and precise measure of exposure is important to the study of dose-response relationship of maternal smoking and neuropsychological development." "Second, a comprehensive and accurate assessment of the child's ability is another important methodological issue.~ "While neuropsychological deficits may be caused by exposure to maternal smoking during pregnancy and/or after birth, the evidence is inconclusive so far. The effects of maternal smoking on neuropsychological development of the offspring found in some studies are so mild that they could easily be attributable to uncontrolled for (or unknown) sodal and environmental factors. In order to gain a dearer understanding of this area, a more predse measure of exposure to tobacco smoke (both in utero and posmatally) and comprehensive assessment of children's ability will be required in future studies." - 47-

ETS AIqD ~.Q. P,.E~EP,_ENCE JULY 1992 I [5] "Passive Smoking and Otitis Media with Effu- sion," G.S. Barr and A.P. Coatesworth, British Medical Journal303: 1032-1033, 1991 [Vol. I, Issue 13, January 3, 1992] Otitis media with effusion, a disorder of the middle ear, has reportedly been associated with parental smoking in some studies. This study, however, reports that the smoking habits of parents of children with otitis media and of parents of children without otitis media did not differ. The authors note the confounding effect between cigarette smoking and sodoeconomic status, and con- dude that smoking is unlikely to be a risk factor for otitis media, although an assodation may be reported. "In all, 115 children (70 boys, 45 girls; age range 17 months to 11 years 6 months, median 5 years 5 months) from the Cheltenharn and Gloucester areas who had otitis media with effusion confirmed by myringotomy were matched according to age (within six months), sex, race, and social class to a control group of healthy children attending the ophthalmology and orthopaedic clinics. The children with otitis media with effusion had hearing loss for at least three months and had been assessed by otoscopy, tympanometry, and audiometry. The control group had no history of ear problems and normal results on otoscopy and tympanometry..." "Data were compared.., for the presence of at least one adult in the household who smoked and whether the mother smoked." "Parental smoking habits in the two groups were the same. There were no differences between the median number of cigarettes smoked in the two groups by mothers alone and by all adults in the household." -. "The prevalence of otitis media with effusion is highest where sodal conditions are poor, and children of non- manual workers have significantly better hearing than do those of manual workers. Cigarette smoking is commoner in those from the poorer socioeconomic classes but it is unlikely to be a risk factor for otitis media effusion, although it may have an assodation." [6] "Influence of Paternal Age, Smoking, and Alcohol Consumption on Congenital Anomalies," Dak. Savitz, P.J. Schwingl and M.A. Keels, Teratology 44: 429-440, 1991 [Vol. I, Issue 13, January 3, 1992] The authors of thls study analyzed data collected during the Kaiser Permanente Child Health and Development Studies on chi]dren born between 1959 and 1966. They reported statistical associations between paternal age over 35 or paternal smoking and a number ofbkth defecm. How- ever, the authors wrote that no "strong or widespread" assodations were repotted. The authors also noted thdr study's impredsion (i.e., in exposure assessment) and the possibility of numerous confounding factors. EXCERPTS,' "[T]he paternal role in the etiology ofbkth defects is of interest and potential public health importance, and mechanisms of effec~ can be postulated. However, the current epiderniologic data are severdy limited. Data from the Child Health and Devdopment Studies are well-suited to an analysis of parernal exposures and birth defects, including defects not immediately identified at birth. This large, broadly representative population of pregnant women was followed prospectively from the initiation of prenatal care, with birth defects carefully evaluated and reported through early childhood. Data were examined concerning paternal age and two prevalent potential reproductive toxins, tobacco smoke and alcohol." "Single live births from 1959 to 1966 among 14,685 Kaiser Foundation Health Plan members who partidpated in the Child Health and Development Studies were analyzed to assess the impact of paternal age, dgarette smoking, and alcohol consumption on the occurrence of birth defecr.s in the offspring. Prevalence odds ratios for anomalies identified by age 5 were analyzed, contrasting exposed to unexposed fathers with adjustment for maternal age, race, education, smoking, and alcohol use_ Advanced paternal age was assodared with increased risk ofpreauricular cyst, nasal aplasia, deft palate, hydrocephalus, pulmonic stenosis, urethral stenosis, and hemangiorna." "Several of those anomalies assodated with older fathers were also increased among fathers who smoked. Three of the adjusted prevalence odds ratios were 2.0 or greater (hydrocephalus [POR = 2.4], ventricular septal defect [POR = 2.0], and urethral stenosis [POR = 2.0] and one was between 1.5 and 2.0 (deft lip +__ deft palate [POR = 1.7]). Cleft lip + deft palate was concentrated among offspring of heavier smokers (POR = 1.9) as was urethral stenosis (POR = 2.4). In contrast, the risk of hydrocephalus and ventricular septal defects was slightly greater among offspring of lighter smokers." "Notable reductions in risk with paternal smoking (crude or adjusted POR<0.7) were found for neural tube defects, patent ductus arteriosus, pyloric stenosis, incom- petent ureterovesical valves, clubfoot, and polydactyly." "These data generally do not indicate strong or wide- spread associations between paternal attributes and birt~ I I I I I I I I I I I I I I

V(BLUMV. I, ISSUES 1-24 JULY 1992 I I I I I I I I I I I I defects. However, because of this study's imprecision, limited ability to isolate defects most likely to be of paternal origin, and the identification of several suggestive associations with age and smoking, further study of this issue would be of value." [7] ~Multiple Chemical Sensitivity," B. Hileman, Chemicaland Engineering Newt pp. 26-42, July 22, 1991 [Vol. I, Issue 7, September 13, 1991] This review discusses a controversial illness, multiple chemical sensitivity, sometimes claimed to be triggered by indoor air components, including ETS. Excerpts: This is an extensive review of what is known and unknown about multiple chemical sensitivity (MCS) - "this enigmatic syndrome." "There is a lack of a dear definition of what it is. There is antagonistic and divisive infighting among several medical specialties as they attempt to deal with a class of patients with MCS-like symptoms while defending their professional turf against competing specialties. Plaintiffs and defense lawyers make extreme claims as alleged victims of MCS seek judicial relief..." "[B]ills currently being considered in both Senate and House on indoor air quality identify MCS as one of the health consequences of indoor air pollution. Those experts who believe that scientific studies have not yet shown MCS to be a true illness are quite alarmed by the government's willingness to legitimize it." "To many traditional allergists and psychiatrists, the onset of MCS is due almost entirely to psychological factors .... Other physidans -- and these are likely to be clinical ecologists and some traditional physicians who practice occupational medicine --believe that chemical exposures are causing these people to be ill." "The number of chemicals the victims respond to is equally wide... [the list includes] passive dgarette smoke • .. But in some of these complex mixtures it is hard to tell what, if anything, the patient is responding to. The majority.., are sensitive to more than one substance, but some react to only one, and the same substance can cause different symptoms in different individuals." "Mary Lamielle, founder and president of the patient advocacy group.., calls for making the current ventila- tion recommendations of... [ASHRAE]... a minimum national mandatory standard for all public and commer- cial buildings. She would also prohibit smoking in all public buildings." The review concludes with a review of research priorities and strategies. This section includes statements attributed to Robert Axdrad of EPA, the Chemical Manufacturers Assodation and the Business Coundl on Indoor Air. [8] "Passlve Smoking by Pregnant Women and Fetal Growth," H. Ogawa, S. Tominaga, K. Hori, K. Noguchi, I. Kanou and M. Matsubara, Journal of ~Epidemiology and Community Health 45: 164-168, 1991 [Vol. I, Issue 6, August 27, 1991] Birth weight and growth retardation were examined in babies born to 6,831 Japanese women who had provided information on ETS exposure during pregnancy. The authors reported a statistically significant decrease in birth weight associated with ETS exposure, but the association disappeared when confounding factors were considered. Nonetheless, the authors speculated that "heavy exposure does induce a reduction in fetal growth" and concluded that "passive smoking in pregnant women is an important public health issue." EXCERPTS= This study consisted of an interview and clinical survey of pregnant women in Aichi Prefecture, Japan. Of the women interviewed, 6,831 delivered a llve baby without malforma- tion and were induded in the study. Bkrth wright and the prevalence of growth retardation were examined. "The rate of passive exposure to cigarette smoke was 62.2% for at least 1 min per day, and 35.3% for at least 1 h per day... Mean exposure time for the exposed women was 3.1 h per day .... About 30% of women whose husbands smoked during pregnancy were not exposed at all to dgarette smoke at home, and about 13% of women whose husbands never smoked were exposed at home." "[W]omen who were exposed to smoke tended to be young and nulliparous, to drink alcohol, to be employed, and to having smoking husbands, compared with those who were not exposed .... [T]hese characteristics were included as possible confounding factors..." "There was a small but statistically significant decrease in birth weight from passive smoke exposure in mothers who had never smoked, while no significant change was observed in smoking mothers who stopped or continued to smoke... "The crude effect of passive smoke exposure on birth weight reduction was found to be statistically significant, but the effect was diminated by adjustment for confounding factors. Multivariate analysis ofbkth weight showed significant effects of parity and age. These factor~ were also significantly associated with passive smoke exposure. - 49-

ETS AND LAQREFERENCE JULY 1992 Therefore the observed crude effect can probably be explained by the younger age at birth and the higher proportion of first deliveries for passively exposed mothers." "[A]n estimation of passive exposure using only paternal smoking involves a sizable miscalculation. Furthermore, about one-third of the total passive exposure time was at the workplace. These results suggest that the paternal smoking habit may be an inadequate index of maternal passive exposure to cigarette smoke." "Although this study did not show a significant effect of passive exposure on fetal growth, it is expected from our results that heavy exposure does induce a reduction in fetal growth, just as active smoking does. Therefore passive smoking of pregnant women is an important public health issue, especially in Japan where the preva- lence of cigarette smokers among young adults is high." [9] "Risk Factors for Birth Defects: Data from the Adanta Birth Defects Case-Control Study," J.D. Erickson, Teratology43: 41-51, 1991 [Vol. I, Issue 5, August 9, 1991] Based on a population-based case-control study of 4,900 babies born with major birth defects and 3,000 babies born without defects, the author reported an assodation of maternal smoking (unadjusted OR = 1.12, 1696 cases) and paternal smoking (OR = 1.12, 1890 cases) with all birth defects combined. "The Atlanta Birth Defects Case-Control Study data comprises information obtained from interviews with parents of 4,900 babies born with major birth defects and with the parents of 3,000 babies born without defects. The source of cases is the Centers for Disease Control's Metropolitan Atlanta Congenital defects Program; the case-control study is population based. Birth defects are classified into 92 groups and cross-tabulated by 105 exposure/risk factor variables." The study reported associations of maternal smoking [OR = 1.12, 95% CI 1.0-1.2 (1696 exposed cases)] and paternal smoking with all birth defects combined [OR = 1.12, 95% CI 1.0-1.2 (1890 cases)] with all birth defects combined, based on the unadjusted data. [10] "A Comparison of Active and Passive Smoking During Pregnancy: Long-Term Effect," J. Maldn, Pall Fried and B. Watkinson, Neurotoxicology and Teratology 13: 5-12, 1991 [Vol. I, Issue 4, July 12, 1991] In this ongoing, prospective, longitudinal study of the development of children, it was reported that maternal -50- • exposure to ETS is assodated with long-term negative effects that are similar, but milder in degree, to the long- term effecv.s of maternal active smoking. Subjects were "from the Ottawa Prenard Prospective Study (OPPS), an ongoing, prospective, longitudinal study of the devdopment ofchi/dren born to mothers who used 'soft' drugs (i.e., marihuana, dgarerr.es, and/or alcohol) during pregnancy." Ninety-one children between the ages of six and nine years were divided into three groups - active smoking, passive smoking, and nonsmoking - according to the status of thdr mothers during pregnancy and were tested in a batxery of neurobehavioral tests. "This study indicates, for the first time, that maternal passive exposure to dgarettes is assodated with long-term negative effects that are similar to but milder in degree to the long-term effects of maternal active smoking." "[C]hildren of nonsmoking mothers generally were found to perform better than the two smoking groups on tests of speech and language skills, intelligence, visual/ spatial abilities and on the mother's rating of behavior." "Of the 19 potential confounds considered, the following four variables were associated with smoking status... : examiner, maternal education, maternal age and SES [sodoeconomic status] .... These four potential confounds were not significantly related dr_her individually or as a group to the outcome variables. "In considering the results of the passively exposed group a number of factors must be recalled. For almost half of the nonsmoking women, involuntary smoking occurred outside the home, i.e., paternal smoking was not a factor. This is important as all of the previous studies examining the rdationship between passive smoking and pregnancy outcomes, with exception of Martin and Bracken, have used the father's smoking as the only index of secondhand smoke. Not only does this approach disregard sources of secondary smoke for a large number of women but it also creates problems in terms of the potential comparison groups who may well be exposed to sources of smoke during pregnancy that are unknown to the researcher." "In the present work, it was statistically impossible to isolate postnatal involuntary smoking by the children of women who smoked during pregnancy as 97 percent of these children were exposed after birth to secondhand smoke. Thus the pronounced effects found in the offspring of active smokers may be due to the combination of in urero exposure and postnatal secondary smoke. The role of I I I I I I I I I I I I I I I I I I

VOLUME I, ISSUES 1-24 JULY 1992 i I I I I I I I I i ! I I I I semndhand smoke among these children remains to be determined and will need studies with a large number of subjects in order to parcel out this variable_" '1 [I 1] "Effect of Passive Smoking During Pregnancy on Selected Perinatal Parameters," F. I_az~roni, S. Bonassi, E. Manniello, L. Morcaldi, E. Repetto, A. Ruocco, A. Calvi and G. Cotellessa, International Journal of Epidemiology 19(4): 960-966, 1990 [Vol. I, Issue 2, May 17, 1991] Prospective study of 1004 deliveries, in 11 Italian cities, to investigate the effect of passive smoking (PS) during preg- nancy on various perinatal parameters. PS exposure was determined by questionnaires which asked for hours of exposure at home and at work (-25% percent (sic) of study subjects were exposed to PS). "A mean reduction of 16g (p < 0.07) in birthweight and a decrease in birth length of 0.05 cm (p < 0.08) were found for each hour of antenatal passive smoke exposure. No or slight effects were reported for other parameters mnsidered." "The findings of this study suggest a possible effect of maternal PS on birthweight, although both.., analyses failed to achieve statistical significance. A new hypothesis, suggest- ing a decrease in the mean length at birth following mammal PS was revealed as well." "IT]he present study showed a mean duration of daily exposure to passive smoking almost double at work (5.02 hours [hours-minutes]) compared to domestic PS (2.41 hours). These data confirm the evidence of an underesti- mated exposure for women working outside the home during pregnancy, when the parmer's smoking habits only are considered as the source of exposure to sidestream smoke." "A more precise definition of exposure, including the choice of the most appropriate marker for PS monitoring, laa'ge sized studies and consideration of a large number of potential confounding factors are needed for conclusive studies on this topic." [12] "Editorial Comn~ntary: New Effects of Active and Passive Smoking on Reproduction," J.M. Samet, American Journal of Epidemiology 133(4): 348-350, 1991 [Vol. I, Issue 1, April 30, 1991] The studies included in this issue address reproductive consequences of smoking that have not yet been exten- sively investigated. Stergachis et al. conducted a case- control study of tubal pregnancy and found an increased risk (relative risk (RR) = 1.4, 95% confidence interval (CI) 1.0-2.0) of tubal pregnancy in current smokers compared with never smokers. Ahlborg and Bodin conducted a prospective cohort study of active and passive smoking and feral death, preterm birth, and low birth weight. The study confirmed that active smoking increased the risk of preterm birth and low birth weight. For nonsmoking women who worked during pregnancy, passive smoking at work in- creased the risk of fetal death (RR. = 1.53, 95% CI 0.98- 2.38); the extent of the increased risk was greater than that for active smoking (RR = 1.30, 95% CI 0.85-1.96, for active smokers of>_ 20 dgarettedday). "Both studies are illustrative of the merahodological problems that may arise in investigating weaker assodations of smoking and disease." "The investigation of Stergaehis et al. illustrates the difficulty of fully disentangling the correlated dimensions of human behavior that determine disease risk." "It is unlikdy that confounding by life-style correlates of smoking can be fi.fl/y contro/led by simply including indicator variables in multivariate models. Such concern about confounding has led to a reluctance to judge dgarette smoking to be causally assodated with cervical cancer, in spite of numerous studies showing this assodation." "The findings of Ahlborg and Bodin need careaCul interpre- tation but not immediate replication as suggested by these authors." "lilt seems biologically incorrect to postulate effects of passive smoking on reproduction that have not been demons'mated for active smoking. The evidence on passive smoking in this study also lacks internal consistency. Passive smoking was not associated with low birth wright, in spite of its association with early fetal death." "The relative importance of the home and work environ- ments for working women has not been determined, although recent data from New Mexico showed higher levels of nicotine in personal samples obtained in the workplace than in living room samples taken in homes. Thus, the finding of an adverse effect of workplace exposure but none of home exposure on early feral death may be consistent with the rdative contributions of exposure in these two environ- menu; however, comparable measurements are needed from Sweden." [13] "Chronic Fetal Hypoxla and Sudden Infant Death Syndrome: Interaction Between Maternal Smoking and Low Hematocrlt During Pregnancy," M.G. Bulterys, S. Greenland and J.F. Kmus, 86(4): 535-540, 1990 [-Vol. I, Issue 1, April 30, 1991] This study is based on data collected from the U.S. Collaborative Perinatal Project - a prospective, multicenter -51-

JULY 1992 study of 56,000 pregnandes enrolled between 1959 and 1966. The hypothesis that chronic feral hypoxia contributes to sudden infant death syndrome (SIDS) through a possible interaction between the effect of maternal dgarerte smoking and low hematocrit during pregnancy was analyzed by comparing 139 identified SIDS cases with 1930 controls randomly sdexxed from infants who survived the first year of /ire. AFter adjustment for matema/age, infants born to mothers who smoked 10 or more dgarettes per day and who were anemic (hemaczit less than 30%) during pregnancy were at much higher risk of SIDS than infants born to mothers who did not smoke and were not anemic (odds ratio = 4.0; 95% confidence limits, 2.1 and 7.4). "[W]e found that after controRing for rdevant confounders [including maternal education, family income, number of prenatal visits] maternal hematocrit did not affect the risk of SIDS among infants born to nonsmoking mothers." "Few investigators have been able to distinguish between the effects ofsmoldng during pregnancy and passive inhalation of smoke during posmatal life. Respiratory morbidity in infancy appears independently assodated with both maternal smoking during pregnancy and, to a lesser extent, passive infant smoking after birth in infancy. It is possible that SIDS deaths are related more to passive smoking than to smoking during pregnancy; this issue deserves further research." ETS ExeosuR~ aND MONITOmNG [1] "The Personal, Indoor and Outdoor Concentra- tions of RSP and Nicotine Measured in Six Smoker's Families in Taiwan," C.C. Chart, Huang, Y.C. Chen and J.D. Wang, Presented at the American Industrial Hygiene Association/ American Congress of Governmental Industrial Hygienists Conference, June 3, 1992, Boston,/VIA [Issue 23, Item 29] This abstract reports on personal monitoring for RSP and nicotine of six smokers' families. Reportedly, RSP concentrations were higher indoors than outdoors, and smokers were exposed to more RSP than were their nonsmoking relatives. Both RSP and nicotine concentra- tions were positively correlated with the number of cigarettes smoked at home, and according to ume-activity data, relatives of smokers were most frequently exposed to "passive smoking" between 7 and 11 p.m. "Studies have documented that passive smoking has strong impact on human health in developed countries. -52- In addition to workplace, people were usually exposed to passive smoking in their residence. The problem of active and passive smoking has potentially worsened in Taiwan since the government a/lowed tobacco import in 1987. This study examines [the] smoker's impact on his indoor air quality and his rdatlve's [sic] exposure to respirable suspended particulates (RSP) and nicotine. The daily indoor/outdoor RSP and nicotine concentrations of six smokers' homes were concurrently measured over a week in a rural area of Taiwan in 1991. Personal exposure to RSP and nicotine of fourteen members from these six families were [sic] also measured. Indoor RSP concentra- tions were higher than outdoor. The RSP smokers were exposed to was about 33.6 ug/m~ higher than the non- smoking relatives. Indoor nicotine concentrations of six homes were averaged at 6.7 + 5.8 ug/m~ and personal exposure were [sic] about 4.6 + 4.6 ug/mt The RSP concentration difference between indoor and outdoor was positively correlated with the ratio of number of cigarettes smoked at homes [sic] divided by home volumes. The indoor nicotine concentrations were also positively correlated with the number of cigarettes smoked at homes [sic]. The time-activity data of 14 persons indicate that 83% of their time were [sic] spent indoors. Relatives of smokers were most frequently exposed to passive smoking between 7 to 11 pm during the study period." [2] ~Toxic and Trace Elements in Tobacco and Tobacco Smoke," M. Chiba and R. Masironi, Bulletin of the World Heatth Organization 70(2): 269-275, 1992 [Issue 23, Item 32] The authors daim that cigarette smoke may be a "substan- tial source" for the smoker of elements including arsenic, selenium and polonium-210. Moreover, they claim that nonsmokers may also be exposed via "passive smoking." The second author, tL Masironi, was formerly the coordinator of WHO's Tobacco or Health Programme, ~While the harrnfu/health effects of carbon monoxide, nicotine, tar, irritants and other noxious gases that are present in tobacco smoke are wetl known, those due to heavy metals and other toxic mineral elements in tobacco smoke are not sufficiently emphasized. Tobacco smoking influences the concentrations of several elements in some organs. This review summarizes the known effects of some trace elements and other biochemically important elements [aluminum, arsenic, cadmium, chromium, copper, lead, manganese, mercury, nickel, polonium-210, selenium, and zinc] which are linked with smoking. Cigarette smoking may be a substantial source of intake of these hazardous elements not only to the smoker but

I JULY 1992 I ! 1 I I I I I I ! t I I I I i I I also, through passive smoking, to nonsmokers. The adverse health effects of these toxic elements on the fetus through maternal smoking, and on infants through parental smoking, are of special concern." "Environmental smoke pollution consists not only of the smoke exhaled by the smokers but also of sidestream smoke emitted by the burning cigarette. Sidestream smoke, which is inhaled by nonsmokers (passive smok- hag), usually contains relatively high concentrations of many noxious substances including heavy metals. As particle sizes are smaller in sidestream smoke than in mainstream smoke, the deposition in the lung tissue of passive smokers reaches deeper into the alveolar spaces." "[A] positive association was found between urinary [arsenic] levels in children and parental smoking habits. The mean [arsenic] level in the urine of children of non- smoking parents was 4.2 ug/g creatinine, in children with one parent smoking it was 5.5 ug/g and in children with both parents smoking it was 13 ug/g creatinine." "Most of the [cadmium in the dgarette] passes into sidestream smoke, thus posing a risk to passive smokers." "Passive smoking plays an important role in exposure of children to lead. Parental smoking, but no other environ- mental or dietary factors, was found to be related to the blood [lead] level in children; 30 ug/l on average in children on non-smoking parents, 37 ug/1 if only the father smoked, and 47 ug/1 if the mother, or both parents, smoked. Even if children lived near a lead smelter parental smoking had a significantly stronger influence on blood lead levels (35 ug/1 in children with non-smoking parents, 38 ug/l if only the father smoked, 43 ug/1 if the mother or both parents smoked, and 46 ug/1 if the mother smoked more than 15 dgarettes/day. The number of cigarettes smoked by the parents was negatively correlated with the British Ability Scales scores, which measure the ability and attainment of small children." "Tobacco smoke containing [polonium-210] combines with household dust, settles on surfaces and clothing, and contributes to household radon-emitted alpha-radioactivity." [3] "Measurement of Salivary Cotinine to Judge Tobacco Smoke Exposure in Wheezing Children," . P.W. HeTmann, A.L. Duff, E. Pomeranz, L.E. Gelber, A.H. Farris and Tdi.E. Platts-Mills, Abstract No. 347, Journal of A~g~ and C~nical Immunology89(1): Part 2, 1992 [Vol. I, Issue 21, May 8, 1992] In a population of 61 children treated for wheezing, cotinine was measured in saliva. In children under the age of two who were reportedly exposed to ETS, the mean cotinine concentration was 19.9 ng/ml; in children age two and over, it was 7.1 ng/ml. The difference between the two groups was reportedly statistically significant. ExcHlr~: "Multiple factors contribute m acxite wheezing episodes in childhood including tobacco smoke extmsure. Since 3040 percent of adults are smokers, children in these households are commonly exposed to passive tobacco smoke and are reported to be at greater risk for more frequent respiratory tract infections and increased airway hyperresponsiveness. Cotinine~ a major metabollte of nicotine, can be measured.. • in saliva in patients exposed to tobacco smoke. In 61 children treated for wheezing in a Pediatric ER, 60 percent under the age of 2 and 50 percent over the age of 2 were exposed to household tobacco as judged by questiormaire. In saliva from patients < [less than] age 2, the mean cotinine concentration was 19.9 rig/m] in exposed children (n = 14) and 2.9 ng/ml in non-exposed children (n = 7). In patients > [greater than] age 2, mean values for cotinine were 7.1 ng/m] lind 3.0 ng/ml in exposed (n = 33) and non-exposed (n = 18) chil&en, respectivdy. The mean levd of cotinine in exposed patients was significandy highe~ in patients < age 2 than in patients _> age 2 and similar to levels previously reported for individuals who activdy smoke. These results demonstrate that passive exposure to tobacco smoke l_~cls to significant levels of mtlnine in saliva from young wheezing patients in whom smoke exposure has been reported to be a risk factor for symptoms." [4] "An Objective Assessment of Environmental Tobacco Smoke (ETS) Exposure in 5-7 Year Old Children," S. Clark, T. Assadullahi and J.O. Warner, Abstract No. FC28, CBnicalandF_ape~- mentalAIK, rgy22: 109-142, 1992 [Vol. I, Issue 2I, May8, 1992] The authors used liquid chromatography to measure salivary cotinine in a group of asthmatic children and thdr age-matched controls. Although 31 percent of cases were reportedly exposed to ETS according to questionnaire responses, the ootinine assay suggemed that 68 percent were actually exposed, the authors claimed. "We have devdoped a sensitive assay of ootinine using high Performance liquid cah_romatography (HPLC) to quantitate [ETS] exposure in a group of 5-7 year old asthmatic children compared to a group of aged matched controls." "We chose to use.., saliva collected by absorption into dental rolls ha the mouth for 5 minutes. Our modified - 53-

ETg AND IAQ_REFERENCE JULY 1992 extraction procedure was highly reproducible .... The parents were asked to fill a questionnaire on atmospheric pollutants to obtain an estimate of declared ETS exposure in the home." "Results showed 31 percent of the asthmatic patients according to the parents were exposed to ETS but by HPLC 68 percent had been so exposed (n = 19). From the control group the figures were 40 percent and 51 percent of patients respectively." "Therefore an objective assessment is essential as ETS is more ubiquitous than is apparent from questionnaire alone. Large studies are required to establish assodations between ETS exposure and atopic disease." [5] =Exposure to Passive Smoking Among Bar Staff," M.J. Jarvis, J. Foulds and C. Feyerabend, British JournalofAddiction 87:111-113, 1992 [Vol. I, Issue 18, March 20, 1992] The authors measured saliva mtinine as a quantitative measure of exposure to ETS among 42 nonsmoking bar staffers in central London and Birmingham. Based on measured saliva cotinine concentrations, the authors estimated that nicotine intake was about 0.6 mg per day, which they stated was equivalent to 0.6 ofa dgarette. "We have investigated passive smoking in non-smoking bar staff using saliva cotinine as a quantitative measure of their exposure to environmental tobacco smoke." [Forty- two employees of pubs in central London and Birming- ham, England, participated.] "The mean saliva cotinine concentration [in saliva] was 9.28 ng/ml." "Using a formula.., and on the basis that saliva cotinine concentrations are 1.3 times higher than in blood, the median cotinine concentration observed corresponds to an estimated daily nicotine intake of about 0.6 mg." "The levels observed in bar staff are about double those in children with two smoking parents, and are about four times higher than in adults reporting recent exposure." "Since smokers take in, on average, about 1 mg nicotine from every cigarette smoked, our findings suggest a dally nicotine intake equivalent to 0.6 of a cigarette. It should be stressed that this estimate applies to nicotine only, and exposure to other smoke components could be tither relatively higher (as in the case of tobacco-spedfic nitrosamines) or lower. Furthermore, since it is not known which components of the smoke are responsible -54- for particular health effects, it cannot be assumed that the overall risk of smoking related disease among non- smoking bar staffwould be similar to that run by a smoker who smoked and inhaled halfa dgarette a day. However, the results do provide prima fade evidence of likely harm to health. Among children codnine concen- trations much lower than those found here have shown associations with glue ear and with impairment of some aspects of lung function which were statistically signifi- cant and independent of a variety of potential confound- ers." [6] "Tea Drinking, Passive Smoking, Smoking Deception and Serum Cotinine in the Scottish Heart Health Study," H. Tunstall-Pedoe, M. Woodward and Ca~l. Brown, Journal of Clinical Epidemiology44(12): 1411-1414, 1991 [Vol. I, Issue 18, March 20, 1992] This study set out to investigate a recent claim by Idle that the reliability of cotinine in assessing ETS exposure was confounded by nicotine from certain foods, particu- Iarly tea. In this survey of more than 3,000 men and women, serum cotinine reportedly showed no association with self-reported daily average tea consumption. Accord- ing to the authors, Idle's claim is not supported by the results of this study. EXCERI'~: "In a recent review in this Journal, Idle questioned the use of cotinine to identify 'smoking deceivers', and claimed that its reliability in assessing passive smoking was confounded by nicotine in certain foods, particularly tea. In [another] paper we report on smoking habits and smoking biochemistry in 10,350 Scottish men and women in the Scottish Heart Health Study, one of the largest cross-sectional studies to use serum cotinine. We have now used this study to test Idle's suggestion con- cerning tea, because this population has a high prevalence both of tea drinking, the traditional national beverage, and of passive smoking. In addition, we have looked at the distribution of the range of serum cotinine readings in self-reported smokers and non-smokers to demonstrate 'deception'." "Our first analysis concerned passive smoking. We have analyzed median serum cotinine levels, for each degree of self-reported passive smoking, against the reported average number of cups of tea consumed daily." "In the second analysis, to demonstrate differently the existence of smoking 'deceivers', we have analyzed the distribution of serum cotinine both in self-reported non- smokers and in those who smoked any tobacco product."

VOLUME I, ISSUES 1-24 JULY 1992 ! I I I I I I I I I l I I i I I l "There is a possible step-up of serum cotinine in those drinking 10 or more cups of tea a day, but the effect.is inconstant and there is no consistent gradient below this. Passive smoking however, shows a consistent gradient in median co,nine levd which is much more powerful: 9 of the 10 sub-groups of men and women who deny exposure to tobacco smoke have median cofinine levels of 0 or 0.01 ng/ ml; whereas 7 out of 10 sub-groups exposed to 'a lot' of tobacco smoke had median leeds of 1.0 ng/rnl or more." "We are unable to substantiate the claim that tea drinking will confound the effects of passive smoking on serum codnine. Imperfect though it is, serum cotinine appears to be an indicator of passive smoking. This is shown by the strong gradient in median level, for differ- ent levels of reported tobacco smoke exposure, in self- declared non-smokers with low levels of expired air CO and serum thiocyanate. Even the low but measurable cotinine levels in non-smokers who deny exposure to tobacco smoke could, nonetheless, come from tobacco, if they have forgorten their exposure, or because nicotine can be absorbed through the skin, as well as the lungs, from environmental sources." "The range of cotinine readings shows that some self- reported non-smokers are absorbing the same amount of tobacco products as the heaviest smokers." "In our opinion these results support the use ofcotinine both to assess passive smoking and to identify 'smoking deceivers'.... While not denying the possibility of some dietary contribution to serum cotinine, and the variation in individual metabolism described by Idle we assert nonethe- less that passive smoking is the major cause of measurable low levels of cotinine in true non-smokers, and that high leeds are virtually pathognomic of furtive smoking, other- wise known as smoking deception. With soda] pressures to conceal active smoking, and current interest in the harm to health of passive smoking0 the value of cotinine measure- ment should not be underrated." [7] "Nicotine and Cotinine in the Cervical Mucus of Smokers, Passive Smokers, and Nonsmokers," M.F. McCa.nn, D.E. Irwin, L&. Walton, B.S. Hulka, J.L. Morton and C.M. Axelrad, Cancer Epidemiology, Biomarkers & Prevention 1 : 125- 129, 1992 [Vol. I, Issue 18, March 20, 1992] After studying cervical mucus obta_ined from 50 women visiting a cancer clinic in North Carolina, the authors reported that nicotine was detected in the cervical mucus of all smokers and that cotinine was detected in the mucus of 84% of smokers. Nicotine and cotinine levels for women reportedly exposed to ETS versus those women not reporting ETS exposure were much lower than for women who currently smoked. In addition, little difference was reported between the nicotine and cotinine levels among women reporting ETS exposure and those who did not report exposure. "Although epidemiological studies suggest that dgarette smoking is a risk factor for cervical cancer, further evidence is required to document the biological plausibil- ity of this relationship. This study obtained cervical mucus, using a cervical flush technique, from 50 patients in a neoplasia clinic." "The women were divided into three smoking status categories: (a) smokers (at least 1 cigarette/day); (b) passive smokers (exposed to cigarette smoke at work or at home in the last 24 h but who do not smoke themselves); and (c) nonsmokers (reporting no exposure to cigarette smoke)." "All smokers had detectable levels of cotinine in their mucus, as did 58% of the passive smokers and 67% of the nonsmokers. Cotinine was present in 21 of the 25 smokers (84%) but was not der.ected in any passive smokers or nonsmokers." "The mean nicotine level for smokers was significantly different from that for passive smokers and for nonsmok- ers. The mean cotinine level was also significantly different for smokers compared to both passive smokers and nonsmokers. Comparison between the passive smokers and the nonsmokers revealed no significant differences in mean nicotine or mean cotinine levels." "The present study, together with the earlier studies of nicotine and cotinine in cervical mucus, supports the biological plausibility of the assodafion between exposure to tobacco products and risk of cervical cancer. Future epidemiological research should evaluate the correlation between nicotine or cotinine levds in cervical mucus and the severity of dysplasia. = [8] "Exposure to Environmental Tobacco Smoke in Naturalistic Settings," ICM. Emmons, D.B. Abrams, R.J. Marshall, R.A. Etzel, T.E. Novotny, B.H. Marcus and M.E. Kane, American Journal of Public Health 82(1): 24-28, 1992 [Vol. I, Issue 16, February 17, 1992] In this study, 186 nonsmokers used an exposure diary to self-monitor ETS exposure and also completed a ques- tionnaire assessing patterns of ETS exposure. The authors reported that the subjects' assessments of exposure were consistently lower on the questionnaire than in the diary. -55-

ETS AND IAQ.R_EFERENCE JULY 1992 I They also reported that the primary source of ETS exposure was the workplace, unless the subject lived with a smoker, in which case the household was the primary source. "The presence of a smoker in the household resulted in [reports of] higher levels of exposure both at work and in other locations when compared with subjects without household exposure," according to the study. "The data presented here are from a larger study designed to develop interventions to reduce ETS exposure in the workplace. The purpose of the present study was to develop a method for self-monitoring ETS exposure continuously over a 7-day period, to use this method to document the frequency of exposure in natural settings, and to document selected exposure parameters. Results obtained with this self-monitoring technique were also compared with global ratings typically used in survey "The exposure questionnaire provided a global estimate of subjects' exposure to ETS. The questionnaire was designed to didt information about the respondents' perception of their level of ETS exposure (number of minutes per day, intensity of exposure) at various loca- tions (work, home, and other)." "... Subjects were instructed to define 'exposure' as contact with an individual who was actively smoking. The parameters assessed by the diary included number and duration of exposures, location of exposure (work, home, other), intensity of exposure (number of smokers), and distance from the source of exposure ('near' = less than 5 feet; 'far' = more than 5 feet)." "This study represents one of the first attempts to develop a prospective, diary measure of daily exposure to ETS, to examine sources of exposure for a 1-week period, and to compare the diary with a global questionnaire measure of exposure. We found the exposure diary to be a relatively simple and effective method of assessing ETS. Furthermore, the data were quite stable." "The majority of ETS exposure occurred in the work- place. The percentage of subjects reporting exposure at work (75.6%) replicates that found by Cummings et al. (75%) in a different sample. Furthermore, both the duration and the intensity of exposure were greater for subjects with a smoker in the household, regardless of the degree of exposure in the workplace. Importantly, subjects with smokers in the home had virtually equiva- lent exposure across all se~tings, whereas subjects who did not live with smokers had the majority of their exposure at work and very litrde in other settings." "The subjects' own assessments of thdr exposure levels on the questionnaire were oonsistendy lower than the levels determined by the diary, suggesting that studies utilizing global exposure ratings from surveys may underestimate actual exposure. Tl'6s may be due in part to the insidious nature of exposure -- a high level of vigilance is necessary if accurate estimates are to be obtained." "These data may be useful in developing effective public health polities and in designing intervention programs targeted at both individuals and organizational settings to reduce nonsmokers' exposure to ETS." [9] "Misdassification of Smoking Status By Self- Reported Cigarette Consumption," E.J. P~rez- Stable, G. Mar~n, B.V. Marln and N.L. Benowitz, American Review of Respiratory Disease 145: 53- 57, 1992 [Vol. I, Issue 16, February 17, 1992] Self-reported dgarette consumption and serum codnine levels were compared in a sample of 743 participants in the Hispanic Health and Nutrition Examination Survey. Of 189 stir-reported nonsmokers, 12 were defined as "biochemical smokers" via their cotinine levds, and were considered to be misclassified. As only one of the 12 misdassified nonsmokers reported living with a current smoker, the authors found it "unlikely" that the cotinine levels in the biochemical smokers were a result of environ- mental tobacco smoke exposure. "To evaluate possible misdassification of smokers and nonsmokers, we compared self-reported dgarette con- sumption and serum cotinine levels in a sample of 743 Mexican American partidpants in the Hispanic Health and Nutrition Examination Survey (HHANES) .... We defined biochemical smokers as persons with serum cotinine levels > 0.084 bt/vt/L (14 ng/ml). Misclassification was defined as a discrepancy between self-reported smoking and the serum cotinine level used to define a biochemical smoker." "Of 189 self-reported nonsmokers, 12 (6.3%) were found to have serum cotinine levels > 0.084 IAVI/L and were defined as biochemical smokers .... These 12 subjects had a mean cotinine level of 0.354 g.M/L, including 3 with measurements greater than 0.600 blM/L. Of the 65 subjects who claimed to be former smokers, 7 (10.8%) were found to be biochemical smokers, com- pared with only 5 [of] 124 (4%) never smokers .... Only 3 of the 189 (1.6%) self-reported nonsmokers had cotinine levels close to the overall mean for smokers." "Only 1 of the 12 self-reported nonsmokers with serum cotinine above the cutoff responded that there was a I I I i I I I i I I I I I I I

! VOLUM~ ~, ~.SSUE£ ~.-24 JULY 1992 I I I i I I I I I I I i I I I I I I smoker living at home, and he had a serum cotinine of 0.330/~2v~/L... of the 177 stir-reported nonsmokers with serum cotinine < 0.084 hIM~L, 55 (31.1%) reported living with another smoker at home." "... Using serum cotinine values to assess smoking status, we found that self-reported behavior may result in misdassification of some smokers as nonsmokers, and vice versa. These proportions are small, and occasional smoking was not identified by the questionnaire used. However, misdassification of tobacco eXlX)sure has potential epidemiologic implications in determining assodations between active or passive smoking and disease." "Our analyses of Mexican American participants in HHANES show that approximately 6% of reported nonsmokers had levels of serum cotinine above our cutoff for nonsmokers .... However, the majority of defined biochemical smokers had cotinine levels consistent with light or occasional smoking." "The possible contribution of environmental tobacco exposure to the serum cotinine level of our study sample can also be estimated. HofFman and collaborators found that the maximum serum cotinine achieved in nonsmoking volun- teers exposed to sidestream smoke in an experimental setting was only 0.020 g.M/L. Jarvis and cotleagues reported that environmental tobacco exposure in a sample of 100 non- smokers resulted in average plasma cotinine levels of less than 0.018 bI.M/L.... Thus, it is unlikely that cotinine levels in biochemical smokers were a result of environmental tobacco exposure However, it is possible that more detailed questions on environmental tobacco exposure than available from the HHANES would identify self-reported nonsmok- ers with additional exposure that may account for a portion of the serum cotinine levels." "Misdassification of a person as a nonsmoker who is really a smoker has implications in estimating the health risk associated with exposure to environmental tobacco smoke. Such misdassification increases the apparent relative risk of smoking-related diseases in nonsmokers and needs to be corrected for in calculating risk estimates for environmental tobacco smoke exposure. Based on published studies, the proportion of misclassification is estimated to range from 1 to 7%. Our data support the assumption ofa misclassification rate at the higher end of this range .... [M]ost misclassified nonsmokers are probably very light smokers or occasional smokers who 'binge.' As such most misclassified nonsmokers would be expected to be at lower risk for smoking-related diseases, such as lung cancer, than would most self-reported smokers, who have much greater exposures. This observa- tion should be considered in making corrections for misdassification for risk estimation of lung cancer in n onsm o~iers, ~ [10] "Dietary Nicotine: A Source of Urinary Cotinine," R.A. Davis, M.F. Stiles, J.D. DeBethizy and J.H. Reynolds, Fa[ Chem, Tox/c. 29(12): 821-827, 1991 [Vol. I, Issue 16, FebruatT 17, 1992] In this paper, tLJ. Reynolds scientists report on analyses of common food items for the presence of nicotine. Of the foods analyzed, the highest nicotine contents were found in instant teas. Nicotine also was de~ecred in black tea, tomato and potato, but not in eggplant or green pepper. The authors present a range of potential cotinine values that could be obtained in persons eating average amounts of these foods, and propose that such levels could confound epidemiologic studies where cotinine measurements are used. "Recently, doubt has been cast on the widely accepted bdief that nicotine is unique to tobaoco. Plant sources other than tobacco have been reported to contain varying levels of nicotine; many of such plants are common dietary constitu- ents, for example egg plant, tomato and green pepper." "The purposes of this research were to confirm the presence of nicotine in common foods and to assess the use ofcotinine as a biomarker for ETS in the light of the results." "Foods, prindpally from plants in the family Solanaceae, and a number of teas were examined for the presence of nicotine. Dietary nicotine would give rise to cotinine in urine and compromise estimates of exposure to tobacco smoke that depend on urinary cotinine. All foods were homogenized, ex'n-acted and analyzed for nicotine and cotinine by gas chromatography with nitrogen-sensitive detection (GC) and/or GC/MS (mass spectrometry). Weak acid and aqueous extracts of the teas were analyzed in a similar manner." "Our results confirm that certain vegetables, black tea and instant tea contain nicotine. The nicotine concentration for market tomatoes was about one-half of the nicotine concen- nation in fresh locally grown tomazoes. The concentration of nicotine in the flesh of the potato was approximardy three-fold greater than the concentration in the peal. Nicotine was also detected in caukhqower, a non-solanaceous vegetable. No nicotine was observed above background in green pepper and egg plant -- two solanaceous vegetables previously reported to contain nicotine." "The source of the nicotine found in these plant products poses an interesting question. Either the plants possess the - 57-

ETS AND IAQ_REI::ER.ENCE JULY biosynthetic pathway for the production of nlcotine, or it results from some contamination, perhaps insecticide residue." "If only average quantities of [these] foods are constlmed, then the daily intake of nicotine would be 8.8 lag. The resulting urinary cotinine concentration is estimated to be 0.6 ng/rnl. Maximal food and tea consumption results in an approximately ten-fold increase in nicotine intake (100 day) and an estimated urinary cotinine concentration of 6.2 ng/ml. This range of estimated urinary cotinine concentra- tions (0.6 to 6.2 ng/ml) is very similar to that observed in non-smokers who claim no exposure to ETS, and in ETS- exposed non-smokers." "Diet is not usually controlled in most studies of ETS. Most investigators assume that many of the uncontrolled variables will 'average out' in daily-life exposure studies." "We have demonstrated the presence of nicotine in foods and beverages common to the diet and discussed the implications of this finding for studies related to ETS. We concur with Idle that studies in which cotinine is used as a biomarker for nicotine exposure must address individual physiological, genetic, and environmental factors, including diet." [11] "Gas-Paxticulate Phase Distribution and Decay Rates of Constkuents in Ageing Environmental Tobacco Smoke," G.B. Neurath, S. Petersen, M. D/inger, D. Orth and F.G. Pein, Environmental Technology 12: 581-590, 1991 [Vo|. I, Issue 12, December 10, 1991] The authors report on chemical characteristics of ETS components produced by the burning of 27 cigaret~ for 11 minutes in an experimental chamber. While some of the components could potentially be used as markers, the authors caution that the components' behavior at high experimental levds may vary from "real-life" situations. "This paper gives some basic data on important smoke constituents that could be envisaged as markers: distribution and time variation of the respective partitions between the particulate and gaseous phases for solanesol, nicotine, neophytadiene, limonene, and 3-ethenylpyridine. These have been measured in five minute samples over fifty five minutes aging ofsrnoke generated from burning dgarerres. The dynamics of the phase distribution of nicotine has been observed related to such short sample-collection times." "For the high concentration model experiment 27 filter cigarettes were lit wir2fin 60 seconds by two persons with lighters and were allowed to statically bum to a butt length of ca. 23 mm. The dgarettes were extinguished by turning the glowing cone into holes of an ashtray. The combustion -58- took about 11 minutes. No persons were present in the room during the experiment." "Solanesol and protonated nicotine behave like particu- late matter (UV-PM). Free nicotine evaporates from the particulate phase during the 5 minute sampling time and decays rapidly from the gaseous phase. Neophytadiene leaves the particulate phase faster than predicted by deposition. Limonene, 3-ethyenylpyridine, and naphtha- lene are present only in the gaseous phase and have slightly different decay rates." "Systematic studies on measurable constituents of ETS, like those reported here, may fadlltate cautious conclusions to the behaviour of other components which can not be determined at suffident sensitivity to study thdr dynamics in ETS at present. Even in this study the determinations requir~ higher concentrations than are likely in real IL~e situations in the indoor environment. Further, the be_ha'dour of the smoke constituents at the very low concentrations occurring in indoor situations may differ from that at higher leeds in modal experiments. The limitations of the work axe acknowledged at this time. However, the improvement of analytical methods should provide more realistic data in the [12] "Measuring Personal Exposure to Airborne Mutagens and Nicotine ha Environmental Tobacco Smoke," N.Y. Kado, S.A. McCurdy, S.J. Tesluk, SAC Hammond, D.P.H. Hsieh, J. Jones and M.B. Schenker, Mutation Research 261: 75-82, 1991 [Vol. I, Issue 11, November 22, 1991] A bacteria] assay was utilized to measure the mutagenic- ity of extracts of persona] sampling filters worn by persons exposed to ETS. Paxtides colleczed on the filters were reportedly mutageni~ mutagenic activity also corrdated well with measured nicotine levels. "Extracts of the particulate matter collected from ETS have been reported to be mutagenic in Salmondla and measurements of mutagenic acti-dry may provide an estimate of individual exposure to mutagenic compounds in the complex mixture of ETS, especially if measure- ments are obtained from personal sampling." "We report here the use of the [Salmonella] microsuspension assay for measuring muragenidty of organic solvent extracts ofpersona.l sampling filters. The filters were placed near the breathing zone and connected to low volume sampling pumps worn by individuals exposed to ETS. These portable sampling pumps axe worn on the waist and are used widely in occupational exposure studies. These portable units were also placed at I I I I I I I I I t !i I I I i

I VOLUME I, ISSUES 1-24 JULY 1992 I I I I I I I I I I I I I I I I I I a single location (fixed-site) throughout the sampling period. In a pilot study, we examined two separate indoor environments for exposure to the mutagens present in ETS. The mutagenic activity was compared to levels of airborne particulate matter and to airborne nicotine, a marker of ETS." "Mutagenic activity of particles collected from personal filters was detected for sampling times of 40 n'fin to 6 h in the two environments where ETS was present. The mutagenic acr.iviry correlated well with the total particle concentrations present in the locations studied based both on airborne concentrations and absolute amounts per filter, suggesting that a similar source was responsible for these activities. The concentrations of particles at the casino and bingo fadlides that represented a zero level of mutagenic activity would be of interest to study indoor air that has confirmable ETS or other combustion products present, and would reveal what level and range of background particulate matter is mutagenic in these environments2 "The strong correlation of nicotine and mutagenicity from personal filters suggests that the level ofmutagenic activity attributed to ETS in these environments can be estimated based on nicotine levels." [13] "Passive Smoking By Humans Sensitizes Circulat- ing Neutrophils," R. Anderson, AJ. Themn, Gall. Richards, M.S. Myer and A.J. Van Rensburg, American Review of Respiratory Disease 144: 570- 574, 1991 [Vol. I, Issue 10, November 1, 1991] Healthy, young nonsmokers were exposed to active smoking in a poorly-ventilated room for three hours. Blood samples were tested for several parameters, e.g., number of white blood cells, oxidant production and neutrophil activity, as surrogates for the inflammatory response. The authors reported that several of the parameters were in- creased following ETS exposure, and suggested that these indicated a "prokn.flammatory response." "The primary objective of the present study was to investigate the effects of inhalation of sidestrearn smoke on the numbers and activity of circulating neutrophils and levels of proinflammatory cymkines in plasma of healthy, young nonsmokers. A secondary component of the study was the investigation of acute dgarerte smoking on the same parameters in active smokers who had refrained from smoking for 14 [hours]." Sixteen nonsmokers were exposed to actively smoking individuals in a poorly-ventilated room for periods of three hours. Blood samples were drawn before and after exposure, and tested for several parameters, e.g., number of white blood cells, oxidant production, neutrophil activity, as surrogates for the inflammatory response. "Passive inhalation of cigarette smoke for 3 [hours] was associated with increased numbers of circulating leuko- cytes as well as increased oxidant production and chemo- taxis of circulating neutrophils .... [C]onfirmation of these apparent proinflammatory effects of passive smok- ing was obtained in a follow-up investigation designed to exclude serial bleeding [blood samples] on a given day as a possible complicating factor." "[W]e do concede that the present study was designed primarily to investigate the effects of acute, experimental passive smoking on neutrophil numbers and fanctions and that additional investigations are necessary to establish the long-term effects of chronico low-levd exposure to sidestream cigarette smoke on leukocyte numbers and activities." [14] '~Assessing Exposure to Environmental Tobacco Smoke: Is It Valid to Ex*rapolate From Active Smoking?," M.J. Reasor and J.A. Will, Journalof Sraoking-Related Diseases 2(1 ): 111-127, 1991 [Vol. I, Issue 8, October 1, 1991] This review paper examines the question of whether exposure to ETS can be assessed by extrapolation from active smoking and concludes that it is extremely difficult, if not impossible, to do so with any degree ofreliability. "This review examines the question of whether exposure environmental tobacco smoke (ETS) can be assessed by extrapolation from active smoking." "Since ETS has not been adequately characterised, there are insuffident data on which to base a hazard analysis. Accordingly, there are not enough data available on which to base an exposure assessment for ETS. Due to the dynamic nature of ETS, it is impossible to relate ETS to MS chemically or physically. In the absence of this relationship, it is inappropriate to make any extrapola- tions from what is reported about the effects of active smoking to possible effects of exposure to ETS. There- fore, any calculation of risk from exposure to ETS based on extrapolations from calculated risks of active smoking is, at best, not reliable and, most probably, of no value whatsoever. It is important, therefore, to consider ETS as a distinct entity, and further research is needed to test hypotheses based on valid protocols that meet the criteria established for epidemiology of weak associations." -59-

ETR AND IAO.~R_EFERENCE JULY 1992 [15] "Mutagenicity of Indoor Air Containing Environ- mental Tobacco Smoke: Evaluation of a Portable PM-10 Impactor Sampler," P.E. Georghiou, P. Blagden, D.A. Snow, L. Winsor and D.T. Wil- liams, Environmental S~ience and Technology 25(8): 1496-1500, 1991 [Vol. I, Issue 7, Septem- ber 13, 1991] This paper describes an indoor air sampling impactor which can be used to collect particulate matter samples. The authors report that the mutagenic responses in the Ames assay of particulate extracts from environments containing ETS were greater than those for environments that did not contain ETS. The investigators are affiliated with the Memorial University of Newfoundland, and Health and Welfare, Canada. "It is apparent from these field tests that the IASI [indoor air sampling impactor] can be used to collect indoor PM-10 [inhalable suspended particulate matter having median diameters < 10 grn] samples in sufficient amounts for gravimetric analyses and mutagenidty testing with a modification [according to Kado et al.] of the Ames S. typhimurium assay. The IASI proved to be quiet and unobtrusive and was well-tolerated by the inhabitants of the test houses/locations that were sampled. The fact that it can sample PM-10 at 10.0 L/min enables it to be used in virtually any indoor setting for periods of as low as 1 h if samples are needed for mutagenidty assays." "The mutagenic responses.., of the extracts of PM-10 from the environments containing ETS were 2-16 times greater than those observed for environments that did not contain ETS." [The availability of this sampling device may make it easier for people to collect samples of indoor air particu- lates for mutagenidry testing. Several groups of investiga- tors have previously claimed that ETS is the major contributor to the mutagenicity of indoor air.] [16] "Environmental Tobacco Smoke in Public Places," U. Brynnel and G. Lofroth, Meeting Abstract, Third Europe~cm Meeting of Environmental Hygiene, Dusseldorf, Germany, June 24-27, 1991 [Vol. I, Issue 6, August 27, 1991] ETS was studied in shopping malls and restaurants. Nicotine was reported to be a poor quantitative indicator for ETS because it disappeared faster than did other components. The mutagenic acdvity of particulate matter was reportedly much higher in the indoor environments than in the outdoor air. EXCERPt: "Environmental tobacco smoke (ETS) is a major indoor air pollutant wherever tobacco smoking takes place. The measurement of ETS is hampered by the fact that most of its components, such as particles, carbon monoxide, nitrogen oxides and gas phase hydrocarbons, have other sources in the present age of combustion. In addition, tobacco smoke specific compounds, such as nicotine and solanesol, require relatively elaborate sample collection or analysis. In a study of ETS in shopping malls and restaurants we selected to use two parameters as indicators for ETS, nicotine and mutagenic activity of particulate matter .... The initial work showed, however, that nicotine in real world samples is present in both the gas phase and the particulate matter contrary to the published data obtained in test chamber analyses. Subsequent studies, including aging of ETS in a room, showed that the ratio of gas phase to particulate nicotine decreases with the age of the smoke and that gas phase nicotine disappears much faster than other ETS components... The mutagenic activity of particulate matter collected in restaurants and shopping malls was much higher than that of simultaneously collected outdoor ambient air. The total nicotine concentrations in these public places... were grossly related to the mutageni¢ activity .... The conclusions of this stud)" are that nicotine cannot be used as quantitative tracer of ETS as it disappears faster than other ETS components and that air pollution, as mea- sured by mutagenic activity of the complex mixture in particulate matter, is much higher in indoor smoking environments than in the outdoor ambient air of a busy urban center." [17] "A Comparison of Methods of Assessing Exposure to Environmental Tobacco Smoke in Non- Smoking British Women," C.J. Proctor, N.D. Warren, Mael.J. Bevan and J. Baker-Rogers, Environment International 17: 287-297, 1991 [Vol. I, Issue 5, August 9, 199I] Four methods for assessing exposure to ETS were compared: personal monitors for nicotine and volatile organic compounds (VOCs); saliva analysis for cotinine; and questionnaires on lifestyle and ETS exposure. The personal monitor data suggested that most ETS exposure was from the domestic environment, but that mean exposures were very low. Cotinine, although considered to be a useful marker for ETS exposure, was detected in some subjects not exposed to ETS. VOCs were not .,.t significantly correlated with either nicotine or cotinine. Questions regarding spouse smoking habits did not correspond well with either levels of exposure to nicotine or salivary cotinine measurements. (/3 I I I I I I I I I I I I I I I i -60 -

VOLUME I, ISSUF_S 1-24 JULY 1992 t I I I I I I I I I I I I I I I [ A study based on 52 non-smoking Bridsh women was conducted in 1989. This is the prototype for a study recendy conducted by RJR sdendsts in Columbus, Ohio and a study proposed for conduct in Japan. The partidpants wore active (i.e., pump-driven), personal monitors for nicotine and volatile organic compounds (VOCs) over a 24-hour period; supplied samples of saliva for cotinine analysis; and answered questions regarding lifestyle and ETS exposure. Airborne Nicotin¢. "Our results suggest, as has been indicated by previous studies, that the most influential source of exposure to ETS is in the domestic environment and that contributions at work, leisure or during travel are much smaller. Moreover, mean exposures are low... the median daily exposure over all subjects would be less than the nicotine equivalent of 0.001 dgarette." Salivary Cotinine:. " ... levels tend to increase with exposure to airborne nicotine, but many subjects not exposed to ETS were found to have detectable salivary cotinine levels at values that overlapped with subjects that were exposed." VOC~. '%vvrallace has suggested that exposure m environ- mental tobacco smoke corresponds to significantly increased personal exposures to benzene, xylene, ethylbenzene and styrene .... If it were m be the case, then it might be expected that there would be a sr.mng correlation between personal exposures to nicotine and to these aromatic VOCs. As shown in Table 3, there is no significant corrdation between either benzene and nicotine or between benzene and salivary cothaine levels. The same holds true for xylene, ethylbenzene, and styrene." "The study evaluated various objective and subjective methods for assessing populations' exposure to ETS. It was foumd that questions regarding spouse smoking habits, as commonly used in epiderniological studies, did not oorre- spond wd/with either levels of exposure m nicotine or salivary cotinine measurements, although trends did exist." [18] "Correlations Between Urinary Nicotine or Cotinine and Urinary Mutagenicity in Smokers on Controlled Diet," C~.. Rahn, G. Howard, E. Riccio and D. J. Doolittle, Environmental and Molecular Mutagenesis 17: 244-252, 1991 [Vol. I, Issue 5, August 9, 1991] Although ETS is not mentioned, this study presented dam suggesting that nicotine and codnine can be used as markers for tobacco smoke exposure in smokers. The study reported that mutagens were present only in the urine of smokers of tobacco-burning dgarettes, and not in the urine of those who smoked tobacco-heating dgarettes, suggesting that the mutagens are pyrolysis products, not nicotine or coti~na EXCERPTS-" RJR sdentists a.re coauthors on this paper and Dr. N. Benowitz is acknowledged for the determination of urinary nicotine and cotinine. No mention of ETS. This paper is a follow-up to an earlier one which demonstrated the importance of controlling for diet in studies of the relationship between smoking and urinary mutageniciry. "In the present study, urinary nicotine and cotinine were used to provide a reliable assessment of tobacco smoke exposure, and correlations between urine mutage- nicity and tobacco smoke exposure were evaluated for each dgarette type [i.e., for conventional dgarettes and for dgarettes which heat, but do not burn, tobacco] 2' "The results of this study dearly show that, whereas nicotine and cotinine are useful as indicators of tobacco smoke exposure, they are not the mutagens in smokers' urine. The mutagens in smokers' urine are likely to be pyrolysis products from tobacco, since they are present in urine of smokers of tobacco-burning dgarettes but not in the urine of smokers of tobacco-hearing dgarettes." [19] "Evaluation of Vapor-Phase Nicotine and Respi- rable Suspended Particle Mass as Markers for Environmental Tobacco Smoke," B.P. Leaderer and S.K. Hammond, EnvironmentalSclence and Techno/ogy 25(4): 770-777, 1991 [Vol. I, Issue 2, May 17, 1991] Combined environmental chamber and field study. Chamber study evaluated vapor-phase nicotine (VPN) rdadve to particle mass concamtrafions for different brands of dgarettes. Fidd study (conducted in winter of 1986) assessed 1-week-average VPN concentrations (passive monitor) and pardde mass concentrations (gravimetric method) in residences as a funczion of the number of cigarettes smoked and the presence of other sources. "No trend in RSP emission rates by FTC tar and nicodne rating were observed... The 10 U.S. brands of cigarettes emitted similar amounts of RSP." "The consistency in the RSP/VPN rados observed in the chamber suggests that V'PN is a good predictor of ETS- assodated RSP .... It is not known if vapor-phase nicotine is in a constant rado with other important gas- and particle- phase ETS constituents (e.g., nitrosamines) among various dgarette brands. Additional chamber experiments to address this question are currently underway.~ "Detectable levds of nicotine were measured in 47 (49%) of the 96 residences. In these homes there is a considerable shift: in the RSP distribution toward higher concentrations -61-

ETS AND bexQ R_EFER_ENCE JULY 1992 with the average concentration (RSP = 44 ug/m3) 3 times those in homes without measurable nicotine levels (RSP = 15 ug/m3) .... Nicotine was measured in 13% of the residences that reported no smoking, while nicotine was not detected in 28% of the residences that reported smoking." "Measured nicotine concentrations were found m be good predictors of both RSP and the reported number ofdga- rettes smoked in the residences. The dominance of the source in controlling RSP and VPN concentrations in the residences is demonstrated in the very small additional explained variation in RSP and VPN attributable to other factors (house volume, [measured air-exchange rate], etc.)." [20] "Elevated Nicotine Levels in Cervical Lavages from Passive Smokers," C.J. Jones, M.H. Schiffman, R. Kurnan, P. Jacob and N.L. Benowitz, American Journal of Public Health 81(3): 378-379, 1991 [Vol. I, Issue 1, April 30, 1991] This study was conducted to confirm the presence of nicotine in the cervical mucus of nonsmoldng women who were part of a larger study of cervical neoplasia in three Washington, D.C. area hospitals. One hundred forty-five women who were found to have routine Pap smears were interviewed regarding ETS exposure and a 3 ml saline lavage was collected for measurement of cervical nicotine levels by gc-ms. "The assodation of recalled environmental tobacco smoke exposure in the past 24 hours and detectable nicotine in cervical lavage was highly significant (p = 0.001)." "Exposure in the home resulted in the highest nicotine levels observed; however, women exposed only outside the home also had elevated levels of nicotine compared to those not exposed." "Among the 38 (sic) women with exposure at home in the past 24 hours, there was no association observed between nicotine levels and the number of smokers to whom the women had been exposed, the usual smoking intensity of those active smokers, or the number of hours since last exposure. Moreover, for women exposed to smoke in the home, additional exposure outside the home was not related to nicotine level. Cigarette smoke ac- counted for virtually all of the exposure at home. Among those women exposed only outside the home, there was no relationship observed between time since last exposure to environmental tobacco smoke and nicotine level." "The biological significance of cervical nicotine levels resulting from passive exposure to tobacco smoke is not kllown." -62- "It is difficult to reconcile the relatively lbw levels of nicotine that we observed in passive smokers with the large magnitude of the relative risks reported by S/artery, et al., who found a risk from passive smoking [for cervical cancer] similar to that of actively smoking women." "[P]assive smoking should be evaluated further as a risk factor for cervical disease." INDOOR AIR QUALITY [1] Letters to the Editor Regarding "The Measure- ment of Environmental Tobacco Smoke in 585 Ofl:i~e Environments," S. Turner, L. Cyr and A.J. Gross, Environment International 18: 19-28, 1992 [Issue 24, Item 4] Environment [nternationalrecently published two letters concerning this article, which was summarized in issue 14 of this Report, Jan uary 17, 1992. The authors of the letters were (i) James L. Repace and Alfred H. Lov, a'ey and (ii) Turner, er al., authors of the original paper. The letters appear at Environment International 18:311-314 and 318-320, 1992. In their original a.r~icle, Turner, et al., reported that low levels of ETS did not significantly affect indoor air quality and that acceptable indoor air quality could be main- tained via good ventilation with "moderate" amounts of smoking. In their letter, Repace and Lowrey stated that the results of the study were "meaningless" and that its conclusions were "unjustified" because, they claimed, Turner, et al., did not report air exchange rates in the buildings they studied. Repace and Lowrey also characterized the Turner, et al., paper as having a "massive omission of experimental data." Repace and Lowrey further claimed that the smoker densities reported by Turner, et al., were not representative of a typical office building, and claimed that the RSP levels reported in the paper were associated with a substantial lung cancer risk for nonsmokers. Turner, et al., replied by noting that their paper used a statistical method called discriminant analysis, which is used to predict an outcome. They stated that "the proper use and application of this model does not appear to have been grasped" by Repace and Lowrey. Turner, et al., noted that other scientific literature had produced results similar to their study. They also responded to Repace and Lowrey's statement about omitting sdentific data by noting that it is normal to publish only the pertinent data; in fact, reviewers of the paper requested that Turner, et al., omit certain data plots. Turner, et al., called for further research to turn to indoor pollutants that are not ! I ! I ! I ! I i i I i I i i

I VOLUME 1, 1SSUES 1-24 JULY 1992 I I I I I I I I I I I I I I I I I as well understood, such as microorganisms, rather than to continue focusing on ETS. [2] "Sick Buildings and the Experimental Approach," D.P. Wyon, Environmental Technology 13: 313- 322, 1992 [Issue 24, Item 57] This article reported on a series of field experiments conducted at a Swedish hospital where a large number of employees had reported sick building syndrome symp- toms. An experimental approach investigating several technical remediations reportedly found a decrease in sick building syndrome symptoms with the experimental application of certain remediations, i.e., reduced tempera- ture, reduced glare from lighting, increased humidity, and ionization, but not with increased or decreased ventila- tion, reduced use of cleaning chemicals, or reduced static electricity. "In order to justify expensive and energy-consuming technical measures, it should ideally be demonstrated that they will reduce complaints indirectly, that is, by affecting the indoor climate, and not just by their effect on opinion, for in the latter case PR measures might be more cost-effective. They would certainly be quicker and would use less energy. An agreed procedure for proving that the proposed technical measures really do alleviate reported symptoms by means of their effects on the indoor climate is therefore required. This paper addresses that require- ment." "A large proportion of the [approximately] 1000 people working in the surgical wing of Maim_ General Hospital (MAS) in 1987 were dissatisfied with the indoor condi- tions in this relatively new building, in fact over 250 of them used the formal procedure available in Sweden for registering claims of industrial injury to document their view that working in the building was musing them to experience symptoms assodated with Sick Building Syndrome (SBS)." "All of the reported symptoms were of the SBS type, and as the building was new, built and equipped to a very high standard, and apparently in good working order, the probability of being able to identify the causative factor or factors simply by measuring every available physical and chemical parameter of the indoor environment was judged to be very low." "The nine experimental conditions were established for three-week periods." "1. Reduced airflow: supply and exhaust airflows were reduced by 30 percent." "2. Increased airflow: supply and exhaust airflows were increased to their design maximum, an increase of 40 percent from normal practice." "3.A.ir filters: 40 free-standing aircleaners were installed in the various rooms of the experimental ward. Each unit was equipped with a fan which passed room air through a pre-iqlter and a fine filter." "4. New cleaning routines: floors, furniture and shelves were cleaned only with water, i.e. without the normal cleaning and disinfectant agents." "5.Reduced room temperatures: set values for supply air were reduced by 3°C, and radiator thermostatic valve se~tings were reduced by 2°C. The measured reduction in room temperature averaged 1.5°C." "6.Antistatic treatment: conventional measures were taken to reduce static electric charging by treating floors, equipment and clothing." "7. Reduced lighting glare: the lighting units in the wards were fitted with baffles to reduce glare .... In the experimental ward these baffles were exchanged for baffles painted a brownish yellow." "8. Ionisation: 37 small air ionisation units were installed in the various rooms of the experimental ward, each unit supplying negative ions to the room air and having a small positively charged external panel." =9. Humidification: steam humidification was provi- sionally installed in the pressure chamber of one of the two ventilation systems of the building." "SBS-symptoms were not significantly affected by increas- ing or decreasing the ventilation rate, by redudng the amount of chemicals used in cleaning, or by the reduction of personal static charge. A significant positive affect was obtained by reducing the room temperature by 1.5%2 and by redudng lighting glare. Humidification of the air significantly reduced the measured severity of SBS-symp- toms .... Ionisation with negative ions.., had a highly significant and benefidal effec~ on SBS-symptoms." "[S]ubjective symptoms of SBS were found not only to be correlated with objective signs in a given Population, . but "" also to co-vary with them in response to changes in the physical conditions under which the subjects worked." "Field experiments, in which the effeczs of various technical measures of SBS are studied empirically, can provide a cost- effective basis for investment decisions, whether or not the underlying cause of the problem is understood." - 63-

[3] "IAQ and Energy-Management By Demand Controlled Ventilation," F. Haghighat and G. Donnini, Environmental Technology 13: 351-359, 1992 [Issue 24, Item 58] Two identical floors of an office building were com- pared: one had a conventional HVAC system, and the other had a demand-controlled system, where outdoor air flow was linked to indoor CO2 level. Energy consumption was reportedly less with the CO>controlled system, while IAQ constituent levels and occupant comfor~ perception did not deteriorate. F..XCERPTS t "The purpose of this study is to test a carbon dioxide controlled ventilation system in a oommerdal building as a method of controlling indoor air qua/iV, occupant comfort, and energy consumption, and to compare them with the air quality, thermal comfort, energy demands, and occupant satisfaction resulting from a conventional system." "The energy demand for the floor that had the CO2. based controller was lower than for the floor with the conventional control system. The greatest energy saving occurred during the month of October, when little outdoor air was supplied on the CO~.controlled floor, indicating that the supply air temperature and the CO2 levels were adequate." "This study was undertaken to show that a supply air temperature and carbon dioxide sensor that measures the occupant generated carbon dioxide could be used to control the use of outdoor air in a more effident manner." "The study verified that the energy consumption lessened with the CO~.controlled ventilation system as compared to the conventional ventilation system. It was also observed this did not worsen the quality of indoor air and thermal comfort, and the occupants did not perceive a deterioration in their working environment." [4] "Indoor Air Pollution and Its Health Effects in China p A Review," B.H. Chen, C.J. Hong and X.Z. He, Environmental Technalogy 13:301-312 [Issue 24, Item 59] This article reviewed Chinese studies on indoor air pollution and epiderniology. Chinese indoor air quality issues are reportedly substantially different from those in the West. The authors noted that ETS and active smok- ing were generally controlled for in epidemiologic studies which concluded that indoor air pollution was considered to be related to several adverse health effects. EXCERPTS: "Indoor air pollution in titles in China is mainly from domestic coal combustion, and from coal and biomass fuel combustion in rural areas. Factors affecting indoor air pollutant concentrations include fuel, ventilation, floor plan of house, and season, etc. Pollutants emitted from coal combustion are significantly higher than those from gas or liquefied petroleum gas. Among biofuels used in China, biogas is the least, firewood the second, and cow/ sheep dung the most polluting biofuel in terms of indoor pollution levels. Adverse health effects of indoor air pollution due to fuel combustion in China are diverse in nature and vary in severity, ranging from pre-dinical decrease in local non-specific immune function and pulmonary function, to elevated prevalence of respiratory diseases and symptoms, lung cancer, espedally female lungcancer, birth defects and fluorosis. In most of the epidemiologic studies cited below, due precaution was taken to stratify for the effect of smoking and passive smoking. Indoor air pollution is particularly harmful to the health of women and children, especially aged women and young children. Intervention measures have been taken to ameliorate indoor air pollution and need to be strengthened." "Coal combustion products, e.g. suspended particulates (SP), sulphur dioxide (SO2), polycydic aromatic hydro- carbons (PAH), carbon monoxide (CO) eta, are the major indoor air pollutants from cooking. The indoor air pollution problem in China is thus quite different from that in developed countries, where gas combustion produc~s, smoking~ and organics emit-ted from furniture are the major sources of indoor air pollution." [5] "Nasal Deposition ofIrthaled Acrolein and Acetaldehyde Vapors," J.B. Morris, CIAR Currents 1(3): 1, 4, 1992 [Issue 23, Item 20] Acetaldehyde and acrolein have been identified in indoor air, and apparently have many potential sources. The author measured nasal uptake of acetaldehyde in rats, and found that uptake differed greatly at levels of 1000 ppm acetaldehyde, the concentration reported to produce a carcinogenic response, from uptake at 1 ppm, a concen- nation potentially found in indoor air. Moreover, when animals were exposed to acrolein while bring exposed to acetaldehyde, acetaldehyde uptake was inhibited. The author noted that such interactions must be evaluated when considering exposure to mixtures of agents such as may be found in indoor air. "Acetaldehyde and acrolein, two aldehyde vapors which are present in indoor air, are derived from many sources. I i i i I I I I I I I I I I I l I

I VOLUME I, iSSUES 1-24 JULY 1992 I i I I I I I I I I I I I I =1 I I I In.halation exposure of rodents to high concentrations of dr_her aldehyde results in epithelial cell damage in the upper airways with minimal damage to tissues in the deep lung. Acetaldehyde is carcinogenic in rodents, with chronic exposure to concentrations in excess of 1000 ppm produdng laryngeal tumors in hamsters and nasal tumors in rats. Since indoor air concenmadons are several orders of magnitude less than 1000 ppm, the potential adverse effec-cs of typical indoor air exposure scenarios are unclear." "Research in this laboratory is aimed at examining the dosimetry of inhaled acetaddehyde and acrolein in rat respiratory tract. These studies are focused on the nasal c~vity as this is the site of lesions in acetaldehyde- or acrolein-exposed rodents." "IT]he kinetic~ of uptake differ greatly at 1000 ppm, the concentration needed to produce a carcinogenic response, from that at 1 ppm, a concentration which more closely approximates the levels which might occur in indoor air. Thus, direct linear extrapolation of the data from the animal carcinogenicity studies to predict effects at low concentrations is not appropriate." "Indoor air pollution represents a complex mixture of agents. Acetaldehyde uptake was measured . . . in animals co-exposed to acrolein to determine if acetaldehyde dosimetric relationships were altered by simultaneous presence of another aldehyde .... [A]crolein produced a concentration-dependent inhibition of acetaldehyde uptake .... The~e results indicate that strong dosimetric interactions occur for these aldehydes. The potential for such interactions must be considered in evaluations of potential effects from exposure to complex mixtures of these agents such as typically occur in indoor air." [6] =Modeling the Indoor Environment," B.S. Austin, S.M. Greenfield, B.R. Weir, G.E. Anderson and J.V. Behar, Environ. SH. TechnoL 26(5): 851-857, 1992 [See Appendix A] [Vol. I, Issue 22, May 22, 1992] The authors reviewed currently available computer models that estimate exposure to indoor air substances, and advocated that persons using such models evaluate their strengths and limitations. "In this paper we briefly review currently available computer models that estimate human exposure to indoor Pollutants. We examine representative indoor air models and their input data requirements from the standpoint of their relative effectiveness in a variety of problem condi- tions. This paper is not a definitive, systematic evaluation of the currently available databases and models. Rather, it aims to help the model user (decision maker) develop a process by which to evaluate the strengths, limitations, and range of applicability of available models and data. Understanding the inherent accuracy of our analytical tools is particularly relevant when these tools are used to assess health risks." "Many regulatory issues require the use of risk assess- ments containing human exposure estimates. The significant contribution of indoor air quality to an individual's total exposure (most people spend well over 90 percent of their time indoors) requires a more system- atic approach for choosing how to estimate exposure than is currently available. A number ofmodellng tools can be used to estimate indoor pollutant concentrations and human exposure." "A/though each model has its strengths and weaknesses and in most cases is continually being improved, few risk assessors are aware that mos~ available models were designed for a specific set of conditions and may not be the best model for their particular purpose. This potential probterrr can result in errors ranging from gross to subtle. This section presents an initial step in choosing the most appropriate mode/for a given application: to examine a general set of rdevant parameters." "A key to choosing the best model for an application is understanding how it addresses various parameters relevant to the application." "[A] new generation of'total human exposure' models has been developed recently to simulate human exposure to environmental pollutants rather than just concentra- tions in various geophysical media. As discussed, a vital component of these models is their utilization of activity pattern information." "In the context of exposure assessment, human activity patterns are quantifiable descriptions of the things people do and when, where, and how long they do them during the conduct of theia" daily fives. People spend the majority of their time in a varieD" of indoor environments; therefore, human activity patterns -- appropriatdy combined with microenvironmental pollutant concentrations -- are a vital dement in determining the exposure of populations to indoor pollution." "To consider the 'total exposure' of a population to various contaminants, it is important to describe both the indoor environment and human population activity par'terns as accuratdy as possible. This information vdll enable derision makers to develop strategies for minimizing health risk. Many of the components needed to perform a comprehen- sive indoor air quality analysis are available."

AND IAQ REFERENCE JULY 1992 "However, despite the progress to date, model input data (source emissions, sinks, distributions, and population activity patterns) are lirrfited and restrictive; collateral information that would permit a degree of generalization is lacking; evaluation of models and input data with regard to uncertainty, limitations, and other comparative attributes is not adequate;, and relationships between airflow (ventilation, filtration, streamlines, etc.), sources and sinks, concentration, and exposure have not been established. The demand for decisions regarding the design of risk-reduction strategies has resulted in the utilization of available models and data despite their shortcomings." "/Analysis, and thus decisions, would be significantly improved if models and databases were better evaluated, and if the attributes of the analytical system were matched to the user's need to support the selection of risk-reduc- tion strategies. We recognize the need to improve our knowledge of the indoor environment, the complex determinant factors, and the models and databases that permit effective analyses. However, the development now of'objective selection and application standards for analytical approaches is essential in light of the prolifera- tion of risk-reduction programs and guidelines." [7] ~Air Quality During the Winter in Quebec Day- Care Centers," S. Daneault, M. Beausoleil and K. Messing, American Journal of Public Health 82: 432-434, 1992 [Vol. I, Issue 19, April 10, 1992] This study reported that over 90% of 91 day care centers surveyed in Montreal had CO2 levels in excess of 1000 ppm. Some scientists consider this level of CO~ to be indicative of poor IAQ. High levels of CO2 are associated with respiratory tract and other symptoms, according to the study. "[A]iz quality was mor~itored in 91 day-care centers, representing 86% of such centers in the South Shore and eastern regions of Greater Montreal. The study was designed to assess air pollution and other physical conditions in the centers, and, in a second stage, to relate them to the health of the children amending the centers. We present here the assessment of physical conditions." "In this study, the temperature, percentage of relative humidity, and levels of carbon dioxide (CO2) were chosen because they are (1) easy to measure, (2) reliable indica- tors of ventilation quality, (3) recommended by regulat- ing agencies as the first step of tight building evaluations, and (4) used by Scandinavian researchers, providing a basis for comparison. It should be noted, however, that toxic levels of unmeasured substances may persist in the presence -66- of low levds of these indicators, partioAarly in the case of multiple-vocation buildings." "The COa levds for the 91 day-care centers are shown. There is no prescribed limit for day-care centers, but the average level for all 1672 measurements was 1505 ppm, wed/ over the office building standard of 1000 ppm... Only 9 of the 9I centers never exceeded 1000 ppm, and 13% exceeded 2500 ppm." "The levels of relative humidity were in the recorm-nended range in 76 day-care centers (83%). Temperature was in the comfort zone (20%2 to 23.6~2) in 92% of the centers." "The main predictor [of COs levels] was the density of children in the center when the measurements were made. Each child present per cubic meter increased the level of CO2 by nearly 70 ppm. The use of an electric heating system (in 63% of the centers) increased the mean COx level by 250 ppm. The presence of a ventila- tion system (in 32% of centers) reduced CO2 levels by a mean of 562 ppm. Each yeaz of building age decreased the level of COa by 12 ppm." "The level of COi is important in itself, since it has been related to various symptoms, such as nasal discharge, eye irritation, throat dryness, headache, dry sldn, and leth- argy. In addition, the level of COa is an indicator of the quality of ventilation in general, and may thus serve as a marker for other pollutants, not verified in the present study, such as carbon monoxide and formaldehyde." "If the COl levels and some of the humidity levels de- scribed in this study had been found in office buildings, immediate action would have been recommended according to the limits suggested by the American Society for Heating and [sic] Refrigerating and Air Conditiorting Engineers." ~Adaption to Indoor Air Pollution," L. Guamarsen and P.O. Fanger, Environment International 18: 43-54, 1992 [Vol. I, Issue 18, March 20, 1992] Using a climate chamber, the authors exposed 32 subjecm to differing concentrations of human bioeffluents (odors produced by humans), tobacco smoke and emissions from building materials. The change in the subjects' perception of polluted air was studied during the first 15 minutes of exposure. The authors reported that the subjects' perception ofbio~luents decreased after a few minutes, and that adaption to tobacco smoke apparently occurred, although the subjects' perceptions depended on the concen- tration of the exposure. The authors propose that the level of ventilation necemary for comfort may be reduced if occu- pants are allowed to adapt to their environment. I I I I I I I I I I I I I

l ~OLUME 1, ISSUES 1-24 JULY 1992 I I ! I I I ! I I i t I I I I I I "The purpose of the present investigation is to study discomfort caused by typical indoor pollution before, during, and after a transient period of adaptation." "[H]uman bioeffluents, tobacco smoke, and emissions from building materials were studied. Perception of pollution from one source, as well as perception of pollutants after previous exposure to other pollutants were studied, with subjects in different adaptational states." "Bioeffiuents. The adapted subjects voted low intensity and high acceptahifity rather independent of pollution level. The non-adapted subjects fdt the odor to be more intensive and less acceptable at an increasing air pollution levd." "Tobacco smoke. Intensity increased and acceptability decreased with increasing concentration both for non- adapted and for adapted subjects. But the adapted felt the intensity slightly less intensive and more acceptable than the non-adapted subjects." "The air is percdved least acceptable immediately after people enter a space with air pollution. After some minutes, people may adapt and the air is felt more acceptable. Adaptation improves acceptability of air quality considerably when humans pollute the air; some improvement occurs when moderate tobacco smoking is the source .... When designing ventilation for adapted people, human bioeffluents may be neglected ira mini- mum ventilation rate per person of 1-2 Us is assured, and the ventilation requirements based on moderate tobacco smoking may be reduced to 50%." [9] "Occupational Stress in the Aircraft Cabin," M. Edwards, Cabin Crew Safety 26(5): 1-6, 1991 [Vol. I, Issue 16, February 17, 1992] This article addresses the issue of stress experienced by cabin crew members during the course of normal duties. The author proposes that stress may be attributed to such conditions as the quality of cabin air, the working space and equipment used on board, the hours of work and the social organizational aspects of the job. The magazine Airline Executive [nternationa~ in its December 1991 issue, characterized this study as reporting that "It]he effects of air pressure, low humidity, ozone and carbon dioxide rob flight crews of productivity and could often be the cause for the discomfort that is attributed to tobacco smoke." "The present article addresses the issue of stress experi- enced by cabin crew members during the course of normal duties." "Under typical flight conditions, the cabin crew may be subject to stress that can be attributed to a number of sources. These include the quality of the a/r in the cabin, the working space and equipment used on board, the hours of work, and the sodal organizational aspects of the job." "The confined spaces of the aircraft cabin must be well- ventilated to ensure the comfort and well-being of its Occupants°" "Although aircraft fly at heights considerably higher than 10,000 feet, these harmful effects [of reduced air pressure] are minimized because the cabin is generally pressurized to an equivalent of about 6,000 feet. How- ever, this does not mean that there are not effects from the change of pressure. The feelings of'bloated stomach' and swollen legs reported by the cabin crew are assodated with carrying out physical work at reduced air pressure." "Low levels of relative humidity lead to dry skin, dry eyes and dry mucous membranes in the nose and throat. Dry skin may lead to early aging effects; dryness of the eyes produces discomfort and it may cause pawticular problems for wearers of contact lenses which are not designed for use in dry air; dry mucous membranes in the respiratory tracz may account for the observed increased susceptibility of cabin crew to upper respiratory tract infections such as colds and sore throats." "... Symptoms typical ofozone-toxidty have been reported three-to-four times more frequently by flight attendants in aircraft flying long distances at high altitudes than by those in aircraft flying short distances at lower altitudes." "Ozone affects the respiratory system and causes coughing, irritation of the throat, chest discomfort and difficulty in breathing .... Those who suffer from asthma appear vulnerable to more asthmatic attacks when exposed to increased levels of ozone. People who are active during exposure to ozone, and therefore breathing more air, are more vulnerable to its effects than people who are sedentary. Cabin crew members consequently suffer from the effects of ozone more than passengers." "Levels of cosmic radiation increase as altitude increases, thus creating a potential hazard for aviation personnel. Cabin crew members flying an average schedule could experience an increase in the annual radiation dose by 250 totem. There is a greater hazard to those flying frequently at very high altitudes, and a busy schedule on supersonic flights could result in an annual increase of up to 700 mrem." "The quality of cabin air is the focus of numerous complaints by cabin crew members and it is possible that - 67-

F_T~ AND LAO~I:LEFEP~NCE JULY 1992 ETS may cause some annoyance among non-smokers. However, the symptoms attributed to ETS, such as eye discomfort, coughing and respiratory complaints, are those known also to be associated with ozone and low levels of humidity. Further study is required [in] order [to] measure the effects of ETS both in the aircraft cabin and elsewhere." "The consequences of a poorly-designed working environment are the various musculoskeletal disorders of which cabin crew members complain. These include increased fatigue and a greater likelihood of injuries such as cuts, bruises, bums and scalds." "Work schedules that demand wakefulness when the body expects to be asleep lead to the disturbances of body rhythms with consequential digestive problems, menstrual irregulari- ties, difficulties in keeping awake and general feelings of malaise. These schedules also require that sleep take place during periods of natural wakefulness. The difficulties in sleep experienced by many people in these circumstances result in sleep deprivation and ~onsequent fatigue." "Compared with other types of employment .... the flight attendant is less able to maintain the sort of social contacts and ~:tivides which fit in with a ground-based job that involves regular daytime hours." [10] "The Los Angeles TEAM Study: Personal Expo- sures, Indoor-Outdoor Air Concentrations, and Breath Concentrations of 25 Volatile Organic Compounds," h Wallace, W. Nelson, R. Ziegenfiis, E. Pellizzari, L. Michael, R. Whitmore, H. Zelon, T. Hartwell, R. Perritt and D. Westerdahl, Journal of Exposure Analysis and Environmental Epidemiology 1 (2): 157-192, 1991 [Vol. I, Issue 15, January 31, 1992] This article reports on a 1987 study conducted in Los Angeles as part of the EPA-sponsored Total Exposure Assessment Methodology (TEAM) studies, which assess human exposure to volatile organic chemicals (VOCs). The 1987 study assessed VOC levels in personal air (measured by an air monitor carried by the participant), indoor air (measured in the home) and outdoor air (measured in the back yard). The study reported that personal air concentrations of VOC_s exceeded outdoor concentrations for all chemicals except carbon tetrachlo- ride, and that indoor concentrations were "nearly always greater" than outdoor concentrations. "Since 1980, the U.S. Environmental Protection Agency (EPA) has supported a series of studies of human exposure to volatile organic chemicals (VOCs). These studies, known as -68- TEAM (Total Exposure Assessment Methodology) Studies, have discovered several major sources of exposure for a number of toxic and carcinogenic chemicals. Large-scale studies were carried out in 1981-83 in New Jersey and in 1984 in Los Angeles and Antioch-Pittsburgh, California. In 1985, the Congress appropriated resources for further studies of exposure to volatile organic compounds. In response, three studies were carried out in 1987 in Balti- more, New Jersey and Los Angeles." "The major goal of the 1987 Los Angeles TEAM Study was to determine the personal, indoor, and outdoor air concentrations of 25 VOCs for a set of Los Angeles residents." The study's objectives were to: "Determine the relation- ships between personal exposure, indoor and outdoor concentrations"; "[d]etermine the suitability of using exhaled breath measurements to estimate previous exposure"; "[d]etermine the variability oFVOC concen- trations within a home"; "[d]etermine the 'emission rates' of homes"; "[e]xtend the coverage of chemicals and the sampling and analysis methods"; and "[d]etermine seasonal and multi-year variability." "Parfidpants carried a personal air monitor .... Two consecutive 12-hour samples were collected. Concurrently, two consecutive 12-hour outdoor air samples were collected in the back yards of the homes. Breath samples were collected at the beginning, middle, and end of the 24-hour monitoring period... Indoor air concentrations were measured in the kitchen.., and in the living area... All analysis of personal air, breath, and indoor and outdoor air samples.., by [gas chromatography/mass spectrometry]." "Earlier findings confirmed by this study include the following: (1) personal air concentrations exceed outdoor concentrations for all chemicals except carbon tetrach]orid¢ (2) personal air and outdoor air concentrations are greater in winter than in spring or summer; (3) overnight outAoor air concentrations exceed daytime values in winter but not in spring or summer; and (4) breath values reflect previous personal exposures. New findings included: (1) indoor concentrations were nearly always greater than outdoor concentrations; (2) overnight indoor concentrations were about the same as overnight personal air concentrations; (3) daytime personal air concentrations often exceeded daytime indoor concentrations; (4) air exchange rates were found to be 2-3 times as high in summer as in winter; (5) 12-h average indoor concentrations varied line from room to room; (6) breath concentrations were stable from morning to everting0 from season to season, and from year to year, suggesting that breath measurements collected after an hour or more in the home may reflect long-term near-eq,~lihrium concentrations. I I I I l I I I I I i I I I I l I I

I "VOLUME I, iSSUES JULY 1992 I I l i I I I I I I I I I I I I I "Common daytime activities include travel, work, and smoking, all of which have been found in previous TEAM studies to be significant contributors to personal exposure to the target chemicals... However, only 11 of the 51 partidpants in February smoked, and only 11 went to work on the day they were monitored; thus it appears likely that other personal activities accounted for a substantial portion of personal exposure." [11] ~Indoor Climate and Air Quality in New Offices; Effects of a Reduced Air-Exchange Rate," B.M. Berardi, E. Leoni, B. Marchesini, D. Case.eRa and G.B. Ra.ffi, International Archives of Occupatlonal andEnvironmentalHealth 63: 233-239, 1991 [Vol. I, Issue 15, January 31, 1992] An indoor air quality investigation was initiated among personnel of an Italian bank who had recently moved into a new building. Employee health complaints (i.e., symptom prevalence), were evaluated via questionnaire; ventilation effectiveness, contaminant concentrations and microdimate variations were studied as well. The authors concluded that reported health complaints could not be ascribed to any of the substances measured. They re- ported that insufficient ventilation, namely an inadequate amount of "fresh" outside air, was related to the com- plaints. "In response to a request from the personnel of a credit bank that had been moved into a new building, an evaluation was made of employee health complaints. The prevalence of symptoms was determined using a self- administered questionnaire. Ventilation effectiveness, contaminant concentrations (which included chemicals and bioaerosols) and microclimate were studied to establish their relationship to environmental discomfort and to the health problems mentioned by the employees." "The 20 symptoms that appeared in the questionnaire were as follows: eye heaviness, flicker, eye-burning sensation, redness, photophobia, watering, vision blunting, color flinging, visual spots, double images, visual blurring, eye irritation and itching, contact lens problems, dry throat, skin dryness, nose irritation/itching, nasal congestion, dripping nose, drowsiness and headache." "The mean number of symptoms reported per respondent was 4.3, and the maximal number of symptoms reported by an individual was 15." "The most frequently reported symptom was a dry throat (48.6%), which, together with dry skin (28%), was thought to be caused by the low humidity of r.he environment. The symptoms of a possible 'allergic' reaction in the nose and eyes showed the next highest inddence. Headache was mentioned by 29% of the employees." "After the initial study of prevalence of symptoms, we carried out a survey of occupational hygiene in the building and of the problems eventually associated with the perfor- mance of individual parts of the [HVAC] equipment." "The results of indoor climate measurements provided evidence of satisfactory thermal conditions for the credit bank personnel in the new building offices. It was shown that although the overal] air-exchange rate was in accordance with the ventilation standards, defident dilution of indoor air with 'fresh' outside air resulted in increased contaminant concenn'ations. The reported health problems could not be ascribed to any of the contaminants studied, since the values measured lay well below the current health standards for exposure; nevertheless, an insuaq:ident rate of air exchange is commonly found in new or remodeled buildings. We would conclude that although a causative agent was not identified, ventilation dd:idendes may be a primary factor contributing to employee discomfort; the establishment of optimal ventilation conditions should not be underestimaxed in the design of fully enclosed working places." [12] "The Measurement of Envlronmental Tobacco Smoke in 585 Office Environments," S. Turner, L. Cyr and A.J. Gross, Environment International 18: 19-28, 1992 [Vol. I, Issue 14, January 17, 1992] The authors used a statistical procedure based on data on respirable suspended particles (RSP), nicotine, carbon monoxide, carbon dioxide, room size and occupancy to sample and predict the smoking status of 585 offices. Very few nonsmoking rooms were misdassified as smoking, suggesting little "ETS spillover" from smoking areas. On the or_her hand, a substantial number o£smoking rooms were classified as nonsmoking, suggesting that levels of ETS in offices may be low. "In order to provide information on levels of environ- mental tobacco smoke (ETS) in office environments during 1989, a total of 585 offices was sampled for a number of factors, including respirable suspended particles (RSP), nicotine, carbon monoxide, carbon dioxide, room size, average number of room occupants, and number of ciga- rettes consumed. Each data set was collected over a one-hour sampling period." The authors used discriminant analysis, a statistical procedure that analyzes a number of variables (here, the air sampling data) to predict a case's (i.e., an office's) member- skip in a group ("smoking" vs. "nonsmoking" status). "The - 69-

JULY 1992 sdected model was able to dassiaCy properly 96.1% of the nonsmoking rooms as nonsmoking. However, only 41.4% of the smoking rooms were dasskfied as smoking; 58.6% of the smoking rooms were dassified as nonsmoking. Overall, 65.1 [%] of the cases were properly classified." "The most significant results from the discrknLnant analysis are the following: (1) RSP and nicotine contribute to the prediction of • room type--smoking or nonsmoking. (2) most (96.1%) nonsmoking rooms are classified as nonsmoking rooms, demonstrating very little evidence ofETS spillover from smoking areas, and (3) a significant number (58.6%) of total smoking rooms are classified as nonsmoking rooms." "[S]moking rooms can be separated as 'light' or 'heaw'. The light-smoking rooms appear equivalent to nonsmok- ing rooms when considering, in a multivariate context, the important factors of RSP and nicotine. The heaW- smoking rooms do have elevated levels of nicotine and RSP. This indicates that there may be a rough working range of smoker density in which smoking activity does not seem to influence ETS levels significantly, as mea- sured by RSP and nicotine." "These 585 data sets reveal some other interesting information. If one examines the absolute levels of items of particular concern, such as nicotine, RSP and carbon monoxide, they show relatively conservative values. For instance, most of the RSP data in smoking-observed areas show levels well under that reported in Repace and Lowrey's early papers; in general, about four times less than their mathematical model predicted for office environments of 200 gglmL" "These 585 measurements of some components of ETS and other related parameters sampled during 1989 suggest overall concentrations of ETS in typical office workspaces ro be considerably lower than estimated ten years previously. Some parameters, such as carbon monoxide, appear to be only weakly related to smoking activity. Discriminant analysis shows that when 'blind- folded' for presence or absence of smokers, in most cases realistic smoking levels do not significantly influence the aspects of air quality that were measured, and spillover from smoking areas into nonsmoking areas appears to be minimal. This work further reinforces the position the American Sodety of Heating, Refrigerating and Air Conditioning Engineers (ASHRAE) has taken on ETS in of~ce buildings in ASHRAE Standard 62-89, in that acceptable air quality can be maintained in properly -70- ventilated offices with a moderate amount of smoking, even without smoker segregation. These data help to further define the limits of moderate smoking." [ 13] "Relationship Between Occupant Discomfort as Perceived and as Measured Objectively," F. Haghighat, G. Donnini and R. D'Addario, Indoor Environment 1:112-118, 1992 [Vol. I, Issue 13, January 3, 1992] The authors experimentally varied ventilation rate in a building marked by numerous occupant complaints. They reported that the frequency of complaints was not associated with the ventilation rate or with measurements of indoor air constituents, but rather with the occupants' perceptions of the indoor environment. "The main objective of this paper is to show that in some cases the complaints reported by occupants are assodated with perceived rather than measured levels of indoor environmental parameters." "The building in which the investigation was carried out is an eleven-stoW building... Occupants of this building expressed symptoms believed to be assodated with poor air quality and irritating odors." "Experiments were designed to find out whether there is any relationship between the amount of fresh air (ventila- tion rate), and occupants' complaints. To answer this question, the position of the outdoor supply air dampers was varied from 0° (total recircuiated air) to 25°, 75° and 90° (total fresh air). The damper modulation was per- formed in a random order so as to ensure that the occupants were unaware of any pattern of change." "Most noteworthy in the responses was that more than 34% of the occupants expressed that the air is dry. As indicated, the measured relative humidity ranged from 40 to 65%. Another interesting finding in the responses was that more than 32% of the occupants expressed that in general, the thermal environment was unsatisfactory, even though almost all the measured thermal comfort param- eters complied with the standards." "[M]ore than 41% of the occupants complained of dry throat, regardless of the ventilation rate. Similarly, more than 29% of the occupants responded positively to eye irritation; more than 22% to breathlessness; more than 29% to drowsiness, and more than 26% to difficulty in concentration. In response to odor, 25-53% responded that they did not perceive odors; while 6-32%o respond that they percdved odors regularly." I I I I I I I I I I I I I I I I I l

I VOLUME I, ISSUES 1-24 JULY 1992 I I I I I I I I I I I I I I I I I "The office building under consideration, during our study, was considered 'sick' due to the fact that more than 20% of the occupants complained of health problems such as dry throat, eye irritation, breathlessness, drowsi- ness, and a difficulty in concentration. There is no correlation between the questionnaire responses and the actual environmental conditions. The measurements of the parameters satisfied the air quality and thermal environment standards. Therefore, this investigation showed that complaints reported by the occupants were associated with perceived rather than measured levels of indoor environmental parameters." [14] "Indoor Air Pollutants From Unvented Kerosene Heater Emissions in Mobile Homes: Studies on Particles, Sernlvolatile Organics, Carbon Monox- ide, and Mutagenicity," J.L. Munfford, R.W. Williams, D.B. Walsh, R.M. Burton, D.J. Svendsgaard, J.C. Chuang, V.S. Houk and J. Levcva~, Environmental Science and Technology 25: 1732-1738, 1991 [Vol. I, Issue 12, December 10, 1991] Eight mobile homes where kerosene space heaters were used were included in this study. Air quality measure- ments indicated significant increases in levels of CO and organic compounds. Pollutant emissions exceeded U.S. ambient air standards in half the houses. "This study assessed human exposure to air pollutants from unvented kerosene heaters in mobile homes. Eight electric homes with no smokers were monitored for airborne particles of < 10 lain in diameter (PM,0), semivolatile organics, and carbon monoxide with the kerosene heaters on and off. The organic emissions were assayed for polycyclic aromatic hydrocarbon (PAIl), nitro-PAH, and for mutagenicity in a Salmone[da ~vphimurium reverse mutation assay." "Mobile homes are a fast-growing segment of the housing market. Few studies have been reported on the effects of usage of unrented kerosene heaters on indoor air quality in mobile homes. The objectives of this study are (a) to investigate the effects of kerosene heater usage on indoor air quality by comparing the indoor air pollutants, including airborne particles of < 10 ~ in diameter (PM,0), semivolatile organics, CO, and muta- gens in mobile homes when kerosene heaters are on and off, and (b) to assess the human exposure to these air pollutants from u.nvented kerosene heater emissions in the real-life home environment and relate pollutant levels to the U.S. National Ambient Air Standards." "Increased CO concentrations and organic emissions containing carcinogenic PAH and nitro-PAil were found in mobile homes when kerosene heaters were used. TCO and GRAV organics were emitted at high concentrations from kerosene heaters in these homes. The results of chemical analysis of the semivolatile emissions from the kerosene heater suggested the presence of evaporated, unburned kerosene fuel. Four of eight homes had CO and/or PM~0 concentrations that exceeded the U.S. National Ambient Air Standards in some days when the kerosene heaters were on. Five homes had increased mutagenic activity when the kerosene heaters were in use. The results from this study indicated that kerosene heater emissions can significantly impact indoor air quality, and the organic emissions from the kerosene heater contained carcinogerfic compounds that can be potentially carcino- genic in humans. Further e~posure assessment and health effects studies on the emissions from unvented kerosene heaters in homes are needed." [15] "Cabin Air Quality and Headth Risks: A Review and Synthesis of the Available Data," L.C. Holcomb and W. Crawford, 1TA Magazine 66: 17-22, 1991 ['Vol. I, Issue 11, November 22, 1991] This review paper appeared in the magazine of the Institute of Air Transport in Paris. The authors propose that airline cabin air quality would be improved with increasing ventilation, because banning smoking does not remove other substances which influence air quality. The authors also report that ETS levels in airliner cabins are low. "Data on airline air quality is [sic] reviewed with particular reference to possible health risk. Cabin air quality is determined by a variety of physical and chemi- cal factors. Most chemical factors investigated easily meet existing or proposed standards; however, carbon dioxide levels indicate that cabin air can be stuffy and air quality could be improved by increased ventilation. The levels of environmental tobacco smoke components which have been monitored indicate that ETS is unlikely to pose a significant health risk to passengers or flight attendants." "Only the visible ETS and its odor would be removed by banning smoking leaving the invisible factors to affect comfort and health. When an effect or complaint is multifactorial the removal of but one factor of proven small concentration will have litxle effect on health. This could result in deterioration of air quality, for now that the only visible material is removed, airlines can easily continue to decrease fresh air intake." -71-

"Exposure to those components of ETS so far measured in airliners indicate levels which are low even in the smoking sections and in the boundary seats adjoining that section. The most exposed persons for consideration of possible long term effects are the flight attendants. Such staffmay be exposed for some 800 to 960 hours per year partly in the smoking section and mostly in the nonsmoking section particularly ifredrculation of cabin air is practiced and the quantity of fresh uncontaminated air is limited." "The very low levds of components of ETS monitored in airlines do not appear to pose a measurable risk to health of passengers or flight attendants if indeed a risk exists at all. Comfort of passengers and attendants can be maximized by assuring flesh air ventilation meeting recognized standards." "A valid sdentific study of the health and Ell health status of Right attendants is required to assess the value of the judgment to apply resMctive nonsmoking regulations." [16] =Indoor Sources of Mutagenic Aerosol Particulate Matter:. Smoking, Cooking and Incense Bttrning," G. L_fi'oth, C. Stensrnan and M. Br~ndhorst-Satzkorn, Mutation Research 261: 21-28, 1991 The authors examined the mutagenic activity of aerosol pa2"tides produced during several indoor activities. Extracts of particulate matter produced by dgarettes, incense and mosquito-coil burning, and frying food were all reportedly mutagenic in a bacterial assay. "The present study had the objective of determining the mutagenic activity of 2 Salmonella assays of experimen- tally produced aerosol particulate matter from some common and uncommon indoor activities. An additional aim was to compare the chemical nature of the mutagenic components by simple chemical fractionation. In addi- tion, it was feasible to determine the emission of carbon monoxide, isoprene and benzene. The common processes studied were cigarette smoking and frying of proteina- ceous foods, whereas the uncommon processes were the burning of incense and mosquito coils and the smoking of herbal dgarertes." "The emission of aerosol particles and their mutagenic activity as well as the emission of some gaseous pollutants has been studied experimentally in order to compare the emission from some indoor pyrolysis processes. Cigarette (tobacco and herbal) smoking, incense and mosquito-coil burning and frying of experimental lean minced pork emitted particulate matter. Their extracts were mutagenic in the Ames Salmonella test with [strain] TA98 and [metabolic] activation... The frying of some common food items following cookbook redpes also emitted mutagenic aerosol particles but the emitted activity was less than that in the pork experiment. Carbon monoxide, -72- isoprene and benzene were present in the emissions from the smoking and burning processes but were not detect- able in the frying fumes. The results suggest that incense and mosquito-coil burning can cause indoor air pollution akin to that from cigarette smoking. Indoor air pollution from cooking requires further study." [17] "Envlrortmenta/, Occupational, and Personal Factors Rehted to the P~zvalence of Sick Building Syndrome in the General PopM~don," D. Norback and C. Edlin~ British Journal of lndust-dal Medim'ne 48: 451-462, 1991 [Vol. I, Isstm 10, November 1, 1991] The authors report that data from questiormaires given to 633 Swedish adults suggest a rdationship between SBS symptoms and urban residency, VDU work, static dectric- ity, psychosocial work environment, exposure to fresh paint, and presence of preschool children at home. They also suggest that childhood exposure to maternal tobacco smoke and urban environments might "enhance" the prevalence of SBS symptoms. This study included 633 subjects drawn randomly from the populations of three counties in Sweden. Symptoms, personal factors, dwelling characteristics and ch~dhood exposures were assessed via questionnaire;, occupations were classified according to the Nordic Clamification of Occupa- tions with respect to seven classes of chemical exposure (inorganic dust, organic dust, organic solvents, irritating gases, combustion products, animal allergens and other chemical exposures). "The prevalence of SBS symptoms was in the range of 10- 30% for many common symptoms such as headache, fatigue, and upper airway irritation. Nausea and dermal symptoms were less common." "[The] results indicate a relation between SBS symptoms and urban residency, VDU work, static electridty, psychosodal work environment, exposure to fresh paint, and presence of'preschool children at home. Furthermore, childhood exposure to maternal tobacco smoke and urban environments might enhance the prevalence of SBS symp- toms. Also, a relation was shown between maternal smoking and atopy as well as allergy to nickel among adults." [18] "HVAC Retrofit for Healthy Schools," R.C. Thomp- son, G. Fisher, T. Brertmm, W.A. Turner and F. McKnight, presented at ASHRAE meeting, Indoor Air Quallty'91, September 1991 [Vol. I, Issue 8, October 1, 1991] [See Volume I, Issue 7 (September 13, 1991) of this Report for additional details on the EPA-ALA school survey.] A simple retrofit of the I-t-VAC system was reported I I I I ! I I I I ! I I ! i I i I l

I VOLUME I, ISSUES 1-24 JULY 1992 i I I I I I I ! I I I I I I I I I to be an effective and economical way to improve IAQin several schools in this study. "The Environmental Protection Agency has evaluated the impacts of HVAC systems and building dynamics on radon concentrations in 26 schools across the UMted States. Diagnostic data indicated that radon was not the only indoor air pollutant in these schools." "Elevated levels of CO= indicated that ventilation, and therefore probably the general indoor air quality, was inadequate. Because of this combination of indoor air problems and because the traditional radon mitigation practice would not benefit general indoor air quality, a new approach was recommended to the majority of the schools -- an HVAC retrofit." "lAin extensive diagnostic evaluation requiring between 32 and 40 person-hours was performed at each school. Diag- nostic procedures included plan evaluations, persormel interviews, and site walk-throughs. Diagnostic measure- merits included HVAC system parameters, subslab commu- nication testing, air pressure differentials on the building shell, building shell tighmess, and CO= concentrations throughout the school." "In general, the retrofit work was simple, requiring the repair or replacement of mechanical components such as damper actuators and linkages, the replacement of inopera- tive sensors and control components, and modification of control logic and serpoints." "For these schools, an HVAC retrofit was a very effective and economical means for improving the IAQ. When compared to current radon remediation practice, an HVAC retrofit has an added benefit -- improvement of overall IAQ." [19] "Diagnostics and Remediation for Healthy Schools," W.A. Turner, F. McKnight, G. Fisher, R. Thompson and T. Brennan, presented at ASHRAE meeting, Indoor Air '91, September 1991 [Vol. I, Issue 8, October 1, 1991] [See Volume I, Issue 7 (September 13, 1991) of this Report for additional details on the EPA-ALA school survey.] Although this study set out to investigate radon levels in schools, elevated CO= levels were also found, suggesting the necessity of a more general approach to improving indoor air quality. "The U.S. Environmental Protection Agency School Evaluation Program (SEP) was originally conceived in the summer of 1989 as a technical assistance program in response to an emerging need for information on diagnostic and rnJtigation strategies applicable in schools with devated radon." "The findings of both elevated carbon dioxide levels and devated radon levels in the majority of the 29 schools evaluated suggest that there is a need for a more holistic (building systems) approach to radon control and general indoor air quality in schools." "The evaluations performed on all 29 schools have, to date, strengthened the investigators' opinion that investi- gations of schools have two unique characteristics when compared to residential radon investigations: * Schools are complicated to diagnose and " Schools require that all components affecting the building operating dynamics be weighed in selecting the optimum control strategy." [20] "Passive Cigarette Smoke Increases Working Level Exposures and Lung Cancer Risk to All Occupants of the Smoker's Home," R.H. Johnson, Jr. and E. Geiger, presented at the 36th annual meeting of the Health Physics Society, June 21-26, 1991, Washington, D.C. [Vol. I, Issue 8, October I, 1991] This meeting abstract reporm on a study which claimed to have measured devated radon levels following dgarette smoking. The authors propose that nonsmokers are at increased risk from radon if ETS is present. "[V]ery little attention has been given to the effects of passive dgarette smoke and radon decay product exposures to nonsmokers. Preliminary studies [by these authors] showed that even a single dgarette drastically increased the Working Lewd exposures in homes. Consequently, a dgarette smoker not only increases his/her own risk, but may also increase the risk to all occupants of the same dwelling due to an increase in Working Level exposures." "This paper presents the results of additional measure- ments to evaluate the effects of typicaI cigarette smoking patterns in a home. The studies simulated the smoking habits of a one pack a day smoker. Continuous measure- ments were made on radon gas levels, Working Levels, and corresponding equilibrium ratios. Working Levels were found to increase rapidly after lighting of dgarettes and to remain elevated for several hours. Cigarette smoke provides a significant source of aerosols for attachment of radon decay products in homes. Furthermore, the airborne particles from dgarerte smoke remain in the air for many hours after the visible smoke has dissipated. Therefore, the increase in Working Level exposures and equilibrium ratios persists long after the smoking stops." - 73-

ETR AND IAQ_P,~FER_ENCE JULY 1992 "This paper confirms the potential risks to nonsmokers from increases in Working Levds due to passive smoke in homes and points to needs for further studies to docu- ment other risk factors." [21] "Airborne Glucan and Endotoxln in Sick Buildings," R. Ryiander, C/AR Current~ 1 (1): 34, 1991 [Vol. I, Issue 7, September 13, 1991] Data from an ongoing study in Sweden, presented in the first issue of the Center for Indoor Air Research newsletter, suggest that indoor levels ofendotoxin and glucan may be related to reports of symptoms assodated with sick build- ings. Further research will include challenge tests using aerosols of these substances, which are of microbial origin. "Our hypothesis for this epidemiological research project is that endotoxin and glucan [substances produced by microorganisms] are related to symptoms reported in sick buildings." Several buildings in Stockholm and Gothenburg, Sweden, where people had been reporting symptoms, were identified: two day care centers, one post office and one secondary school. An office building where no complaints were reported was a control. Air samples were collected for measuring amounts of endotoxin and glucan. Symptoms were evaluated using a questionnaire, with results presented as the percentage of persons • reporting symptoms each day. Higher percentages of symptoms were reported in the buildings with higher measured levels of airborne glucan. "[S]tudies were performed in individual houses with and without complaints .... The results show that the amount ofglucan in homes with symptoms was 0.75 nanogram/m~ as compared to 0.05 nanogram]m~ in houses without symptoms." Further studies will investi- gate six day care centers in Gothenburg and a secondary school in Sundsvall, Sweden. "The air sampling technique itself is also evaluated. F_.,xperience from earlier studies demonstrated that glucan was not detected in some localities where high amounts of glucan were present. A methodological evaluation was performed in an empty flat where the person living in the flat had experienced symptoms. The results show that continuous activity needs to be present during measuring and also suggest that a large proportion of the parr.ides aze not respirable." Challenge experiments using aerosolized glucan and endotoxin are also planned. [22] '*Sources of Polyn udear Aromatic Hydrocarbons (PAH) in the Indoor Air ofHong Kong Homes," L.C. Koo, J.H.-C. Ho, H. Matsushita, H. Shirnlzu, T. Mori, H. Matsuld and S. Tominaga, Meeting Abstract, Indoor Air International, "Priorities for Indoor Air Research and Action," Montreux, Switzerland, May 29-June 1, 1991 [Vol. I, Issue 6, August 27, 1991] This pilot study analyzed household dust for PAH levels. Increased PAH levels were associated with activities such as cooking and incense burning. Interestingly, PAH levels were also higher when windows and doors were opened. Cigarettes were described as a "minor contribu- tor~ to PAH levels. The authors suggested that not only sources but also the occupants' behaviors must be considered in assessing indoor air quality. "As a pilot study to investigate the cardnogenidty of household dust in urban homes characterized by high rise buildings in dose proximity, the kitchen and living rooms of 33 Hong Kong homes were monitored by an air sampler.., the collected dust was analyzed by HPLC for 7 PAH compounds.., and then summed for a total PAH count..." "The PAH levels in kitchens increased with cooking activities, incense burning, and smoke from neighbours. The we of gas water heaters and ventilating fans was associated with lower PAH levels, whereas it was higher when windows were opened. Incense burning was the major source of PAH in living rooms; dgarettes being a minor contributor. Homes with opened windows and doors had higher PAil levels. These data indicate that assessments of indoor air quality in homes must investi- gate emission sources as well as compensation behaviors, as the latter may override the effects of the former." I I ! ! [23] "Indoor Dust Exposure: An Unnoticed Aspect of • Involuntary Smoking," H.O. Hein, P. Saudicani, P. Skov and F. Gyntelberg, Archives of Environ- • mental Health 46(2): 98-101, 1991 [Vol. I, Issue 2, May 17, 1991] Nicotine concentrations in samples of house dust from • the homes of 34 smokers and 38 nonsmokers were determined by agc method. A strong positive correlation (r = 0.65, p < 0.0001) between the amount smoked and the nicotine concentration was found when the data from all homes was analyzed, and a fairly strong correlation (r = 0.35, p < 0.05) was found for smokers' homes only. Estimates of the amount of nicotine inhaled during 1 h in ~ i homes containing high concentrations of nicotine in ~-\~ house dust led to the conclusion '.. that howe dust inhalation constitutes only a modest addition~ source of ~" • involuntalT smoking.' ~-~ "[A] nonsmoker may inhale tobacco constituents even if smoking does not occur."

"VOLUME I, ISSUES 1-24 JULY 1992 I ! I I I I I I I I I I I I I I I I "The dust collected in vacuum cleaners consists of both respirable and nonrespirable dust. We were unable to separate these two components and determine [the] nicotine concentration in both types of dust." 'Wacuum cleaning was not conducted in a standardized manner; i.e., participants used various types of vacuum cleaners. This increased the variability of the study results and may have weakened the assodation found between the amount of smoking reported and nicotine concentra- tion in the house dust." SMOKING POLICIES AND RELATED iSSUES [1] "Discomfort From Environmental Tobacco Smoke Among Employees at Worksites with Minimal Smoking Restrictions m United States, 1988," Morbidity and Monality Weekly Report 41 (20): 351-354, 1992 [Issue 24, Item 42] The results of a 1988 survey of 114.1 million employed adults were published in this 1992 article. Of 28.5 million persons working in places where smoking is permitted, 43.5 percent reported "some or moderate discomfort" and 15.7 percent reported "great discomfort" claimed to be related to ETS. Discomfort was more likely to be claimed by never smokers, younger individuals, women, white-collar workers and workers with more education. In another publication, this study's conclu- sions were used to support a position that "Federal regulation and workplace polities need to be imple- mented and enforced to reduce employee discomfort and potential exposure to cardnogens found in secondhand tobacco smoke." See Occupational Satiny and Health Reporter, June 3, 1992. EXCERPTS: "As pazt of the 1988 National Health Interview Survey- Occupational Health Supplement (NHIS-OHS), CDC measured the degree of discomfort caused by ETS in the workplace. The NHIS-OHS collected information on cigarette smoking, workplace smoking restrictions, and perceived discomfort caused by ETS at the workplace. This report summarizes survey findings and describes efforts to reduce ETS at the workplace." "The survey asked the following question of employed respondents (i.e., persons who reported they had a job during a 2-week period immediately before being interviewed): 'Is smoking allowed in your place of work other than in designated areas?' Respondents who reported that smoking was allowed in designated (if any) and other areas were asked: 'Do you find that cigarette smoke in the workplace causes you no discomfort, some discomfort, moderate discomfort, or great discomfort?'" "Based on the survey findings, among 114.1 million employed adults in 1988 (who reported that their workplace was not in their home), 40.3 percent worked in locations where smoking was allowed in designated (if any) and other areas. Among 79.2 million employed nonsmokers (former and never smokers) (who reported thek workplace was not in their home), 28.5 million (36.5 percent) worked at places that permitted smoking in designated (if any) and other areas. Of these, 12.4 million (43.5 percent) reported some or moderate discomfort and 4.5 million (15.7 percent) reporraxt great discornfor~ from ETS at the workplace. Of 16.7 million current smokers, 2.5 million (15.0 percent) reported at least some degree of discomfort from ETS at the workplace." "[W]orkplace ETS exposure was more likely to be reported as a cause of discomfort by never smokers (63.6 percent) than by former smokers (5 t.4 percent) and by women (69.0 percent) than by men (53.9 percent). Nonsmokers in younger age categories were more likely than older nonsmokers to report discomfort from ETS. Prevalence of any discomfort was generally similar by race and ethnicity. The likelihood of any discomfort from ETS increased directly by level of education, from 44.1 percent among nonsmokers with fewer than 12 years of education to 69.6 percent among college graduates. Reported discomfort was more prevalent among non- smoking white-collar workers (65.1 percent) and persons in service occupations (62.5 percent) than among nonsmoking blue-collar workers (50.0 percent) and persons in agricultural/fishing occupations (44.0 per- cent)o~ "Two important considerations influence interpretation of the findings in this report. First, because this survey included only employees for whom smoking was permitted in the workplace in both designated (if any) and or_her areas, the results probably underestimate the number of U.S. non- smokers in 1988 who experienced discomfort from ETS at the workplace (i.e., employees who experienced discom- fort from ETS despite more restrictive worksite smoking policies were not included in this survey). Second, these findings are based on self-reported data and perceptions of discomfort have not been validated, even though self- reported workplace exposures of nonsmokers has been validated biochemically." "Two national health objectives for the year 2000 include efforts to prohibit or severdy restrict smoking at work. The first is to increase to at least 75 percent the proportion of worksites that have a formal smoking policy that prohibits or severely restricts smoking at the workplace. The second is to ena~ in the 50 states comprehensive laws on dean indoor air that prohibit or strictly limit smoking in the workplace and endosed public places (e.g., health-care fadlities, schools, and public transportation)." -75-

ETS AND IAQ_ REFERENCE JULY 1992 "The Environmental Protection Agency is reviewing the health effects of ETS exposure, and OSHA is considering regulatory options regarding indoor environmental quality. Enacting and adhering to workplace policies and regulations regarding worksite exposure to ETS can reduce employee discomfort and the exposure to carcino- gens and other toxic substances from ETS." [2] "The Effects of Ordinances Requiring Smoke Free Restaurants on Restaurant Sales in California," Sak. Glantz and L.R.A. Smith, Monograph Series, Institute for Health Policy Studies -- University of California, San Francisco, March 1992, 21 pages [Vol. I, Issue 20, April 24, 1992] The authors compared data on taxable restaurant sales in four California communities which prohibited smok- Lag in restaurants by ordinance to data from communities where no such ordinances were in effect. The authors concluded that a 100% smoke free restaurant ordinance had no significant effect on total restaurant sales and was associated with a small, but statistically significant, increase in the fraction of total retail sales that went to restaurants. A March 23, 1992, press release from Joan Luther and Associates in Beverly Hills, California, however, claimed that the study's conclusion was invalid because the authors had transposed numbers in the sales data utilized in the study. A reanalysis of the data was reportedly conducted by Louis Masotti, a professor at Northwestern University, who concluded that restaurants in one community, San Luis Obispo, actually lost 26.5% of their business after a restaurant smoking ordinance went into effect. "As the evidence that environmental tobacco smoke endangers nonsmokers has accumulated, more and more communities have restricted or eliminated smoking in public places and work places. Recently, several commu- nities have enacted legislation mandating 100% smoke free restaurants .... Such legislation, however, is not in the interests of the tobacco industry because creation of smoke free restaurants is a highly visible statement that tobacco use is not longer socially acceptable .... The industry uses.., the argument that requiring smoke free restaurants will be bad for business." "There is now enough information available to conduct an objective econometric analysis of the effects of smoke free restaurant ordinances. In addition to Beverly Hills, the California cities of Bellfl0wer, Lodi, and San Luis Obispo have had such 100% smoke free restaurant ordinances in force long enough to assess their effects based on reported taxable sales data available from the California State Board of Equalization. Analysis of total restaurant sales, individual categories of restaurant sales, restaurant sales as a fraction of total retail sales, and restaurants sales in cities with 100% smoke free restaurant ordinances compared to similar cities which do not have ordinances show no negative effects on business. If .anything, smoke free restaurants are slightly beneficial for business by increasing the fraction of total retail sales that go to restaurants." "Smoke free ordinances have no statistically significant effect on total restaurant sales or total restaurant sales in dries with ordinances compared to cities without ordi- nances. There is some evidence that restaurants took a greater share oftovat retail sales in Bellflower after the smoke free restaurant ordinance went into force." [3] "A Conceptual Framework for the Roles of Legislation and Education in Reducing Exposure to Environmental Tobacco Smoke," L.L. Pederson, J.M. Wanklin, S.B. Bull and M.J. Ashley, Amer/- can Journal of Health Promotion 6(5): 105-111, 1991 [Vol. I, Isstm 17, March 6, 1992] The authors prepared a conceptual model incorporating factors that may affect smokers' compliance with regula- tions on smoking in public places, based on the authors' research and a review of the literature. The authors suggested that compliance with regulation may be directly affected by education about the purported health effects of ETS and by consideration of personal attitudes toward legislative measures regulating smoking. "Legislation regulating smoking in public places is being enacted by various levels of government because of concern about the health effects of environmental tobacco smoke (ETS). As a result, increasingly fewer areas remain where smokers are free to smoke as they choose." "[A] conceptual framework incorporating a number of factors that may affect personal compliance with restric- tions is proposed. This model could guide the behavior of public policy makers and educators." "Many factors can be posited to influence individual attitudes toward regulations on smoking. These include knowledge of the effects of exposure to ETS, parameters of smoking such as amount and duration of smoking and degree of dependence, awareness of existing legislation regulating smoking, norms of the society in which the smoker lives, the smoker's philosophical position regard- ing freedom of choice and social responsibility, personal- I I I I I I I I I I I I I I i I I l

I VOLUME I, ISSUES 1-24 JULY 1992 I I I I l I I I I I I I I I i I I ity characteristics, and, finally, past experiences with restrictions on smoking." "One major implication of the postulated modal is that legislative measures should be enacted incrementally based on aoceptability... Initially, measures could be enacted that are relatively moderate in their impact. Once smokers have experience with them and find them acceptable, more slM.ngent legislation could be put into place." "['1] t is important to gather information from smokers about the acceptability of the regulations and where they are found to be unacceptable to determine how they could be made more palatable." "One way to achieve greater compliance is to educate smokers about the detrimental health effects of exposure to ETS and, thus, provide a rationale for policy changes. If smokers are not willing to quit for their own health, as evidenced by their continued smoking, it may be that an emphasis on the health of other will affect their behavior." [4] "Employee Knowledge and Attitudes About a Work-Site Nonsmoking Policy: Rationale for Further Smoking Restrictions," G. Sorensen, Nak. Rigotti, A. Rosen, J. Pinney and R. Prible, Journal of Occupational Medicine 33(11): 1125-1130, 1991 [Vol. I, Issu~ 17, March 6, 1992] This study investigated employee awareness of and satisfaction with a n0nsmoking policy at the New England Telephone Company. The policy was instituted in 1986, and the employees were surveyed 20 months after the policy was implemented. The authors reported that awareness of the rules about smoking was generally quite high, as was employee satisfaction with the restric- tive policy. They claimed that a total ban on smoking might be easier to implement and more satisfactory to employees than are more limited policies. "On March 1, 1986, the New England Tdephone Company instituted a company-wide policy restricting smoking in all its work sites. The policy was announced in July 1985 and implemented in two phases. Begirming September 1, 1985, smoking was banned in conference rooms and classrooms and restricted to designated areas in cafeterias and lounges. Six months later, on March 1, 1986, smoking was prohibited in all work areas, even individual offices. Smoking areas were designated in cafeterias, lounges, and in some hallways and some restrooms." "Employees were surveyed in November 1987, 20 months after full implementation of the policy, to assess their attitudes toward the policy and the impact of the policy on their smoking behavior." "The 23-item questionnaires assessed demographic variables (age, sex, and job status), respondents' awareness of and satisfaction with the poLicy, perception of the policy's effect on air quality, and smoking behavior." "[A]wareness of the policy was highest among upper level managers and lowest among nonmanagers. Aware- ness of the rules about smoking was highest where the policy banned smoking totally, such as in conference rooms and classrooms .... Awareness of the policy was lowest about places where smoking was allowed only in designated areas, such as rest rooms, hallways, and lounges." "Over half the respondents were satisfied or very satisfied with the nonsmoking policy, and fewer than one fifth were not at all satisfied. Of those not at all satisfied with the policy, 57% preferred either more restrictions on smoking or a total ban, although 42% preferred fewer or not restrictions .... [S]mokers were least likely to be satisfied with the policy, but only one-quarter were entirely dissatisfied." "Although satisfaction with the current nonsmoking policy was generally high, when asked to describe the kind of policy they would like to see at the company, balf the respondents said that they would prefer a stricter nonsmoking policy at work.... [P]olicy preference varied by smoking status: current smokers were more likdy to prefer fewer or no restrictions on smoking compared with former smokers and never smokers. In contrast, nonsmokers were more likdy than smokers to prefer additional resMctions on smoking." "This study also suggests that work sites may find even more restrictions on smoking easily implemented. Rules about smoking are more easily understood where smoking is toradly banned, as exemplified here. Employees reported less exposure to environmental tobacco smoke where smoking was prohibited. Also, employee satisfaction with the smoking policy was highest among those reporting less exposure to smoke in their work areas. These findings are likely to be welcome results to work sites seeking to lower their workers' exposure to environmental tobacco smoke by adopting polities that ban or severely limit smoking." [5] "Work-Site Smoking Policies in Srrmll Businesses," G. Sorensen, A. Rosen, J. Pinney, J. Rudolph and N. Doyle, Journal of Occupational Medicine 33 (9): 980-984, 1991 [Vol. I, Isstm 17, March 6, 1992] In this study, 216 small businesses (i.e., 120 or fewer employees) were surveyed. Companies with smoking polides - 77-

were generally lager, reportedly had fewer smokers and more management support for smoking potides. Nearly three-fourths of the businesses without a policy said they would adopt one if required by legislation. EXCERPTS: "We examine the processes small businesses use to implement polities restricting smoking. There is a continuing national trend toward workplace nonsmoking restrictions. This trend was strengthened by recent reports on environmental tobacco smoke and its serious health effects on nonsmokers." "We report the findings of one of the first in-depth examinations of the experiences of small businesses in implementing policies restricting smoking. This descrip- tive study was designed to identify factors that may influence the processes of adopting and implementing nonsmoking policies in small work sites, to compare the characteristics of small businesses with and without formal policies restricting smoking, and to characterize the experiences and concerns of small businesses that have adopted smoking policies." "Small businesses that restricted smoking were identified in 13 states by contacting representatives of the American Lung Assodation and the American Cancer Sodety. Representatives were asked to identify work sites with 100 or fewer employees with which they had consulted on the development of smoking policies. Comparison work sites of the same size were identified from Chamber of Commerce listings from the same areas .... [T]elephone interviews were completed with 116 work sites with policies and 100 comparison work sites without polities." "Small businesses with polities had fewer smokers and greater management support for the smoking policy than did the work sites without policies restricting smoking... • We found few problems with policy implementation. Respondents reported that enforcement was highly effective overall, and that compliance with the policy was high. Both employees and management were generally very satisfied with the policy; only 4% of work sites surveyed reported that any employees had left the company because of the policy." "It was not the purpose of this study to estimate the prevalence of smoking policies in this population of work sites. The generalizability of the study findings is some- what limited by the method used to identify work sites; as a result of nonrandom sampling, the distribution of businesses by industry division is not representative of the national distribution," "The restriction of smoking in the workplace is a community-wide issue affecting the health of smokers and nonsmokers alike. Legislation requiring work sites -78- (regardless of size) to adopt smoking polities is a first step in this process .... Nearly three fourths of the small businesses we surveyed that did not restrict smoking said they would be very likely to adopt a policy if it were required by law, confirming the importance of such legislation." [6] "Occupational Risks Associated with Cigarette Smoking: A Prospective Study," J. Ryan, C. Zwerling and E.J. Orav, American Journal of Pub~c Health 82(1): 29-32, 1992 [Vol. I, Issue 16, Februaxy 17, 1992] This study, a survey of 2,537 postal employees, reported statistically significant elevated risks for accidents, injuries, and discipline problems among smokers on the job. The authors c/aim that their conclusions have relevance for companies formulating smoking policies and establishing smoking cessation programs. The conclusions of this study were reported in a recent issue of the BNA Occupational Safity and Health Reporter and could be raised in the context of the OSHA RFI, "We present a prospective examination of the assodation between dgarette smoking and rates of employee turnover, absenteeism, accidents, injuries, and discipline while controglng for age, gender, race, job category, exerdse habits, and drug abuse. It represents a semndary analysis of data collecr~ in a previous study of the predictive value of preemployment drug testing." [The sample consisted of 2,537 postal employees.] "Smoking status was categorized as smoker or non- smoker using self-reported data. Persons who had smoked three or more cigarettes per day over the previous month were classified as smokers. All other subjects were classi- fied as nonsmokers." "Self-reported information on smoking status was used because of its convenience, simplicity, economy, and reliability." "A number of confounding factors might be assodated with both the risk variable (cigarette smoking) and the outcome variables (turnover, absentedsm, accidents, injuries, and disdpllne). We therffore collected data on age, gender, race, job classification, drug use, and exercise habits." "Smokers were more frequently older, female, nonexerdsers, and drag users and less often Asian or letter carriers. Thus, these variables were potential confounders." "To our knowledge, this study presents the first prospec- tive estimates of assodations between dgarette smoking I I I I ! I I I I ! I

I JULY 1992 I I I I I I I I I I I I I I I I I ! and adverse employment outcomes after adjustment for the potentially confounding covariates of age, sex, race, exercise, drug abuse, and job category. We found signifi- cant positive associations between cigarette smoking and absenteeism, industrial accidents, occupational injuries, and disdplinary action. Smokers had a 34% increase in mean absenteeism that remained significant after adjust- ment for covariates. Smokers had increased covariate- adjusted relative risks for industrial accidents (29%), occupational injuries (40%), and disdpline (55%). We found no significant difference in turnover, voluntary or involuntary. Thus, our study confirms the previously reported elevated rates of absenteeism, acddents, and injuries for smokers." "In this study, we were unable to control for the possible confounding effect of alcoholism. Smoking is reported to be assodated with alcoholism." "... [I]t is possible that some of the assodations demon- strated are am'ibutable to alcoholism rather than smoking." "Our study cannot distinguish adverse employment outcomes assodated w~th dgarette smoking caused by the physiological effects of smoking from those caused by smoking behavior (e.g., decreased attention to driving) or by other sodal or personality characteristics assodated with smoking .... [O]ur findings provide a basis for estimating the magnitude of the increase in costs to industry that results from hiring dgarerte smokers." "The findings of this study may also be useful in assessing the cost-effectiveness of corporate smoking control and smoking cessation programs. To the degree that such programs decrease dgarette smoking in the work force, a health-related effect such as absentedsm might be expe~ed to decrease. It is less dear whether there would be similar improvements in effects such as accidents, iniuries, and disdpline, as these may be linked to social and behavioral characteristics that might persist after cessation of, or a decrease in, smoking." [7] "The Department of Veterans Affairs Smoke-Free Policy," A.M. Joseph and P.J. O'Neil, JAMA 267(1): 87-90, 1992 [Vol. I, Issue 14, January 17, 1992] This article describes the VA hospita] smoking policy, which prohibits smoking in all acute-care facilities. The authors invoke risks purportedly associated with ETS as a reason for developing such policies and as a tool in negotiat- ing with labor. "The Department of Veterans Affairs (VA), the largest health care provider in the United States, implemented an agency policy on January 7, 1991, requiting all 172 acute- care hospitals m be completdy smoke-free-" "This policy was written in a national climate that is increasingly intolerant of environmental tobacco smoke, a/though hospitals commonly continue to ~ow smoking in designated areas .... Because of the national trend in legislation of public smoking (including the Joint Commis- sion on Accreditation of Hea/thcare Organizations revised standard requiring anzredited hospitals to be smoke-free) and new sdentific evidence about the risks of exposure to environmenta/tobacco smoke (including the Environmental Protection Agency reo0rnmendation that mvimnmental tobacco smoke be classified as a known human carcinogen), hospitals in both the public and private sectors will have to implement smoke-flee policies." "The purpose of restrictive smoking polides in hospitals is to reduce patient and employee exposure m environmental tobacco smoke, espedally since the majority are not current 81Tloker$." "The national VA policy prohibits smoking by patients, visitors, or employees in all acute-care fadlities." "Patients are informed of the smoke-free policy prior to admission or dLrfic appointments by computer letzers, building signs, and in some cases overhead speaker systems." "AI/employees share responsibility for policy enforce- meD-t. "The VA smo "king policy permits smoking on the grounds of the hospital." "[M]anagement was required to provide shelters outside the building to protecx smokers from adverse weather condmons. "Unions are sometimes reluctant to accept polides because of thdr obligation to represent all members, including those who smoke. Emphasis on the dangers of environmenta/ tobacco smoke as opposed to the 'rights' of smokers may improve unions' acceptance of smoke-free initiatives." STATISTICS AND RISK ASSESSMENT [1] "Trust & Credibility in Risk Communication," V. Covello, Health & Environment Digesg April 1992 [Issue 24, Item 43] This article reported on research into effective risk communication, described as a complex skill requiring knowledge, training and practice. The author proposed that trust and credibility were key factors in successful risl~ communication. "Why so much interest in risk communication? One - 79-

E.TS .hND LKQREFERENCE JULY 1992 explanation is the increased number of hazard communica- tion and right-to-know laws rdating to chemical exposures. Another stems from increased public fear and concern about chemical exposures and a corresponding demand for risk information. But a third underlies the first two: the loss of mast in government and industry as trusted, credible sources ofinformalion on chemical risks." "Public distrust of government and industry is grounded in several beliefs: that government and industry have been insensitive to public concerns and fears about chemical risks, that they are unwilling to acknowledge problems, share information, or allow meaningful public partidpa- lion, and that they are negligent in fulfilling their environmental responsibilities." "Better risk communication calls for improvements in the credibility of individual spokespeople, in the credibil- ity of organizations with risk assessment and management responsibilities, and enhanced third-party support, defined as better collaboration with trusted institutions and individuals." "A common thread in all three strategies is the need to be proacfive in establishing high levels of trust and credibility. When these are missing or weak, the primary goal of risk communication is to build them up. Only when trust and credibility have been established, can other goals, such as education and the sharing of information, follow." "Risk communication is effective to the degree to which all actions and communications -- verbal and non-verbal J convey caring and empathy, competence and expertise, honesty and openness, and dedication and commitment." "These findings represent only a sampling of results from the emerging area of risk communication research. They suggest, on the one hand, that effective risk communication is a complex ar~ and skill that requires substantial knowledge, training, and practice. They also suggest that there are no easy prescriptions for effective risk communication; that there are limits on what can be accomplished through risk communication alone -- no matter how ski~ed, committed, and sincere an organization or person is. Final/y, they suggest that trust and credibRity are the key factors in successful risk oommunication." [2] "The Interplay of Science, Values, and Experiences Among Scientists Asked to Evaluate the Hazards of Dioxln, Radon, and Environmental Tobacco Smoke," G.L. Carlo, N.L. Lee, K.G. Sund and S.D. Pettygrove, Risk Analysis 12(1): 37-43, 1992 [Vol. I, Issue 19, April 10, 1992] This study investigated the extent to which personal values and experiences among sdenlists could affect their assess- -80- ment of purported risks from dioxin, radon and ETS. Partidpants in a telephone survey were read a vignette describing "mainstream sdentific thinking" on one of the three substances. For half the partidpants, the substance was named; for the other ball, it was described as substance X, Y, or Z. The authors concluded that when sdentists were aware of the substance they were evaluating, they were more likely to perceive a serious hazard than when the substance was not identified. The authors reported that these findings were strongest for ETS. "[W]e conducted an experiment to study the impact of knowing the name of an environmental substance on U.S. scientists' assessments of the substance's stares as an environ- mental health hazard." "To investigate the extent to which personal values and experiences among sdentists might affect thdr assessment of risk from dioxin, radon, and environmental tobacco smoke (ETS), we conducted an experiment through a tdephone survey of 1461 epidernlologists, toxicologists, physicians, and general sdentists. Each partidpant was read a vignette designed to reflect the mainstream sdentific thinking on one of the three substances. For half of the partidpants (group A) the substance was named. For the other half(group B), the substance was not named but was identified only as Sub- sv.ance X, Y, orZ." "The specific null hypothesis tested was that sdentists who were read a basic set of facts describing by name one of three well-known environmental substances would reach similar oondusions about the degree of environmental hazard and health risk posed by the substance as sdentists who were read the same set off-acts but not told the name of the substance." "Knowing the substance had little effect on the sdenlists' evaluation ofdioxin. Those who knew and who did not know the identity of the substance were similar in rating dioxin as an environmental health hazard, rating background exposures as requiring public health intervention and rating above-backq~und exposures as requiring public health intervention. Those who knew the substance to be dioxin were more llkdy to rate the substance as a serious environ- mental health hazard (51% vs. 42%)." "For radon, those who knew the substance by name were significantly more likdy to consider it an environmental health hazard than were those who knew it as substance Z (91% vs. 78%). However, those who knew the identify of the substance and those who did not were similar in rating radon as a serious environmental health hazard and in deriding whether background or above-background exposures required public health intervention." o!

VOLUME I, ISSUES 1-24 JULY 1992 I I I I I I I I I l I I I I I I I "Those who knew they were being asked about ETS were significandy more likely to consider the substance an environmental health hazard (88% vs. 66%), significandy more likely to consider the substance a serious environmen- tal health hazard (70% vs. 33%), significandy more likely to believe that background exposure required public health intervention (85% vs. 41%), and significantly more likely to believe that above-background exposure required public health intea'vention (90% vs. 74%)." "The results of this study suggest the presence of a signifi- cant interplay of scientific facts, values, and experiences among the sdentists asked to evaluate the hazards ofdioxin, radon, and ETS." "Our data show that when scientists were aware they were evaluating dioxin, radon, or ETS, they were more likely to perceive a serious environmental health hazard than when they were read the same facts without the substance being identified. Conversely, when the substances were not identified, mos~ sdentists did not perceive the substance to be a serious environmental health hazard. The findings were strongest for ETS." "One possible hypothesis for the strength of the ETS results lies in the nature of the risk from ETS .... Individuals are more likely to describe a substance or an activity as risky if the hazard is new, if it may lead to severe consequences, and if it appears as if it could be easily controlled. ETS might fit this description more dosdy than dioxin or radon." "A second possible hypothesis for the strength of the ETS • results is the visibility of the issue of dgarette smoking." [3] "Can Meta-Analyses Be Trusted?," S.G. Thomp- son and S.J. Pocock, The Larwet 338:1127-1130, 1991 [Vol. I, Issue 13, January 3, 1992] In a brief overview of meta-analysis of clinical trials, the authors note that potential biases are often "more problematic" in met,a-analyses of epidemiologic studies. In particular, they note the problem of bias when distin- guishing between a small effect and no effect. "Although we concentrate on clinical trials, similar (and often more serious) issues are relevant to meta-analysis of observational studies." "The value of any meta-analysis is totally dependent on lack of bias in its component studies .... In meta-analyses of observational studies, potential biases through con- founding, misdassification, or other causes are often more problematic, especially when distinguishing between a small effect and no effect .... The general purpose of meta-analysis -- to obtain an objective summary of the evidence available J is better served by defining before- hand acceptable standards for the inclusion of individual studies." "Meta-analysis has the potential to remove idiosyncrasy from the evaluation of medical issues but it is unrealistic to imagine that it will produce simple statistical answers to complex clinical problems. A meta-analysis may provide conclusions about a treatment effect that could not be drawn from individual trials because of small numbers. It may provide evidence about a class of drugs or treatments which allow a general qualitative contusion to be drawn. Its results are therefore directly relevant to the formulation of broad medical polities. Meta-analysis cannot tell clinicians how to treat an individual patient but it can provide information that helps decision- making." "The problems we have discussed make it unreasonable to interpret simplistically the numerical result a meta- analysis yidds. Meta-analysis is a most valuable objective descriptive technique which often provides clear-cut qualitative conclusions. Quantitative conclusions require more care and must take into account the practical relevance of the individual studies making up the meta- analysis and the clinical heterogeneity between them." [4] "A Framework for Risk Characterization of Environmental Pollutants," D.F. Naugle and T.K. Pierson, Journal of the Air and Waste Management Association 41(10): 1298-1307, 1991 [Vol. I, Issue 12, December 10, 1991] The authors present their Risk Characterization Frame- work, designed to encourage more systematic analysis and presentation of risk estimates. "Risk characterization is deaqned by both the U.S. National Academy of Sciences and the U.S. EPA as the estimation of human health risk due to harmful (i.e., toxic or cardnogenic) substances or organisms. Risk characterization studies are accomplished by integrating quantitative exposure estimates and dose-response relationships with the qualitative results of hazard identification." "A Risk Characterization Framework has been devel- oped to encourage a systematic approach for analysis and presentation of risk estimates. This methodoIogy subdi- vides the four common components of the risk assessment process into ten elements. Each of these elements is based -81-

JULY 1992 on a term in a predictive risk equation. The equation allows independent computations of exposure, dose, lifetime individual risk, and risk to affected populations. All key assumptions in the predictive risk equation can be explidtly shown. This is important to understand the basis and inherent uncertainties of the risk estimation process." "The systematic treatment of each of the ten dements in this framework aids in the difficult job of comparing risk estimates by different researchers using different method- ologies. The Risk Characterization Framework has been applied to various indoor and outdoor air pollutants of a cardnogenic nature. With further development, it aiso promises to be applicable to noncardnogenic effects." [5] "Meta-Analysis: A Review," R. Rosenthal, Psychosomatic Medicine 53:247-271, 1991 [Vol. I, Issue 10, November 1, 1991] The author proposes to "present the concepts and procedures that are likely to improve the process of the cumulation of evidence in the behavioral and biomedical sciences." Most of the article consists of statistical methodology for meta-analysis, but the author also reports on an experiment comparing the traditional literature review and meta-analysis for summarizing the results of a group of studies. "The most general purpose of this review article is to present the concepts and procedures that are likely to improve the process of the cumulation of evidence in the behavioral and biomedical sciences." Most of the article consists of statistical methodology for meta-analysis, focusing on the issues of defining study results, conceptu- alizing the quantitative summary of research, illustrating such quantitative procedures, reviewing concepts and procedures for dealing with the "file drawer" problem (research that is not available for review), and evaluating the importance of effect sizes. The author reports on an experiment "conducted to assess empiricatl/y the effects of employing meta-analytic procedures on the conclusions drawn by investigators." Forty-one participants were asked to review seven studies on sex difference in task persistence, described by the author as supporting the hypothesis of greater task persistence in males. Some pamidpants reviewed the studies using meta-analysis; others used the traditional literature review. While 73% of the "traditional review- ers" found no support for the hypothesis of a relationship between sex and task persistence (i.e., a conclusion contrary to the author's expectations), only 32% of the -82- meta-analytic reviewers concluded that there was no support. [6] "The Potential and Limitations of Meta-Analysis," T.D. Spector and S.G. Thompson, Journal of Epidemiology and Community Health 45: 89-92, 1991 [Vol. I, Issue 6, August 27, 1991] The authors support the use of meta-analysis with epidemiologic studies, but stress that meta-analysis must be conducted according to rigorous standards. These include criteria for locating studies, including them in the meta-analysis, considering study quality and accounting for publication bias. "It has been suggested by some authors that on~ randomised controlled trials should be subjected to meta- analysis. However this restriction is not desirable... In observational epidemiology, potential bias in individual studies (through confounding, mJsdassification, or other causes) will always remain a problem, especially when effect sizes are small. If such biases are to an extent consistent over different studies, a meta-analysis will reflect both the true effect and the biases. However the increasing use of meta-analysis in observational studies should encourage the more formal reporting of aedological studies, to facilitate the combining of such resLLIts.~ "Although the vast majority of meta-analyses concern the assessment of therapies in randomized control trials, a few studies have addressed contentious aetiological issues such as the quantification of the effect of passive smoking on the risk of lung cancer..." "With the proliferation ofmera-analyses, it has become apparent that their design, methods and publications should be conducted in a rigorous scientific manner, akin to that currently expected of randomised controlled trials .... A meta-analysis should be a research study in its own right." The authors recommend that a thorough search for publications be performed and that the inclusion of studies be based on predetermined criteria. They also write, "quality assessments have also been used in epide- miological studies. The major problem with quality weighting is that it must remain arbitrary and to an extent subjective .... Moreover the procedure goes against the general purpose oiC meta-analysis, that is to obtain an objective summary of the available evidence." "The question of publication bias needs to be addressed in all meta-analyses and its importance considered." I I I I I I I I I !

VOLUME 1, ISSUES 1-24 JULY 1992 I I I I I I I I I I I I I I I I I ! "Meta-analysis is here to stay. Epidemiologists, statisti- cians and clinicians should all be aware of the uses and limitations of the technique .... Despite the potential problems and pitfalls we have outlined, meta-analysis should play a leading role in the review of scientific issues. This necessitates a fuller understanding of meta-analysis as a routine analytical tool, but also a wider appreciation of the issues involved." [7] "Meta-Analysis in Epidemiology, with Special Reference to Studies of the Association Between Exposure to Environmental Tobacco Smoke and Lung Cancer: A Critique," J.L. Fleiss and A.J. Gross, Journal of Clinical Epidemiology 44 (2): 127- 139, 1991 [Vol. I, Issue 1, April 30, 1991] This paper acknowledges support from the Tobacco Institute and is largely based on their written comments submitted to EPA on the ETS risk assessment. The paper contains discussion of the use and misuse of meta-analysis in clinical trials and epidemiological studies and concludes that properly performed meta-analyses can be effectively used to (1) increase statistical power; (2) help make sense out of studies with conflicting conclu- sions; and (3) improve estimates of sizes of effects. The authors are highly critical of the meta-analysis done in the 1986 National Research Coundl (NRC) report on ETS: "It is very unlikely that the biases present in the epide- miological studies of the possible assodarion between exposure to ETS and the risk of lung cancer can ever be removed. The meta-analysis performed by the NRC must tither be completely discounted or... considered a mere 'computational exerdse.'" -83-

VOI_10M.E 1, 7tSSIOF~S 1-24 JULY 1992 I i I I I i I I I I I I I I I I I INDEX LUNG CANCER [1] Letters to the Editor Regarding "Lung Cancer in Nonsmoking Women: A Multicenter Case-Control Study," E.T.H. Font.ham, P. Correa, A. Wu-\Villia_m.s, p. Reynolds, R.S. Greenberg, P,A. ButTler, V.W. Chert, P. Boyd, T. Alterman, D.F. Austin, J. Liffa.nd S.D. Greenberg, CancerEpidemiology, Biornarkem 2B-evention 1: 35-43, 1991 [Issue 23, Item 17] .................................................................................................... .......... 1 [2] "Passive Smoking and Canine Lung Cancer Risk," J.S. Reif, K. Dunn, G.K. Ogilvie and C.K. Harris, American Jour~alof~-taide~iolo~y 135(3): 234-239, I992 [Vol. I, Issue 19, April 10, 1992] ........................................................ 2 [3] "Lung Cancer in Nonsmoking Women: Dietary Antioxidants," F- Fontham, tL. Coates, A. Dilley, P. Reynolds, P.A. Buffler, A. Wu-Williawns, V. Chert, R. Greenberg, P. Boyd, T. Alterrrmn, D.F. Austin and P. Correa, ~tbstract submitted for American Society of'Preventive Oncology meeting, CancerEioidemlology, Biomarkers I~revention 1(3): 429, 1992 [Vol. I, Issue 19, April 10, 1992] ....................................................................................... 2 [4] "Lung Cancer in Women in the Niagara Region, Ontario: A Case-Control Study," E.J. Holowaty, H.A. Risch, A.B. Miller and J.D. Burch, Ca*,adi~nJournaloflaublicHealth82: 304-309, 1991 [Vol, I, Issue 18, March 20, 1992] .............. 3 [5] "Lung Cancer in Nonsmoklng Women: A Multicenter Case-Control Study," E.T.H. Fontham, P. Correa, A. Wu- Williams, P. Reynolds, R.S. Greenberg, P.A~ Buffler, V.W. Chen, P. Boyd, T..~Aterma.n, D.F. Austin, J. Liffa.nd S.D. Greenberg, CancerEioldemiology, t3iornarkers &l-~rewent~on 1 : 35-43, 1991 [Vol. I, Issue 12, December 10, 1991 ] ............. 3 [6] "Passive Smoking and Cancer Risk: The Nature and Uses of EpidemiologicaI Evidence," A. Woodward and A.J. McMichael, European Jour'aal of Cancer 27(11): 1472-1479, 1991 [Vol. I, Issue 12, December 10, 1991] .................. 4 [7] "Is Passive Smoking in the Workplace Hazardous to Health?" A. Woodward, Scandinavian Journal of Work and EnvironrnentalHealth 17: 293-301, 1991 ['Vol. I, Issue 11, November 22, 1991] ........................................................ 5 [8] "Smoking and Health: A R.eview Prepared By the Smoking and Health Subcommittee of the Tobacco Industries Council," a Council Formed By the Minister of Finance of'Japan, H. Kasuga, H. Katsuki, O. Miyagi, W. Mori, S. Takayarna and T. ganagita, The Intewna~onalJournal of the Addicrions 26(4): 423-440, 1991 [Vol. I, Issue 11, November 22, 1991] .................................................................................................... ..................... 5 [9] "An Epidemiologicai Study of'Lung Cancer in Xuan Wei County, China: Current Progress. Case-control Study on Lung Cancer and Cooking Fuel," X. He, W. Chen, Z. Liu and R.S. Chapman, Environmental Health I~ere~ec~ves94: 9-13, 1991 [Vol. I, Issue 9, October 17, 1991] ......................................................................... 6 [t 0] "Mainstream and Environmental Tobacco Smoke," G.B. Gori and N. Manrel, Regu/atory Toxicology and 2~harrnacology 14: 88-105, 1991 [Vol. I, Issue 6, August 27, 1991] ............................................................................... [11] '%'~Teaknesses in Recent Risk Assessments of Environmental Tobacco Smoke," P,N. Lee, Environmental Tecbnolog7 12(3): 193-208, 1991 [Vol. I, Issue 6, August 27, 1991] .............................................................................................. 7 [12] "'Passive Smoking' and Lung Cancer: A Critical Analysis," G. Feuer and D.J. Ecobichon, Modern Medicine of Ca~_a-4_~ 46(4): 26-29, 1991 [Vol. I, Issue 6, August 27, 1991] ..................................................................................... 8 [13] ~Environmental Tobacco Smoke and Lung Cancer in Never Smokers," H.G. Stock'well, E.C. Candelora, A.W. Armstrong and P.A. Pinkham, Meeting Abstract, Society for Epiderniologic Research Conflrence, June 11-14, 1991 [Vol. I, Issue 5, August 9, 1991] .................................................................................................... .............................. 8 [14] "Lung Cancer Risk Associated With Cancer in Relatives," G.L. Shaw, R.T. Faik, L.W. Pickle, T.J. Mason and P.A~ Buffler, .[ournal of Clinical Eioidernlology 44(4- 5): 429-437, 1991 [Vol. I, Issue 4, July 12, 1991] ......................... 9 [15] "Dietary Patterns of Female Nonsmokers With and Without Exposure to Environmental Tobacco Smoke," L. Le Ma.rchand, L~R. Wilkens, J.H. Hankin and N.J. Haley, Cancer Causes and Control 2:11 - 16, 1991 ['Vol. I, Issue 4, July 12, 1991] .................................................................................................... ................................. 9 [16] "Serum Beta-Carotene tn Persons With Cancer and Their Immediate Families," A.H. Smith and K.D. Wailer, ArnericanJournatof2:.pidamiology 133(7): 661-671, 1991 [Mol. I, Issue 4, July 12, 1991] ........................................... 10 [17] "Diet and Lung Cancer in California Seventh-Day Adventists," G.E. Fraser, W.L. Beeson and R.L. PhiLlips, American Journal of Eioidemiology 133(7): 683-693, 1991 [Vol. I, Issue 4, July 12, 1991] ........................................... 10 [18] "Smoking and Other Risk Factors for Lung Cancer in Xuanwei, China," Z. Liu, X. He and R.S. Chapman, International.[ournalofEpid~miology20(1): 26-31, 1991 [Vol. I, Issue 2, May 17, 1991] .......................................... 10 [19] "Comparative Epidemiology of Cancer Bet~'een the United States and Japan,~ E.L. Wynder, Y. Fujita, I~E. Harris, T. Hirayama and T. Hiyaana, Cancer67: 746-763, 1991 [Vol. I, Issue 2, May 17, 1991] ...................... 11 [20] "Passive Smoking and Diet in the Etiology, of Lung Cancer Among Non-Smokers," A. KaIandidi, K. Katsouyanni, N. Voropoulou, G. Bastas, I~ Saracci and D. Trichopoulos, Cancer Causes and Control 1 : 15-21, 1990 [VoL I, Issue 1, April 30, 1991] .................................................................................................... .............................. 11 -85-

ET~ A~D EA.Q.g~EFEP,.ENCE JULY 1992 C_ .~-~,DI OVAS CULAR ISSUES [1] Letters to the Editor Regarding "An Estimate of Adult Mortgtity in the United States from Passive Smoking," A.J. WelLs, Environment internatfonM 14: 249-265, 1988 [Issue 24, Item 36] ............................................................ 11 [2] "Serum Lipoproteins in Nonsmokers Chronically Exposed to Tobacco Smoke in the Workplace," J.IL White) M. Criqui, J'.A. Kulik, H.F. Froeb and P.J. Sinsheimer, Abstract No. 383, 27ighth V~'or/d Conference on Tobacco or Health: Building a Tobacco-Free World~ March 30-April 3, 1992, Buenos Aires, Argentina [VoL I, Issue 20, April 24, 1992] .................................................................................................... ............................ 12 [3] "Passive Smoking and Your Heart," G.L. Huber, R.F~ Brockie and V.K. Ma.hajan, Consumers' Research, April 1992, pp. 13-19, 32-33 [VoL I, Issue 19, April 10, 1992] ................................................................................. 12 [4] ~Passlve Smoking and Carotid Artery Wall Thickness: The A.P,2C Study,"G. Howard, M. Szklo, G. Evans, G. Tell, J. Eckfeldt and G. Heiss, Cardiovascular Disease Epidemiology Abstracts, printed at Annual American Heart Association meeting, March 1992 [Vol. I, Issue 19, April 10, 1992] ................................................ 13 [5] "Indoor Passive Smoking: Its Effects on Cardiac Performance," A. Leone, L. Mori, F. Bertinelli, P. Fahiano and M. Filippelli, InternationalJournal of Cardiology 33(2): 247-252, 1991 [Vol. I, Issue 17, March 6, 1992] ................. 13 [6] **Passive Smoking and the Risk of Heart Disease," K. SteenLand,./'AMA 267(I): 94-99, 1992 [Vol. I, Issue 14, January 17, 1992] .................................................................................................... ........................ 13 [7] "The Association Between Carotid Arterial Wall Thickness and Acdve and Passive Cigarette Smoking," G. Howard, M. Szklo, G. Evans, G. Tell, J. Eckfeldt, G. Heiss and the ARIC Investigators, Arteriosclerosis and Thrombosis 11(5): 1432a, 1991 [Vol. I, Issue 13, January 3, 1992] ............................................................................ 14 [8] =Urinary Cotinine Measurement in Patients W/th Buerger's Disease ~ Effects of Active and Passive Smoking on the Disease Process," M. Matsushita, S. Shionoya and T. Matsumoto, Journal of Vascular Surgery 14(1): 53-58, 1991 [V'ol. I, Issue 11, November 22, 1991] .................................................................................................... .......... 14 [9] "Passive Smoking and the Risk of Heart Attack or Coronary Death," A.J. Dobson, H.M. Alexander, ILF. Heller and D.M. Lloyd, The MedicadJournalofAuatralia 154: 793-797, 1991 [VoI. I, Issue 10, November 1, 1991] ........... 15 [10] ~Passive Smoking Alters Lipid Profiles in Adolescents," J. Feldrnan, I.tL Shenker, I~A. Etzel, F.W. Spierto, D.F-L IA1ienfield, M. Nussbaum and M.S. Jacobson, Pediatrics88(2): 1-6, 1991 [Vol. I, Issue 6, August 27, 1991] .... 15 [11] "Health Scare - the Misuse of Science in Public Health Policy," J.R. Johnstone and C. Ulyatt, Australian Institute for Public Policy, CritfcalIssuesNo. 14, 99 pp., 1991 [VoL I, Issue 4, July 12, 1991] .................................. 16 RESPIRATORY DISEASES AND CONDITIONS - ADULTS [1] "Does Environmental Tobacco Smoke (ETS) Ca.use Adverse Health Effects in Susceptible Individuals? A Critical Review of the Scientific Literature: I. Respiratory Disorders, Atopic Allergy and ReLated Conditions," P. Witorsch, 2int4ronmental Technology 13: 323-340, 1992 [Issue 24, Item 37] .............................................................................. 16 [2] "Passive Cigarette Smoke-Challenge Studies: Increase in Bronchial Hyperactivity,= P. Menon, ILJ. Rando, R.P. Stankus, J.E. Salvaggio and S.B. Lehrer, Journal ofAllerg7 and CllnicalImmunology 89: 560-566, 1992 [Vol. I, Issue 22, May 22, 1992] .................................................................................................... ............................ 17 [3] "Muldpollutant Exposures and Health Responses to Particulate Matter," M.D. Lebowitz, J.J. Quackenboss, M. Kr-zyzanowski, M.K. O'Rourke and C. Hayes, Archives of EnvironmentalHealth 47(1): 71-75, 1992 [Vol. I, Issue 21, May 8, 1992] .................................................................................................... .............................. 18 [4] "Respiratory Irritation from Environmental Tobacco Smoke," R.J. Shephard, Archives ofEn viro}lmenral Health 47(2): 123-130, 1992 [Vol I, Issue 21, May 8, 1992] ............................................. 19 [5] "Environmental Tobacco Smoke: Causative Agent or White Elephant?" R. R.ylander, Archives of Environmental Health 47(2): 102-103, 1992 [Vol. I, Issue 21, May 8, 1992] ............................................ 19 [6] ~Asr.hmatic Resp~bnses to Passive Cigarette Smoke: Persistence of Raeactivity and Effect of Medications," P.K. Menon, ILP. Stankus, tL.I. Rando, J.E. Salvaggio and S.B. Lehrer, Jouwaal of A//ergy and C~'nica/ lrnmunolog7 88: 861-869, 1991 [Vol. I, Issue 14, January 17, 1992] .......................................................................... 20 [7] ~Respiratory Symptoms in Indian Women Using Domestic Cooking Fuels," D. Behera and S.K. Jindal, Chest 100(2): 385-388, 1991 [Vol. I, Issue 12, December 10, 1991] .......................................................................... 20 [8] ~Domesti¢ Pollution and Respiratory Illness in a Himalayan Village," T. Norboo, M. Yahya, N.G. Bruce, J.*L Heady and K.P. Ball, International Journal ofEpidemiolog7 20(3): 749-757, 1991 [Vol. I, Issue 12, Doc~mber 10, 1991] .................................................................................................... ................... 21 [9] ~Respirarory Illness in Nonsmokers Chronically Exposed to Tobacco Smoke in r.he \X~ork Place," J.IL White, H.F. Froeb and J.A. Kullk, Chest 100(1): 39-43, 1991 [Vol. I, Issue 7, September 13, 1991] .................................... 21 [10] "Effects of Air Pollution on Adult Pulmonal-y Function," X. Xu, D. Dockery and L. Wang, Archives of Environmental Health 46(4): 198-206, 1991 [Vol. I, Issue 7, September 13, 1991] .................................................... 22 [11] "The Environment and the Lung: Changing Perspectives," J.M. Samet and M. Utell, ./AMA 266(5): 670-675, 1991 [Vol. I, Issue 7, September 13, 1991] ......................................................................... 22 [12] =Adverse Health Effects Among Adults Exposed to Home Dampness and Molds," R.E. Dales, R. Burnett and H. Zwanenburg, American Review of Res~iratory Disease 143: 505-509, 1991 [Vol. I, Issue 6, August 27, 1991] ........ 23 [13] "Upper Respiratory Tract Environmental Tobacco Smoke Sensitivity," tL Bascom, T. Kulle, A. Kagey-Sobotka and D. Proud, American Review of Respiratory Disease 143:1304-1311, 1991 ['Vol. I, Issue 5, August 9, 1991] ......... 23 -86- I I I I I I I I I I I i I I I

VOLUME I, ISSUES 1-24 JULY 1992 I I I I I I I I I I I I I I i I i ! RESPIRATORY DISEASES AND CONDITIONS - CHILDREN [1] "The Effect of Maternal Smoking During Pregnancy on Early Infant Lung Function," J.P. Hanrahan, I.B. Tager, M.tL Segal, T.D. Tosteson, tLG. Castile, H. Van Vunakis, S.T. Weiss and F.E. Speizer, American Review of Respiratory Disease 145:1129-1135, 1992 [Issue 24, Item 38] .................................................................................... 24 [2] "Asthma in Children," G.L Larsen, The New 2~nglandJournal of,~gediclne 326(23): 1540-1545, 1992 [Issue 24, Item 39] .................................................................................................... ................................................. 25 [3] "Passive Smoking Among Schoolchildren in Israel," A.I. Goren and S. Hell.man, Environmental Health Perspewtives 96: 203-211, 1991 [Issue 24, Item 55] .................................................................................................... ......................... 26 [4] "Indoor Air Pollution and Acute Respiratory Infections in Children," The lancet 339: 396-397, 1992 [Issue 24, Item 56] .................................................................................................... ................................................. 27 [5] "E/:Fects oFPassive Smoking on Lung Growth in Children," M.D. L~bowitz, D. Sherrill and C.J. Holberg, Pedia~ricPulmonology 12: 37-42, 1992 [Vol. I, Issue 22, May 22, 1992] ...................... 27 [6] =The Extent and Impact of Environmental Tobacco Smoke CETS) Exposure in Children with Asthma," LM. Sa.lmun, Bag. ChiLmonczyk, K.N. Megathlin, L.M. Neveu.x, G.J. Knight, A.J. Pulkkinen and J.E. Haddow, Abstract No. 348, Journal ofAllerg7 and ClinicalImrnunology 89(1): Part 2, 1992 [Vol. I, Issue 21, May 8, 1992] ..... 28 [7J ~Effect of Passive Smoking on Asthmatic Children Who Have and Who Have Not Had Atopic Dermatitis," A.B. Murray and B.J. Morrison, Chest 101: 16-18, 1992 [Vol. I, Issue 19, April 10, 1992] ....................................... 28 [8] "Childhood Asthma and Passive Smoking: Urinary Cotinine as a Biomarker of Exposure," I:L F..hrlich, M. Katra.n, J. Godbold, D.S. Saltzberg, K.T. Grimm, P.J'. Landrlgan and D.E. Lilienfeld, American Review of Respiratory Disease 148: 594-599, 1992 [VoL I, Issue 19, April 10, 1992] ............................................................................................... 29 [9] ~Effects of Environment and Passive Smoking on the Respiratory Health of Children," F. Forastlere, G.M. Corbo, P. Michelozzi, R. Pistelli, N. Agabiti, G. Brancato, G. Ciappi and C.A. Peruoai, InternationalJournalofEioiderniology 21(2): 66-73, 1992 [Vol. I, Issue 18, March 20, 1992] ............................................................................................. .-30 [10] "Lung Function In School-Age Children Who Had Mild Lower R~piratory Illnesses In Early Childhood," G.L Strope, P.W. Stewart, F.W. Henderson, S.S. Ivins, H.C. Stedman and M.M. Henry, Ame~qcan Review of Reslairatory Disease 144: 655-662, 1991 ['Vol. I, Issue 17, March 6, 1992] .................................................................. 30 [I I] "Effects of Asthrrm on Pulmonary Function in Children: A Longitudinal Populatlon-Based Stu~iy," S.T. Weiss, T.D. Tosteson, M.R. Segal, I.B. Tager, S. Redline and F.E. Speizer, American Review of Respiratory Disease 145: 58-64, 1992 ['Vol. I, Issue 16, February 17, 1992] .................................................................................. 31 [12] ~Increased Incidence of Asthma in Children of Smoking Mothers," F.D. Martinez, M. Cline and B. Burrows, Ped/-atr/cs 89(1): 21-26, 1992 [Vol. I, Issue 15, January 31, 1992] ............................................................................. 32 [13] ~Impalred Pulmonary Function in Schoolchildren Exposed to Passive Smoking," R. ¢2~_~=te, D. Colantonio, M. Clalente, V. Colorizio, R. Barnabei and P. Pasqualettl, Respiration 58: I98-203, 1991 [Vol. I, Issue 15, January 31, 1992] .................................................................................................... ........................ 32 [14] "Childhood Asthma and the Indoor Environment," C. Dekker, R~ Dales, S. Bartlett, B. Brunekreefand H. Zwanenburg, ChestlO0(4): 922-926, 1991 [Vol. I, Issue 12, December 10, 1991] .................................................... 33 [15] =Quantifying Health Aspects of Passive Smoking in British Children Aged 5-11 Years," S. Chinn and RJ. lq.ona, Jourr~al of Epidemiolog~ and Community Health 45: 188-194, 1991 [Vol. I, Issue 11, November 22, 1991] ................ 33 [16] "Parental Smoking and the Pdsk of Childhood Asthma,n A. Bener, A~R. Ai-Frayh Fachar'zt and T.Q_. AI-Jawadi, fournalofA~t~6ma 28(4): 281-286, 1991 [Vol. I, Issue 11, November 22, 1991] ........................................................ 34 [17] ~Children With Recurrent l~espirarory Tract Infections Tend to Belong to Farrdlie~ With H~alth Problems," M. Soderstrom, B. Hovelius and K. Prellner, Ac~a l~aedlatr. Scan~ 80: 696-703, 1991 [Vol. I, Issue 10, November 1, 1991] .................................................................................................... ..................... 34 [18] "Passive Smoking and Childhood Asthma," S. Willers, E. Svenonius and G. Skarping, A~46: 330-334, 1991 [VoL I, Issue 10, November 1, 1991] ................................................................................ 35 [19] "Indoor Air Pollution Exposure and Lower Respiratory Infections in Young Gambian Children,~ J.R.M. Armstrong and H. Campbell, Interna~qonal Journal of Efoidemiolo~y 20: 424-429, 1991 [Vol. I, Issue 10, November 1, 1991] .... 35 [20] =Children's Exposure to Environmenr.al Cigarette Smoke Before and After Birth,~ M.D. Overpeck and A.J. Moss, Advance Data, No. 202, 11 pp., June 18, 1991, National Center for Health Statistics ['VoL I, Issue 7, September 13, 1991] .................................................................................................... ..................... 35 [21] "Environmental Tobacco Smoke Exposure and Respiratory Health in Children: An Updated Critical Review and Analysis of the Epldemiological Literature," R-D. Hood, J.M. Wu, R.J. Witorsch and P. Witorseh, lndoorEnvironmenr I: 19-35, 1991 [Vol. I, Issue 7, September 13, 1991] ................................................................. 36 [22] "Acute Effect of Pa.ssive Smoking on Lung Function and Airway Responsiveness in Asthmatic Children," M. Oldigs, R. Jorre~ and H. Magnuss~n, Pediatric Pulraonolog7 10: 123-131, 1991 [VoL I, Issue 7, September 13, 1991] ........... 37 [23] "Association of Indoor Nitrogen Dioxide With Respiratory Symptoms and Pulmonary Function in Children," LM. Neas, D.W. Dockery, J.H. Ware, J.D. Spengler, F.E. Speizer and B.G. Ferris, Jr,, American Journal of ~/dem/o/og;y 134(2): 204-219, 1991 [Vol. I, Issue 7, September 13, 1991] ............................................................... 37 -87-

ETS ABID IAQ.REFERENCE JULY 1992 [24] =The Effects of Indoor Environmental Factors on Respiratory Illness in Primary" SchooI Children in Kuala Lumpur," B.H.O. Azizi and R.L. Henry, InternarionalJournalofEpidemialogy 20(1): 144-150, 1991 [Vol. I, Issue 7, September 13, 1991] .................................................................................................... ..................... 38 [25] "Respiratory Health Effects of Home Dampness and Molds Among Canadian Children," tLE. Dales, H. Zwanenburg, R. Barnett and C.A. Franklin, American.]'aurnalofEpidemialogy 134(2): 196-203, 1991 [Vol. I, Issue 6, August 27, 1991] .................................................................................................... ........................... 38 [26] "The Effect of Indoor Air Pollutants on Otitis Media and A.sthma in Children," G.E. Dalgler, S.J. Maxkello and K.M. Curnnaings, Laryngoscope 101: 293-296, 1991 ['Vol. I, Issue 5, August 9, 1991] ............................................. 39 [27] "The Influence of a Family History of Asthma and Parental Smoking on Airway Responsiveness in Early Infancy," S. Young, P.N. Lesouef, G.C. Geelkioed, S.M. Stick, K,J. Turner and L.I. Landau, Th~NewAFngLandJournalof Medicine 324:1168-1173, 1991 [Vol. I, Issue 5, August 9, 1991] .............................................................................. 40 [28] ~Risk Factors for Respiratory SyncytiaI Virus-Associated Lower Respiratory Illnesses in the First Year of Life,= C.J. Holberg, A.L~ Wright, F.D. Martinez, C.G. Ray, LM. Taussig, M.D. Lebowitz and Group Health Medical Associates, American Journal of AFpidemiology 133(11): 1135-1151, 1991 [Vol. I, Issue 5, August 9, 1991] ................. 40 [29] "Relationship of Parenril Smoking to Wheezing and Nonwheezing Lower Respiratory Tract Illnesses in Infancy," A.L. Wright, C. Holberg, F.D. Martinez, L.M. Taussig and Group Health Medical Association, The JaurnalofPedia~qcs 118(2): 207-214, 1991 [Vol. I, Issue 2, May 17, 1991] .............................................................. 41 OTHER CANCER [1 ] "Breast Cancer, Cigarette Smoking, and Passive Smoking," A.J. Wells, American Journal of~Fpiderniolo,~y 133(2): 208-210, 1991 and Letters to the Editor Regarding Same [Vol. I, Issue 22, May 22, 1992] ....................................... 41 [2] Letters to the Editor Regarding "Prermtal Exposure to Parents' Smoking and Childhood Cancer,~ F-~M. John, Savitz a.nd D.P. Sandier, American.[ournalofEpidemiology 133: 123-132, 1991 [V'ol. I, Issue 22, May 22, 1992] ...... 42 [3] "Epldemiologic Evidence for the Role of Indoor Tobacco Smoke as an Initiator of Huma.n Breast Circinogenesis," A.W. Horton, CartcerDetection andJ~re~ention 16(2).' 119-127, 1992 [V'ol. I, Issue 21, May 8, 1992] ...................... 42 [4] ~Exposure-Response ReLationships Between Woodworking, Smoking or Passive Smoking, and Squa.mous Cell Neoplasms of the Maxillary Sinus," K. Fukuda and A. Shibata, Cancer Causes and Control 1: 165-168, 1990 [Vol. I, Issue 15, January 31, 1992] .................................................................................................... ........................ 43 [5] "Parental Occupation and Or.her En-.4ronmenr.il Factors in the Etiology of Leukemias and Non-Hodgkin's Lymphomas in Childhood: A Case-Control Study," C. Magnani, G. Pastore, L. Luzzatto and B. Terracini, Tumori76: 413-419, 1990 [Vol. I, Issue 7, September 13, 1991] .............................................................................. 44 OTHER HEALTH ISSLrES [1] "Childhood Exposure to Environmental Tobacco Smoke and the Risk of Ulcerative Colitis," R.S. Sandler, D.P. Sandier, C.W. McDonnell and J.I. Wur'zelmann, American Journal of Epidonioloxy 135(6): 603-608, 1992 [Vol. I, Issue 22, May 22, 1992] .................................................................................................... ............................ 45 [2] "Demographic Ind Socioeconomic Diff'erences in Beliefs about the Health Effects of Smoking," R.C. Brownson, J. Jackson-Thompson, J.C. Wilkerson, J.R. Davis, N.W. Owens and E.B. Fisher, ArnericanJo~rnalof Pubh'cHealth 82(1): 99-103, 1992 [Vol. I, Issue 22, May 22, 1992] .......................................................................... 45 [3] Letters to the Editor Regarding "Group Day C.Ire and the Risk of Serious Infectious Illnesses," A.T. Berg, E~D. Shapiro and L.A. Capobianco, AmerlcanJo~rrlalofEpidemiology 133(2): 154-163, 1991 [Vol. I, Issue 22, May 22, 1992] .................................................................................................... ............................ 46 [4] "Maternal Smoking and Neuropsychotogical Development in Childhood: A Re~'iew of the Evidence," S. Tong and A.J. McMichael, Developmental Medi~'ne and Child Neurology 34:191-197, 1992 ['Vol. I, Issue 21, May 8, 1992] .................................................................................................... .............................. 47 [5] ~Passive Smoking and Otitis Media with EW-usion," G.S. Barr and A.P. Coatesworth, British Medical Journal 303: 1032-1033, 1991 [Vol. I, Issue 13, January 3, 1992] ................................................................................................. 48 [6] "Influenc~ of'Paternal Age, Smoking, and Alcohol Consumption on Congenital Anomalies,~ D,~. Savitz, P.J. Schwingl and M,ei. Keels, Teratology44: 429-440, 1991 [Vol. I, Issue 13, January 3, 1992] ............................... 48 [7] "Multiple Chemical Sensitivity," B. Hilen-~m, Chemical and En~neering News, pp. 26-42, July 22, 1991 [Vol. I, Issue 7, September 13, 1991] .................................................................................................... ..................... 49 [8] ~Passive Smoking by Pregnant Women and Fetal Growth," H. Ogawa, S. Tomlnaga, K. Hori, iK. Noguchi, I. Kanou and M. Matsubara, Journal of AFpidemiology and Community Health 45:164-168, 1991 [Vol. I, Issue 6, August 27, 1991] .................................................................................................... ........................... 49 [9] "Risk Factors for Birth Defects: Data from the AtLanta Birth Defects Case-Control Study," J.D. Erickson, Terarology43: 41-51, 1991 [Vol. I, Issue 5, August 9, 1991] ...................................................................................... 50 [10] =A Comparison of Active and Passive Smoking During Pregnancy: Long-Term Effect," J. Makin, P.A. Fried and B. Watkinson, Neurotoxicology and Teratology 13: 5-12, 1991 [Vol. I, Issue 4, July 12, 1991] ............................. 50 [11] ~Effect of Passive Smoking During Pregnancy on Selected Perinatal Pa.mmeters," F. La.zzaroni, S. Bonassi, E. Manniello, L. Morcaldi, E. Repetto, A. Ruocco, A. Calvi and G. Cotellessa, International Journal of F_pidemiology 19(4): 960-966, I990 [Vol. I, Issue 2, May 17, 1991] ................................................................................................ 5' -88- I I I I I I I I I I I I I I

! VOLU~ I, ISSUES 1-24 JULY 1992 I I I I I i I i I I I I I I I ! ! I [ 12] ~Editorial Commentary: New Effects of ActVve and Passive Smoking on Reproduction," J.M. Sa.met, American Journal of Epidemiolog7 133(4): 348-350, 1991 [Vol. I, Issue 1, April 30, 1991] ......................................... 51 [13] ~Chronic Fetal Hypoxia and Sudden Infant Death Syndrome: Interaction Between Maternal Smoking and Low Hematocrit During Pregnancy," M.G. Bulterys, S. Greenland and J.F. Kraus, Pediatrics 86(4): 535-540, 1990 [VoL I, Issue 1, April 30, 1991] .................................................................................................... .............................. 51 ETS EXPOSURE AND MONITORING [1] "The Personal, Indoor and Outdoor Concentrations of RSP and Nicotine Measured in Six Smoker's Families in Taiwan," C.C. Chart, S.K. Huang, Y.C. Chen and J.D. Wang, Presented at the American Industrial Hygiene Association/American Congress of Governmental Industrial Hygienists Conference, June 3, 1992, Boston, MA [issue 23, Item 29] .................................................................................................... ................................................. 52 [2] "Toxic and Trace Elements in Tobacco and Tobacco Smoke," M. Chlba and R. Masironi, Bulletin of the World Health OrganizaHonTO(2): 269-275, 1992 [Issue 23, Item 32] ........................................................................ 52 [3] ~Measurement ofSaliva_,-y Codnlne to Judge Tobacco Smoke Exposure in Wheezing Children," P.W. Heymann, A.L. DufF, E. Pomeranz, L.E. Gelber, A.H. Farris and T.A.E. Platts-Mills, Abstract No. 347, Journal of Alderg7 and Clitiicallmmunolog~89(1): Part 2, 1992 [VoL I, Issue 21, May 8, 1992] ................................................................... 53 [4] =An Objective Assessment of Environmental Tobacco Smoke (ETS) Exposure in 5-7 Year Old Children," S. Clark, T. A~sadullahi and J.O. Warner, Abstract No. FC28, Clinical and Experimental Allerg7 22: 109-142, 1992 [Vol. I, Issue 21, May 8, 1992] .................................................................................................... .............................. 53 [5] "P'.~posure to Passive Smoking Among Bar SIn_W," M.J. Jarvis, J. Foulds and C. Feyerabend, Bri~hJournal of Addiction 87:111-113, 1992 [Vol. I, Issue 18, March 20, 1992] ............................................................................... 54 [6] "Tea Drinking, Passive Smoking, Smoking Deception and Serum Cotinine in the Scottish Heart Health Study," H. Tunstall-Pedoe, M. Woodward and C.A. Brown, Jo~rnalofClinicalEpidemiolo~44(12): 1411-1414, 1991 [ Vol. I, Issue 18, March 20, 1992] .................................................................................................... .......................... 54 [7] "Nicotine and Cotinine in the Cervical Mucus of Smokers, Passive Smokers, and Nonsmokers," M.F. McCann, D.FL Irwin, L.A. Walton, B.S. Hulk,% J.L. Morton and C.M.A.xelrad, CancerEpidemiology, Biomarkers dr Preuention 1: 125-129, 1992 [Vol. I, Issue 18, March 20, 1992] ................................................................................ 55 [8] "Exposure to Environmental Tobacco Smoke in Naturalistic Settings," K.M. Earlmons, D.B. Abrams, R.J. Marshall, R.~ Et2el, T.E. Novomy, B.H. Marcus and M.E. Kane, AmergcanJournalofl~liclgealth82(1): 24-28, 1992 [Vol. I, Issue 16, February 17, 1992] .................................................................................................... ...................... 55 [9] "Misclassiflcation of Smoking Status By SeE-Reported Cigarette Consumption," E.J. Pdrez-Stable, G. Marfn, B.V. MaHn and N.L Benowitz, American Review of Respiratory Disease 145: 53-57, 1992 [Vol. I, Issue 16, February 17, 1992] .................................................................................................... ...................... 56 [10] "Dietary Nicotine: A Source ofUrinary Cotinine,= P,.A. Davis, M.F. Stiles, J.D. DeBethizy and J.H. Reynolds, Fa~ Chem. Toxic. 29(12): 821-827, 1991 ['Vol. I, Issue 16, February 17, 1992] ................................................................ 57 [11] "Gas-Partlculate Phase Distribution and Decay Rates of Constltuents in Ageing Environmental Tobacco Smoke," G.B. Neurath, S. Petersen, M. Dtinger, D. Orth and F.G. Pein, Environmental Technolog7 12: 581-590, 1991 [Vol. I, Issue 12, December 10, 1991] .................................................................................................... ................... 58 [12] ~Measuring Personal Exposure to Airborne Mutagens and Nicotine in Environmental Tobacco Smoke," N.Y. Kado, SJk. McCurdy, S.J. Tesluk, S.K. Hammond, D.P.H. Hsieh, J. Jones and M.B. Schenker, Mutation Research 261 : 75- 82, 1991 [Vol. I, Issue 11, November 22, 1991] .................................................................................................... .... 58 [ 13] "Passive Smoking By Humans Sensitizes Circulating Neutrophils," R. Anderson, A.J. Theron, G~A. R.ichards, M.S. Myer and A.J. Van R~nsburg, American Rel~iew of Respiratory Disease 144: 570-574, 1991 [VoL I, Issue 10, November 1, 1991] .................................................................................................... ..................... 59 [14] =Assessing Exposure to Environmental Tobacco Smoke: Is It Valid to Extrapolate From Active Smoking?," M.J. Reasor and Jaei. Wfll, Journal of Smoking-Related Diseaaes 2(1): 111-127, 1991 [Vol. I, Issue 8, October 1, 1991] ..... 59 [ 15] =Mutagenicity of Indoor Air Contalnlng Environmental Tobacco Smoke: Evaluation of'a Portable PM-10 Impactor Sampler," P.E. Georghiou, P. Blagden, D.A. Snow, L Winsor and D.T. Williams, Environmental Science and Technolog7 25(8): 1496-1500, 1991 [Vol. I, Issue 7, September 13, 1991] .......................................................... 60 [16] =Environmental Tobacco Smoke in Public Places," U. Brynnel and G. Lofroth, Meeting Abstract, ThirdAFuropean Meeting of Environmental Hygiene, Dusseldorf, Germany, June 24-27, 1991 [Vol. I, Issue 6, August 27, 1991] ......... 60 [17] "A Comparison of Methods of Assessing Exposure to Environmental Tobacco Smoke in Non-Smoking British Women," C.J. Proctor, N.D. Warren, M.A.J. Bevan and J. Baker-Rogers, F.nvironmen¢ international 17: 287-297, 1991 [Vol. I, Issue 5, August 9, 1991] .................................................................................................... ................... 60 [18] "Correlations Between Urinary Nicotine or Cotinine and Urinary Mutageniciw in Smokers on Controlled Diet," C.A. Rahn, G. Howard, E. RAccio and D. J. Doolittle, Environmental and Molecular Mutagenesis 17: 244-252, 1991 [Vol. I, Issue 5, August 9, 1991] .................................................................................................... .... 61 [19] =Evuluation of Vapor-Phase Nicotine and Respirable Suspended Pirticle Mass as Markers for Environmental Tobacco Smoke," B.P. Leaderer and S.K. Hammond, EnvironmentalScience and Tevhnolog7 25(4): 770-777, 1991 [Vol. I, Issue 2, May 17, 1991] .................................................................................................... ..................... 61 -89-

ETS AND IAQREFERENCE JULY 1992 INDOOR AIR QUALITY [1] I~tters to the Editor Regarding "The Measurement of Environrnental Tobacco Smoke in 585 Office Environments," S. Turner, L. Cyr and A.J. Gross, Environrnentinternationa118: 19-28, 1992 [issue 24, Item 4] ............................... 62 [2] "Sick Buildings Ind the Experimental Approach," D.P. Wyon, Environmental Technology 13: 313-322, 1992 [Issue 24, Item 57] .................................................................................................... ................................................. 63 [3] =IAQand Energy-Management By Demand Controlled Ventilation," F. Haghighat and G. Donnini, ~EnvironmentaI Technology 13: 351-359, 1992 [Issue 24, Item 58] .............................................................................. 64 [4] "Indoor Air Pollution and Its Health Effects in China ~ A Review," B.H. Chen, C2.J. Hong and X~Z. He, Environmental Technology 13:301-312 [issue 24, Item 59] ........................................................................................ 64 [5] "Nasal Deposition of Inhaled Acrolein and Acetaldehyde Vapors," J.B. Morris, CIAR Currents 1 (3): 1,4, 1992 [issue 23, Item 20] .................................................................................................... ................................................. 64 • [6] "Modeling the Indoor Environment," B.S. Austin, S.M. Greenfield, B.R. Weir, G.F_.. Anderson and J.V. Behar, Environ. Sci. Te~hnol. 26(5): 851-857, 1992 [See Appendix A] [Vol. I, Issue 22, May 22, 1992] ............................... 65 [7] "Air Quality During the Winter in Quebec Day-Care Centers," S. Daneault, M. Beausoleil and K. Messing, American Journal of Public Health 82: 432-434, 1992 [Vol. I, Issue 19, April 10, 1992] ............................................ 66 [8] "Adaption to Indoor Air Pollution," L. Gunnarsen and P.O. Fanger, Environment international 18: 43-54, 1992 [Vol. I, Issue 18, March 20, 1992] .................................................................................................... ......................... 66 [9] '*Oc.cupat~ona! Stress in the Aircraft Cabin," M. Edwards, Cabin Crew SaJgOe 26(5): I-6, I991 [Vol. I, Issue 16, February 17, 1992] .................................................................................................... ...................... 67 [10] "The Los Angeles TEAM Study: Personal Exposures, Indoor-Outdoor Air Concentrations, and Breath Concentrations of 25 Volatile Organic Compounds," L. Wallace, W. Nelson, R. Ziegenfus, E. Pellizziri, L. Michael, 1~ Whitmore, H. Z~lon, T. Hart-well, R. Perritt and D. Westerdahl, Journal of~-xpoxure Analysis andEnvironrnental ~'.t,/dern/o/bgy 1(2): 157-192, 1991 [Vol. I, Issue 15,January 31, 1992] ...................................................................... 68 [11] "Indoor Climate and Air Quality in New Offices: Effects of a Reduced Air-Exchange Rite," B.M. Berardi, E. I_=oni, B. Marehesini, D. Ca~cella and G.B. Ra_ffi, International Arrhi~es of C)c~upational and Environmental ]-tealth 63: 233-239, 1991 [Vol. I, Issue 15, January 31, 1992] ................................................................................................... 69 [12] "The Measurement of Environmental Tobaa:x:o Smoke in 585 Office Environments," S. Turner, L. Cyr and A.J. Gross, EnriChment international 18: 19-28, 1992 [Vol. I, Issue 14, January 17, 1992] ....................................... 69 [13] "Relationship Between Occupant Discomfort as Perceived and as Measured Objectively," F. Haghighat, G. Donnini and lq. D'Addario, Indoor Environment 1 : 112-118, 1992 [Vol. I, Issue 13, January 3, 1992] .................................... 70 [14] "Indoor Air Pollutants From Unrented Kerosene Heater E-misslons in Mobile Homes: Studies on Particles, Sea-nivolatile Organic.s, Carbon Monoxide, and Mutageniciry," J.L. Mumford, R.W. Williams, D.B. Walsh, R.M. Burton, D.J. Svendsgaard, J.C. Chuang, V.S. Houk and J, l_~maa, Environmental Science and Technology 25: 1732-1738, 1991 ['Vol. I, Issue 12, Dec.ember 10, 1991] ........................................................................................... 71 [15] =Ca.bin Air Quality and Health Risks: A Review and Synthesis of'the Available Data," L.C. Holcomb and W. Crawford, ITA Magazine66: 17-22, 1991 [Vol. I, Issue 11, November 22, 1991] ............................................... 71 [16] ~Indoor Sources of Mutagenic Aerosol Particulate Matter: Smoking, Cooking and Incense Burning," G. L_froth, C. Stensmasa and M. Brandhorst-Satzkorn, Mutation Research 261 : 21-28, 1991 ....................................................... 72 [17] "Environmental, Occupational, and Personal Factors Related to the Prevalence of Sick Building Syndrome in the General Population," D. Norback and C. Edling, British Journal of Industrial a~4edict'ne 48:451-462, 1991 [-V'ol. I, Issue 10, November 1, 1991] .................................................................................................... ..................... 72 [18] "I-IVAC Retrofit for Healthy Schools," R.C. Thompson, G. Fisher, T. Brennan, W.A. Turner and F. McKnight, presented at ASHRAE meeting, IndoorAir ~_uali~y ~1, September 1991 [Vol. I, Isw, ae 8, October 1, 1991] .............. 72 [19] "Diagnostica and Remediation for Healthy Schools," W~fi.. Turner, F. McKnlght, G. Fisher, R. Thompson and T. Brennan, presented at ASHRAE meeting, IndoorAir "91, September 1991 [Vol. I, Issue 8, October 1, 1991] ....... 73 [20] "Passive Cigarette Smoke Increases Working Level Exposures and Lung Cancer Risk to All Occupants of the Smoker's Home," R.H. Johnson, Jr. asad E. Geiger, presented at the 36th annual meeting of the Health Physica Society, June 21-26, 1991, Waahington, D.C. [VoL I, Issue 8, October 1, 1991] ...................................................... 73 [21] "Airborne Gluc-an and Endotoxin in Sick Buildings," R. Rylander, CZAR Currents 1(1): 3-4, 1991 [Vol. I, Issue 7, September 13, 1991] .................................................................................................... ..... 74 [22] "Sources of Polynuclear Aromatic Hydrocarbons (PAl-I) in the Indoor Air ofHong Kong Homes," L.C. Koo, J.H.-C. Ho, H. Matsushita, H. Shimlzu, T. Mori, H. Matsuki and S. Tominaga, Meeting Abstract, IndoorAir International, "Priorities for Indoor Air Research and Action," Montreux, Switzerland, May 29-June I, 1991 [Vol. I, Issue 6, August 27, 1991] .................................................................................................... ....................................... 74 [23] "Indoor Dust Exposure: An Unnoticed Aspect of Involuntary Smoking," H.O. Heln, P. Saudicani, P. Skov and F. Gyntelberg, An:hives of~.nvironmentalHealth46(2): 98-101, 1991 [Vol. I, Issue 2, May 17, 1991] ....................... 74 -90- I I I I

JULY 1992 I I I I I I I I I I I I I I ;I I il ! SMOKING POLICIES AND RELATED ISSUES [1] ~Discomfort From Environmental Tobacco Smoke Among Employees at Worksites with Minimal Smoking Restrictions -- United States, 1988," Morbid#y and Mona/iF Weekly Report 41 (20): 351-354, 1992 [Issue 24, Item 42] .................................................................................................... ................................................. 75 [2] ~The Effex:ts of Ordinances Requiring Smoke Free Restaurants on Restaurant Sales in California," S.A. Glantz ~lnd L.R.A. Smith, Monograph Series, Institute for Health Polio3," Studies -- UniversitT ofCudifornia, Sua~ Francisco, March 1992, 21 pages [Vol. I, Issue 20, April 24, 1992] ............................................................................................ 76 [3] "A Conceptual Framework for the Roles of Legislation and Education in Reducing Exposure to Environmental Tobacco Smoke," L.L. Pederson, J.M. Wankiin, S.B. Bull and M.J. Ashley, American Journal of Health Promotion 6(5): 105-111, 1991 [Vol. I, Issue 17, March 6, 1992] .............................................................................................. 76 [4] "Employee Knowledge and Attitudes About a Work-Site Nonsmoking Policy: Rationale for Further Smoking Restrictions," G. Sorensen, N~A- tLigotti, A. Rosen, J. Pinney and IL Prible, Journal of Occupational Medim'ne 33 (11): 1125-1130, 1991 [Vol. I, Issue 17, March 6, 1992] .......................................................................................... 77 [5] "Work-Site Smoking Policies in Small Businesses," G. Sorensen, A. Rosen, J. Pinney, J. Rudolph and N. Doyle, Journal of Occupational A4edicine 33(9): 980-984, I991 [Vol. I, Issue 17, March 6, 1992] ......................................... 77 [6] "Occupational Risks Associated with Cigarette Smoking: A Prospective Study," J. Ryu.n, C. Zwerling and E.J. Orav, Araerican.fourna/ of Public Health 82(1): 29-32, 1992 [Vol. I, Issue 16, Febtuary 17, 1992] ...................................... 78 [7] "The Department of Veteraans Affairs Smoke-Free Policy," A.M. Joseph and P.J. O'NeiI,./AMA 267(1): 87-90, 1992 [V'ol. I, Issue 14, January 17, 1992] .................................................................................................... ........................ 79 STATISTICS AND RISK ASSESSMENT [1] "Trust ~`: Credibi~ity in R~sk C~rrtmun~cati~n~" V. C~ve~ Health & Envir~nrnent Diges~ A~r~1992 [Issue 24, Item 43] .................................................................................................... ................................................. 79 [2] "The Interplay of Science, Values, and Experiences Among Scientists Asked to Evaluate the Hazards of Dioxin, Radon, and Environmental Tobacco Smoke," G.I.. Carlo, N.L. Lee, K.G. Sund and S.D. Pettygrove, RiskAnaly~is 12(1): 37-43, 1992 [Vol. I, Issue 19, April 10, 1992] ................................................................................................ 80 [3] "Can Meta-A_nalyses Be Trusted?," S.G. Thompson and S.J. Pocock, The lancet 338:1127-1130, 1991 [Vol. I, Issue 13, January 3, 1992] ......................................................................................... 81 [4] "A Frame~.ork for Risk Characterization of Environmental Pollutants," D.F. Naugle and T.K. Pierson, Journal of the Air and Wa.rte Management Association 41(l O): 1298-1307, 1991 [Vol. I, Issue 12, December 10, 1991] .......... 81 [5] ~Meta-Analysis: A Review," R. Rosenthal, l~sychosorna~ic Medicine 53: 247-271, 1991 [Vol. I, Issue 10, November 1, 1991] .................................................................................................... ..................... 82 [6] "The Potential and Limitations ofMeta-Analysls," T.D. Spector and S.G. Thompson, Journal of~pid~mio/ogy and Community Health 45: 89-92, 1991 [Vol. I, Issue 6, August 27, 1991] ................................................................ 82 [7] "Meta-Analysis in Epidemiology, with Special Reference to Studies of the Association Between F_.xposure to Environmental Tobacco Smoke and Lung Cancer: A Critique," J.L. Fleiss and A.J. Gross, Journal of Clinica/ Epidemiology44(2): 127-139, 1991 [Vol. I, Issue 1, April 30, 1991] .......................................................................... 83 -91-