Tobacco Documents Online

OSHA POSTHEARING SUBMISSION t ~ Submitted by: Helen B. Hubert, Ph.D. Senior Research Scientist Department of Medicine and Health Research & Policy Stanford University Medical Center Stanford, California ~ August 29, 1995 ~ ~ I I I ~ ~ ~ I ~ 2~9 I p.i'

OSHA POSTHEARING SUBMISSION Helen B. Hubert. Ph.D. Senior Research Scientist Departments of Medicine and Health Research & Policy Stanford Universiry Medical Center STanford, California I have obtained and reviewed a series of five deliverables prepared by Kenneth G. Brown. Ph.D.. Incorporated as a subcontractor to Meridian Research Inc. (Task Order No. 3, contract No. J-9-F- 1-0065), entitled "OSHA's Critical Evaluation of Epidemiological Studies on Cardiovascular Disease Risk in Nonsmokers Exposed to Passive Tobacco Smoke." These deliverables were submitted by Dr. E.own over a period of I yeir (November 1993- July 1994) to Ms. Debra Janes at OSHA and include the following: First Deliverable: (1) Updated risk analysis on passive tobacco smoke and lung cancer in the form of revised material from the EPA document, including a revised version of Chapter 5, additions and revisions to Appendix A, and additions to the bibliography. (2) Review and analysis of epidemiologic data on the association between exposure to passive tobacco smoke in the workplace and the risk of lung cancer. Second Deliverable: A report on eptdamiologic studies on passive tobacco smoke and heart disease prepared by Dr. Brown as a subcontractor to Meridian Research Inc. Third Deliverable: Study results on factors that may affect risk of heart disease. Fourth Deliverable: Tier classification scheme for epidemiologic studies on heart disease and passive tobacco smoke. Fifth Deliverable: Tables and figures on heart disease. Although I recognize these deliverables to be drafts of Dr. Brown's analyses on OSHA's behalf, I have some major concerns regarding Dr. Brown's assessments of the epidemioiogic data on exposure to "pagsive" tobacco smoke (PTS, also referred to in this analysis as environmental tobacco smoke, ETS) and lung cancer and cardiovascular disease risks. In particular, I am concerned about errors and discrepancies in the report, inappropriate conclusions drawn from the ~ data, and improper use of statistics. ~ Cdl ~ 1 (~ ~ GO Io

2057837307 ww m

I I I I I I exposures to other occupational factors that may be associated with lung cancer or heart disease risks. Furthetmore. because of widely recognized variability in dietary, lifestyle, socioeconomic conditions, genetics, and other potentially disease-related factors, the relative'irtfluence of confounders may vary in different countries or regions of the world. Another weakness of the report on workplace exposure to environmental tobacco smoke and lung cancer is that none of the published studies adjusted for spousal or other sources of environmental tobacco smoke to derive the independent contribution of workplace environmental tobacco smoke on risk. On page 1 I of the first deliverable, Dr. Brown acknowledges that "the small sample sizes, the low power to detect an effect, the difficulty of assessing exposure in the workplace, the mix of surrogates used for workplace exposure, and the probable correlation of exposure to PTS at (work) with exposure in other environments, leave the evidence only suggestive." The data are obviously of poor quality and Dr. Brown equivocates in his conclusions on page I 1 by stating, "If PTS is a lung carcinogen, as concluded by EPA, then exposure to PTS in the workplace adds to the toxic burden from exposure in other environments to increase the risk of lung cancer." Such a statement is not, in effect, based on any hard, factual data. In comparison to Dr. Brown's assessment of the workplace lung cancer data, LeVois and Layard (1994) performed a meta-analysis of the results of 12 epidemiologio studies that reported risks of lung cancer and workplace exposure to environmental tobacco smoke. Sixteen relative risks were provided in the papers identified; 9 of these 16 were above 1.0 and 7 were less than or equal to 1.0. LeVois and Layard stated that: "Of the 16 reported relative risks, we combined 15 by computing a weighted average of their logarithms, me weights being the inverses of the variances of the log relative risks (the Butler relative risk of 0.0 for males could not be included in the meta-analysis). The summary relative risk for 12 of the 14 worldwide studies was 1.01, with 95% confidence interval (0.92, 1.11). The summary relative risk for 7 of the 9 U.S. studies was 0.98 with 95% confidence interval (0.89, 1.09). Although we could not include the Brownson et al. (1992) and Stockwell et al. (1992) studigs in the meta-analysis, since they did not report relative risk estimates for workplace exposure, the authors' comments ... indicate that including them would not change the conclusion that there is no epidemiologic evidence of an association between workplace ETS exposure and lung cancer." (p. 312) Whereas one could argue that any approach to combining data from these studies may be flawed, use of LeVois and Layard's technique does not present any evidence for an effect of workplace exposure to ETS on lung cancer risk. _ Inappropriate Conclusions Drawn from the Dose-Response Data The issue of an alleged dose-response demonstrated by studies of spousal or workplace exposure ~ to environmental tobacco smoke and lung cancer or cardiovascular disease is one that has ~

I I I I I I I I I I I I t I I I bias and confounding inherent in that design; that is. such a[meta-analysis] will likely be significant [at the conventional 5% level, irrespective of whether a one-sided or two-sided test is used] simply due to artifact" (p. 3 12). Statistical Power of Studies Most of the studies of lung cancer and cardiovascular disease examined by Dr. Brown have insufficient power to detect relative risks of 1.5 or less as shown in Tables 5-9 (first deliverable on spousal exposure to environmental tobacco smoke and lung cancer risk) and Table 6 (second deliverable on spousal exposure to environmental tobacco smoke and cardiovascular disease risk). Rigorous epidemiologic studies should be designed with sample sizes sufficient to achieve 80-90% power to detect the. risk in question, given the expected disease rate or exposure in the population (for cohort or case-control studies) (Breslow and Day 1987). Since most of the data c~ted by Dr. Brov.n were derived from s:udies that were not specifically designed to examine the role of environmental tobacco smoke on these particular health endpoints, power has been seriously diminished. The major drawback is that without sufficient power and with such low risks described, chance cannot be ruled out as an explanation for many of the elevated risks found. Furthermore, power may actually be more seriously compromised than is reported in the tables if, in fact, risks for both lung cancer and cardiovascular disease are lower than 1.5 as the majority of the data suggest. In Table 5-9 of the first deliverable on spousal exposure to environmental tobacco smoke and lung cancer, only 3 out of 12 U.S. studies of lung cancer (Brownson et al. 1992, Fontham et al. 1991, and Garflnkel et al. 1985) for which power could be calculated show sufficient (>80%) power to detect a risk of 1.5. These are considered by Brown to be Tier 1 or Tier 2 studies. Only one (Fontham et al. 1991) shows significantly elevated risk (OR=1.28), but only using a 90% (instead of 95%) confidence interval (1.0.3, 1.50). Brownson et al. 1992 shows an OR=0.91 (CI= 0.75, 1.05) and Garfinkel et al. 1985 shows an RR=1.16 (CI=0.89, 1.52). Of the remaining non- U.S. studies, only Wu-Williams and Samet (1990) has sufficient power, reporting an OR=0.78 (CI=0.63, 0.96). Greater power is also indicative of greater stability of estimates and reliability, if the study methods and analyses are appropriate and have ruled out bias and confounding. With regard to power, on page 5 of the second deliverable on spousal exposure to environmental tobacco smoke and cardiovascular disease risk, Dr. Brown states "If power is small, the lack of significance may be due to small sample size instead of due to ,io effect; if the power is large, so that failure to detect a real effect is unlikely, then lack of significance is more informative." However, Dr. Brown often fails to acknowledge the importance of nonsignificant findings in instances where power is large. " 12

3. Important biases addressed' (al Selection bias 1 1 r • or YES Population-based (geographically defined) study with minimal nonresponse, and/or drop-out rate. O or NO Evidence of differential response or drop-out rate in cases versus controls/exposed versus unexposed or inadequate control group selected. :1 or BLANK Insufficient information to determine bias. (b) Information bias • or YES No evidence of bias when validation of smoking and ETS exposure undertaken in cases and controls, blinding of interviewers to case/control status, and proxy response rate low and nondifferential in cases and controls. O or NO Proxy response greater in cases than controls or no validation of smoking and ETS exposure or unblinded interviewers. a or BLANK Insufficient information to determine bias. 4. Are the data internally consistent? . • or YES Evidence of crude or significant dose-response, subgroup results that go in the same direction (e.g., males v. females, blacks v. whites, home v. work), and exposure preceding outcome. . _ O or NO No evidence of dose-response, subgroup results that go in a different direction, or unclear temporal relationship. O or BLANK Insufficient_data to determine dos.e-response, subgroup results, or temporality. The relative quantities of open circles and blank spaces in Table I suggest that the studies on cardiovascular disease are not sufficiently reliable to make conclusions with regard to the impact of environmental tobacco smoke exposure on cardiovascular disease risk. Of greatest significance is the fact that studies that ruled out chance as an explanation for elevated risks were, without exception, unable to rule out systematic study bias or confounding as explanations for the fmdings. My evaluation as to whether the data collected in each study are reliable to assess the risk of cardiovascular disease associated with exposure to environmental tobacco smoke exposure is described below.

Additionally, as an expert in the epidemiology of cardiovascular disease, I have performed my own assessment of all of the available epidemiologic studies on environmental tobacco smoke exposure and cardiovascular disease, and it is my belief that these studies do not support a conclusion that the reported associations are real. Specifically, as my analysis shows, the 13 published studies, 2 abstracts, and 1 letter that were available to Dr. Brown at the time of his report are not reliable because they do not adequately exclude important sources of bias and confounding nor rule out chance as an explanation for the fmdings. Whereas the copy of the report I reviewed did not contain Dr. Brown's conclusions on the potential effect of environmental tobacco smoke exposure on cardiovascular disease risk (my copy of the deliverable ends on page 7 and the remainder of the document was not available), Dr. Brown's own critique of the epidemiologic studies similarly points out that the data are inconclusive to establish an effect and rule out bias and confounding. Based upon the rndings of my reassessment )f the cardiovascular disease studies and a recognition that the methodological criticisms apply as well to $TS studies involving other disease endpoints, I am confident that a similar unbiased reassessment of all the available lung cancer epidemiology studies would raise similar questions about Dr. Brown's conclusions. My opinions of Dr. Brown's assessments of lung cancer and cardiovascular disease risks, and the results of my independent review of the cardiovascular data form the basis for this posthearing submission to OSHA. KENNETH BROWN'S APPROACH TO ANALYZING EPIDEMIOLOGIC DATA The series of deliverables details Dr. Brown's assessment of the potential association between spousal and workplace exposure to passive tobacco smoke and lung cancer and cardiovascular disease risks. In general, I have several concerns regarding his approach including reliance on studies of limited epidemiologic quality, inappropriate conclusions drawn from his analysis of the available epidemiologic data and of dose-response, limited evaluation of temporality, limited evaluation of confounding and misclassification, limited evaluation of study heterogeneity, improper use. of statistics, and use of subjective tier classifications. In addition, I was able to identify several errors, discrepancies, and misinterpretations in the deliverables that could compromise the validity of the reports' conclusions. Reliance on Studies of Limited Epidemiologic Quality Dr. Brown readily admits to the limited quality of the epidemiologic studies that evaluated spousal and workplace exposure to environmental tobacco smoke and lung cancer and heart disease risks. His comments can be found in the Appendices accompanying each deliverable. For example, among the lung cancer studies reviewed by Dr. Brown to assess workplace. exposure (first deliverable), he comments that the Akiba study "precludes drawing any firm conclusion regarding workplace PTS and lung cancer" (App-1); the Buffler study has "little

t ' l I I I In Table 2 of the first deliverable, relative risks presented are not comparable since they do not incorporate uniform definitions of exposure or study population. For example, in the analysis of workplace exposure and lung cancer, the crude RRs for the Kabat and Wynder and Lee studies are calculated for both genders combined rather than separately for males and females, whereas odds ratios, by gender, are in fact presented in the table footnotes and/or in the appendices. In the Kabat study, these odds ratios by gender show a statistically significant increased risk for neversmoking males exposed to tobacco smoke at work (2.57) but not for females (0.68). In the Lee study, the crude ORs for nonsmoking males and females separately are 1.6 and 0.63, respectively, neither statistically significant. As another example, in the Akiba study, exposure is defined as outside the home vs. housewife, whereas other studies define exposure differently. On p. 4 of the analysis of workplace exposure and lung cancer, Dr. Brown states that "Most. (9 of 12) [crude] values are above 1.0, two are at or slightly below 1.0, and only a single value falls well below the null." I believe that this overstates the strength of the data. According to his Table 2, crude RRs are reported for 13 studies, 4 of the 13 values fail below 1.0, including CHAN 0.77, GARF 0.93, LEE 0.99, and BUTL 0.6. Six are below 1.5, including AKIB 1.08, FONT 1.12, KABA 1.11, KOO 1.36, SVEN 1.26, and WU WI 1.22. Only LAMW is statistically significant. 4. Table 3 of the analysis of workplace exposure and lung cancer, the adjusted risk reported for BUTL in Table 2, 1.06 does not match that reported for BUTL in Table 3, 1.08. 5. On p. 6 of the analysis of workplace exposure and lung cancer, second paragraph, Brown does not report whether these odds ratios are crude or adjusted. This is an important omission since Dr. Brown notes on p. 5-23 of the first deliverable, a revision of Chapter 5, that'`an adjusted RR is considered preferable to a crude RR unless the study review .... indicates a problem with the adjustment procedure.... our choice of RR is the smaller of the crude and adjusted values in ... studies providing both estimates." 6. On p. 8 of the analysis of workplace exposure and lung cancer, fourth line, Dr. Brown omits BROW from the list of elevated ORs. Again, Dr. Brown does not specify in the text which of these ORs are crude and which are adjusted. 7. Despite the observation that 90% confidence limits were calculated for the analysis of lung cancer and passive tobacco smoke exposure, on p. 6 of the second deliverable, passive tobacco smoke and heart disease, and in the accompanying figures, 95% confidence limits were reported instead. 8. On p. 6 of the second deliverable, passive tobacco smoke and heart disease, despite Dr. Brown's observation of the contrast of estimates of RR for males and females exposed to PTS in the home, he suggests no explanation for this observation. 14

employment status was used as a sutroeate for workplace exposure) are consistent with the hypothesis that workplace exposure to environmental tobacco smoke increases lung cancer risk. On page 6, of the first deliverable on workplace exposure and lung cancer risk. Dr. Brown inappropriately deletes a quite accurate statement, that "workplace exposure to other carcinogenic substances could arguably give rise to the same observation." Employment status is not a convincing measure of environmental tobacco smoke exposure since other risk factors may also be associated with working outside the home. 1 I I I Inappropriate Conclusiops I)rawn Erom the Analyses I have identified several examples of instances in which Dr. Brown's conclusions are not reflective of the data he presents. For example, in his analysis of the spousal lung cancer risk by tier and country, he states on page 5-71, "It is concluded that the association of ETS and lung cancer observed rrom the analysis of 33 epidemiologic studies in eight different countries is not due to chance alone and is not attributable to bias or confounding." However, even if one is prepared to accept the tier ranking approach and discount the Chinese studies, Dr. Brown's statements are inappropriate since he fails to acknowledge that pooling of first and second best ranked studies produces no statistically significant effect of environmental tobacco smoke on lung cancer in the United States or Europe. Only in Greece, Hong Kong, and Japan are results significant and consistent over tier pooling. The fact that these are not U.S. studies renders them of limited utility, given different cultural differences and smoking patterns. In the first deliverable on workplace exposure to environmental tobacco smoke and lung cancei risk, four studies (Brownson et al. 1987, Chan and Fung 1982, Svendsen et al. 1989, and Lam 1985) collected data on workplace exposure but presented only an index of exposure to all sources. Therefore; it is incorrect for Dr. Brown to deduce on page 6, "The observation of elevated ORs in 3 out of 4 studies using a measure of total PTS that incorporates workplace exposure is consistent with the hypothesis that workplace PTS contributes to risk of lung cancer ... Lacking analyses of workplace exposure :.. the observed association is implicit rather than explicit " There is really no way of determining anything about the relationship of workplace exposure to lung cancer from these data. Workplace exposure could have shown no elevated risk and the same results could have been achieved given the exposure classifications in _. these studies (i.e., 4 or more hours per day spent in the presence of a smoker; exposure at home or at work). In addition, observing that three of four studies reported elevated risks, only one of which was statistically significant, also does not rule out chance in establishing an association. This faulty reasoning is apparent throughout the documents on spousal and workplace exposures to environmental tobacco smoke and lung cancer and cardiovascular disease. On page 10 of the first deliverable on workplace exposure to environmental tobacco smoke an lung cancer risk, Dr. Brown points out the inconsistencies of the U.S. data on workplace exposure compared to Asia and Europe and states: "there is no reason to think that if workplace exposure is a risk factbr for lung cancer, it would be so only in European and Asian populations: " Consideration should also be given to regional variations in working conditions and potential 4 I

(1994) appropriately concluded that each smoking study is confounded by uncontrolled influences and produces a biased estimate of ETS effect suggesting that weak spousal smoking- lung cancer risk elevations in the individual studies may well be the result of artifact alone. I I I I I I I I I Use of One-tailed Statistical Tests Dr. Brown's continued reliance on one-tailed tests to assess the relationship between environmental tobacco smoke exposure and disease is improper. For example, on page 5-2 of the f rst deliverable on spousal exposure to environmental tobacco smoke and lung cancer risk, the revision to Chapter 5 states that: "Throughout this chapter, one-tailed tests of significance (p=0.05) are used, which increases the statistical ability (power)'to detect an effect. The 90% confidence intervals used for the analyses performed are consistent with the use of the one-tailed test. The justification for this usage is based on the a priori hypothesis (from the plausibility of a lung cancer effect documented in Chaptets 3 and 4) that a positive association exists between exposure to ETS and lung cancer." However, the data contained in Chapters 3 and 4 of the EPA report concern levels of exposure to environmental tobacco smoke (Chapter 3) and discussions of lung cancer in active smoking and laboratory animals (Chapter 4), and, therefore, cannot be relied on as the basis of a decision to utilize a one-tailed test. Second, Dr. Brown's rationale for the use of the one-tailed test (90% confidence limit) for an overall estimate in a meta-analysis, rather than the conventional 95% confidence limit, is not supported by statistical or epidemiological theory or methodology in this instance. Statistical theory grew out of experimental science, which measures variables or responses. The typical test of significance for measured responses is a t-test, which tests differences in sample means. Two-sided (two-tailed) tests of significance are generally employed in the analysis of simple contrasts in experimental studies. However, one-sided (one-tailed) tests of significance (in which the investigator only considers deviations from the null hypothesis in one direction, ignoring deviations in the other direction) may be appropriate when the investigator either knows in advance that a treatment will affect outcome in only one direction or if the investigator has no interest in the outcome of a treatment unless it is superior to an existing treatment (such as in testing the efficacy of a new drug). In contrast to experimental science, epidemiolo;ical and observational studies are generally concerned with categorical or discrete data. Standard practice in these studies is to calculate odds ratios or relative risks. Statistical significance in epidemiologic studies is normally assessed using Chi-square tests, or equivalent procedures, which generate a confidence interval around the odds ratio or relative risk. Although a Chi-square test uses only one-tail of the Chi- square distribution, the test is in fact two sided, since a significant Chi-square can be generated by deviations from expected values in either direction. 10

and sedentary lifestyle. These factors are often related to one another and to other characteristics that could be associated with heart disease risk. Thus. anv state-of-the-art assessment of risk of heart disease associated with exposure to environmental tobacco smoke should consider and adjust for most, if not all, of these factors in statistical analysis in order to exclude alternative explanations for any associations found. As an extension of my review of the epidemiologic studies of cardiovascular disease and exposure to environmental tobacco smoke presented in my oral testimony at the OSHA hearings (November 1994), I have undertaken a more in-depth assessment of the quality of the studies. In any epidemiologic study, there are several important criteria by which one can independently judge study reliability. These primarily relate to study bias and confounding and chance occurrence. In an effort to critically evaluate these studies according to a standardized list of criteria, I developed a rating sheet that includes the major criteria for study reliability as well as questions whose answers allow determination of whether each of the criteria are iulfilled for a particular study (Appendix A). Detailed summanes of each of the 15 available studies (16 ` papers) on cardiovascular disease and environmental tobacco smoke available to Dr. Brown at the time of his report are provided in Appendix B. Table 1 summarizes my ratings of the quality of each available study according to the rating sheet evaluations found in Appendix A. The criteria on the rating sheets used to create Table I are as follows: 1 Statistically sienificant elevated risks? D or YES Adjusted relative risk (or crude, if adjusted not available) was statistically significant by two-sided test, p<0.05 O or NO Crude or adjusted risk not elevated or risk was not statistically significant by two- sided sided test, p<0.115 O or BLANK No tests or confidence intervals presented for elevated risks. Z Major confoundine considered? • or YES Adjustment for age and gender as well as 4 of the 5 listed risk factors or study ascertained which of the listed risk factors were important and appropriately controlled for them. O or NO No control for age and gender and at least 4 of the 5 risk factors and study did not ascertain important risk factors and appropriately control for them. '-J or BLANK Study did not fully specify what risk factors were controlled for or risk factors unclearly defined (e:g, exercise stress test in the He et al. 1989 study). 16

I I I I I utility for the assessment of workplace PTS effects° (App-2); the findings of the Butler study "are based on too few cases to be reliable" (App-2); the limitations of the Chan study are "sufficient to preclude reliance on the study's data to evaluate the effects of workplace PTS exposure" (App-3); the Kabat and Wynder data "are intriguing, but limitations of the exposure measurements and sample size, in combination with lack of direct control for age or other risk factot•s, undermine their utility". (App-5); and the Lam study's results "lack of specificity and potential for distortion of results by other risk factors are too great for the information to carry much weight" (App-8). Among the cardiovascular studies reviewed by Dr. Brown to assess spousal and workplace environmental tobacco smoke exposure, he comments that the Butler data are "at most suggestive" (A-4); in the Dobson study, ".the potential for bias and confounding leave a causal link with PTS exposure uncertain" (A-9); the Garland study evidence "indicates only a tenuous association" (A-13); the He et a1. (1989) article leaves "room for ambiguity or misunderstanding" (A-18); the Humble study "is more suggestive than conclusive" (A-35); and the Lee study results for ischemic heart disease and stroke are "questionable, particularly so for exposure in all-places from all-sources" (A-44). It is imperative that data be reliable and sufficient enough to reject the null hypothesis of no effect of environmental tobacco smoke exposure. In several instances, Dr. Brown suggests that the data are "consistent" with an effect, but nowhere in his critique does he clearly reject the null hypothesis, nor are the data clearly sufficient to do so. Dr. Brown also acknowledges the absence of adequate exposure assessment data in many of the epidemiologic studies of environmental tobacco smoke. On page 2 of the first deliverable on _ workplace smoking and lung cancer risk, Dr. Brown acknowledges the difficulty in accurately measuring workplace exposure to environmental tobacco smoke, which is a function of the intensity of the exposure due to number of smokers, room size, ventilation, proximity to smokers, etc. For example, in the appendix to this deliverable, he comments that in the Akiba study "it is ... not possible to specify how much of the observed association for work outside the home is due to workplace PTS exposure and how much is due to other occupational exposures" (App-1); in the Buffler study "it is not possible to separate effects attributable to workplace PTS exposure, if any, from those due to other occupational exposures" (App-2); in the Chan study "no differentiation was made betweerr exposure in the home and in the workplace"(App-3); in the Garfinkel study "extremely heavy reliance on proxy respondents ... heightens the potential for exposure misclassification" (App-4); in the Kabat study ". . . only current passive exposure was estimated, rendering the exposure classifications potentially misleading..:'(App-5); and in the Lee study "significant misclassification of relevant exposure is ... a real possibility" (App-9). With regard to lung cancer, Dr. Brown further acknowledges the current absence and need for long-term exposure information due to the lengthy latency period. He states on pages 2 and 3 of the first deliverable on workplace exposure that "efforts at determination of workplace PTS estimates were generally minimal" and that the accuracy of proxy respondents "as sources of workplace PTS exposure estimates is far more questionable." Dr. Brown suggests that the elevated risks associated with employment status outside of the home (or studies in which 3 I

limited number of studies, Dr. Brown does not acknowledge that study results may depend upon the specific disease or death endpoint evaluated. He also does not address the importance of the definition of exposure to ETS in analyzing and comparing study results. For example, some studies are only concerned with current spousal exposure by definition; others combine exposure from spouses of exsmokers and light smokers in the analysis (e:g:, Hirayama). In studies in which risks are elevated (greater than 1.0) in one subgroup only, Dr. Brown makes little attempt to explain such inconsistencies. Examples include male versus female comparisons (Butler et al. 1988, Dobson et al. 1991. Jackson 1989), black versus white (Humble et al. 1990), low socioeconomic status versus high (Humble et al. 1990), and home versus work (Dobson et al. 1991) I I I 1 I I I I I I Improper Use of Statistics There are three fi::.damental problems with Dr. Brown's choice of statistical methodology and the conclusions subsequently derived. These include the incorrect assumption that there is a small probability of so many studies yielding positive results, the inappropriate use of one-tailed statistical analyses, and the reliance on studies with insufficient power to detect relative risks of 1.5. Small Probability of Elevated Risk Occurrence In the first deliverable concerning spousal and/or workplace exposure to environmental tobacco smoke and lung cancer risk, Dr. Brown observes that many studies demonstrate elevated risk '- (i.e., the study results are in the same direction and are of similar magnitude) thereby supporting his conclusion that the association is not due to chance alone. For example, he notes on page 5-43 of the first deli verable on spousal exposure to environmental tobacco smoke that, "li'the points lie more toward the right side of the normal curve than could be likely to occur by chance alone, then the hypothesis of no effect is rejected in favor of a positive association between ETS exposure and lung cancer," and on page 5-52 that "the possibility of chance accounting for the observed associations between ETS and lung cancer has been virtually ruled out by the statistical methods previously applied." - In his conclusions on workplace exposure and lung cancer risk, Dr. Brown comments on page 9 of the first deliverable that, "While few individual studies attain nominal statistical significance, ..., this failing is largely overshadowed by the number of studies observing results in the sstne direction and of similar magnitude (p=0.03 is the probability of 11 or more positive studies out of 14): " It is inappropriate to con^.lude that chance is not a likely explanation for this association on the basis of the probability of positive studies. This exercise is akin to tossing a coin and counting "heads" or ".`tails.° The probability of 11 or more heads out of 14 coin tosses is 0.03. Implicit in this statement, however, is the assumption of a "fair" or unbiased coin, the independence of successive coin tosses, and the assumption that the 14 coin tosses represented a sample of an infinite series of such tosses. Clearly, one cannot equate the 14 datasets on , workplace exposure to envirorunental tobacco smoke with 14 coin tosses. LeVois and Layard 9

Svendsen. K.H., Kuller. L.H.. Martin, M.J.. and Ockene. J.K. 1987. Effects of passive smoking in the Multiple Risk Factor Intervention Trial. Am. J. Epidemiol. 126(5):783-795. Thompson, D.H. and Warburton, D.M. 1993. Dietary and mental health differences between never-smokers living in smoking and non-smoking households. J. Smoking-Related Dis. 4(3):203-211. Woodward. M. and Tavendale, R. 1995. Passive smoking by Tunstall-Pedoe, H., Brown. C.A., self report and serum continine and the prevalence oi respiratory and coronary heart disease,in the Scottish heart helath study. J. Epidemiol. Comm. Health 49:139-143. Tweedie, RL. and Mengersen, K.L. 1995. Meta-analytic approaches to dose-response relationships, with application in studies of lung cancer and exposure to environmental tobacco smoke. Stat. Med. 14:545-559. U.S. Environmental Protection Agency (U.S.EPA). ' 1989. Workshop Report on EPA Guidelines for Carcinogen Risk Assessment: Use of Human Evidence. Office of Research and Development. EPA 625/3-90/017. I !1. i I I I

received considerable attention. Dr. Brown's assessment of dose-response is what Tweedie and Mengersen (1995) referred to as an"unsystematic or 'eve-ball' approach," in that Dr. Brown relies on qualitative comparisons of oddss ratios and occasional reports of statistically significant positive trends in odds ratios in the individual studies alone as evidence of a dose-response. _ _... Tweedie and Mengersen commented that "the use of such qualitative evaluations, without some consideration of the variability in the data, is.liab[e to..lead to misinterpretation." In fact, on page 8 of the first deliverable concerning workplace exposure to environmental tobacco smoke and lung cancer, Dr. Brown acknowledges that the information on dose-response "is too limited in quantity and quality to produce a clear nicture" of any relationship. I I I I 1 ' I I I I have several concerns with Dr. Brown's approach. First, in many of the studies of exposure to environmental tobacco smoke and disease, a trend in odds ratios for different exposure levels is assessed by the investigators or by Dr. Brown using the Mantel extension or other test of linear trend. Such tests, however, do not test the appropriateness of the implicit linear model fitted by these techniques. Modem statistical practice is well described by Breslow and Day (1987) as follows: "When the value of [the goodness of fit statistic] exceeds its degrees of freedom by an amount significantly greater than expected under chi-square sampling, we conclude that the fit is inadequate. Either there are systematic effects that have not been accounted for by the model or else the random variation in disease rates among neighbouring cells is greater than that specified by the Poisson assumption. Agreement between the [goodness of fit statistic] and its degrees of freedom does not guarantee that the fit is good, however, particularly when the degrees of freedom are large. Systematic patterns or trends in the residuals that may be.indicative of departures from model assumptions, and large residual values for individual cells, often are not reflected in the summary measure. Also, a good fit for a model based on a cross-classification that ignores relevant covariables does not imply that such variables are unimportant or should be considered." (pp. 137-138) Such an examination of goodness of fit and residual errors is essential in applying the appropriate statistical model, whether logistic, loglinear, or a model implicit in a Chi-square analysis (Agresti _ _ . . , 1990). However, in none of the 19 studies on spousal exposure to environmental tobacco smoke and lung cancer (Table 1 I of the first deliverable) or I 1 studies on home and workplace exposure to environmental tobacco smoke and cardiovascular disease (Table 7 of the fifth deliverable) for which linear trend was assessed was goodness of fit tested prior to trend testing. If a test for goodness of fit failed for any of these studies, it would have been inappropriate to conclude that a linear trend was present in the log odds ratio. Second, as noted above, some of the studies cited in Table 7 of the fifth deliverable on spousal and workplace exposure to environmental tobacco smoke, reported statistically significant trends with no apparent increase in heart disease risk, including Butler's examination of male workplace exposure (RRs of 1, 1.26, and 0.76 with increasing exposure), He et al. (1994) (RRs of 1, 1.16, 6

I I I I I I I report indicating that "Only the first criterion (temporal relationship) is essential to a causal relationship; with that exception, none of the criteria should be considered as either necessary or sufficient in itself." In reality, however, Dr. Brown cannot definitively demonstrate in many of the studies that a temporal relationship exists since many of the studies he evaluates are case- control (28 out of 32 lung cancer studies. 12 out of 13 workplace studies, and 5 out of 12 , cardiovascular disease studies). Further, some lung cancer studies only measured current exposure without regard for the.long latency petiod which is known to precede the development of lung cancer. Limited Evaluation of Confounding and Misclassification Despite Dr. Brown's efforts to evaluate the relevant confounders, on page 4-16 of the third deliverable concerning confounders of studies on cardiovascular disease and exposure to environmental tobacco smoke, Dr. Brown ac`.cr_owledges that ". .. the influence of some cofactors and the magnitude of their effects have not been fully investigated." Of note, in his analyses ofconfounders, Dr. Brown rigorously evaluates the statistical significance of associations between risk factors and disease and expresses concerns about multiple comparisons within studies. This appropriate level of examination, however, was not exemplified in his evaluations of environmental tobacco smoke exposure and disease in which he relied on a more qualitative, "weight of the evidence`' approach based on the number ofstudies reporting elevated odds ratios. In addition, on page 4-6 of the third deliverable, Dr. Brown acknowledges the study by Thompson and Warburton (1993), suggesting possible spousal concordance of risk factors, i.e., that nonsmoking individuals living in smoking households consume fats more frequently, drink more alcohol, eat less root vegetables and cereals, etc. Dr. Brown does not sufficiently address the issue of misclassification and its effect on relative risk. On page 3 of the first deliverable concerning workplace exposure to environmental tobacco smoke and lung cancer risk, Dr. Brown makes the assumption that "the substantial potential for imprecise exposure estimates and resultant nondifferential misclassification would tend to bias the results of workplace PTS studies toward the null hypothesis (no effect)."+ However, Dr. Brown fails to acknowledge the possibility that in these studies, the misclassification may not be nondifferential, since cases and proxies of cases may tend to overestimate workplace exposure in an attempt to find a cause for the disease. Limited Evaluation of Study Heterogeneit l Dr. Brown payslittle attention in his review to differences in details between studies, and differences in results within studies. For example. different endpoints are addressed from study to study in studies of cardiovascular disease, including ischemic heart disease mortality, total cardiovascular disease mortality, nonfatal heart disease, myocardial infarction only, and myocardial infarction or confirmed coronary stenosis on arteriography. Perhaps due to the 8 I

5.06, and 4.11 with increasing exposure). and Lee's examination of ischemic heart disease in males (RRs of 1, 0.41, and 0.41 with increasing exposure). - Third, a recent meta-analytic approach to dose-response conducted by Tweedie and Mengersen (1995) on epidemiologic studies of lung cancer and exposure to environmental tobacco smoke revealed "little indication of a consistent dose response." In this investigation. Tweedie and Mengersen first examined the dose-response in individual papers using the Armitage statistic for equality of response to different doses and two parametric models (exponential and direct linear). They found that inclusion of the unexposed group may lead to "invalid conclusions about the relationship between an increase in dose and the corresponding response" likely to be the result _ . included unexposed _ individuaTs in all of his analyses. Tweedie and of confounding; Dr. Brown . Mengersen also considered three approaches to meta-analysis--a test for equality of response across dose levels using a combination of the Armitage test statistic, imposition of a random effects model, and imposition of a fixed effects model. They demonstrated that the dose- response is flat above the zero level of exposure, implying that the only real difference is between unexposed and exposed subjects; this finding directly contradicts Dr. Brown's statements that a dose-response relationship between exposure to environmental tobacco smoke and lung cancer exists. Tweedie and Mengersen warned that a number of issues must be considered in any assessment of dose response using epidemiological data, including , _ . standardization of dose levels, the use of appropriate models, and the role of the unexposed group in inference. Fourth, in his qualitative approach, Dr. Brown also does not consider relative risks adjusted foF other risk factors for lung cancer within studies. However, any confounding factor associated with an increase in lung cancer risk also may be associated in a dose-response fashion. One must a:..o question whether it is legitimate to test for trend in studies where the test for effect is not significant and no a priori hypothesis regarding dose-response is put forth. On page 5-48 of the first deliverable on spousal exposure to environmental tobacco smoke and lung cancer risk, Dr. . Brown comments that ". . . three of the U.S _ . studies.... are statistically significant for a test of tn.nd, providing evidence for an association between ETS exposure and lung cancer even though neither was significant in a test for effect ... this occurs because the data supporting an increase in relative risk are largely at the highest exposure level." Finally, claims of a dose response must take into account the fact that neither dose nor exposure was measured in any of these studies. Moreover, the exposure categories are based on recall and are subject to bias. Limited Evaluation of Temporality I I Dr. Brown evaluates the evidence for a causal association between environmental tobacco smoke and lung cance'r according to seven specific criteria developed by a U.S. Environmental Protection Agency workshop (U.S. EPA 19$9) which include (1) temporal relationship, (2) consistency, (3) strength of association, (4) dose-response, (5) specificity of association, (6) biologic plausibility, and (7) coherence. On page 5-72 of the first deliverable on spousal exposure to environmental tobacco smoke and lung cancer risk, Dr. Brown quotes the workshop 7 ,

among never smokers and no trend in risk with the number of years exposed. The Scottish cross- sectional survey (Tunstall-Pedoe et al. 1995) relating measures of ETS (defined by self-reports of none to a lot) to self-reported coronary disease showed a statistically elevated odds ratio of 2.4, 95% Cl=1.1-4.8 (adjusted for age, housing tenure, cholesterol, and diastolic blood pressure) for doctor-diagnosed disease among never smokers who reported "a lot" of exposure. While the odds ratio for diagnosed coronary disease at the highest level of serum cotinine was consistent with that at the highest level of ETS exposure, there are problems in interpreting results from this cross-sectional survey including the temporality of exposure and outcome, possible misclassification of former smokers as never smokers, inconsistencies in results across categories of heart disease, and incomplete control for potential confounders. Thus, neither of these studies present convincing evidence for a true ETS/heart disease association and their results do not alter my conclusions stated above. I 1 Of further interest is the paper by LeVois and Lzyard (1995) in which the authors assessed publication bias in the ETS/heart disease controversy. They compared pooled relative risk estimates from 14 published studies (relative risk=1.29, 95°/aCI=1.18-1.41) and unpublished results from the prospective American Cancer Society's Cancer Prevention Studies CPS-I and CPS-II and the National Mortality Followback Survey done by the National Center for Health Statistics (relative risk=1.00, 95% CI=0.97-1.04), The pooling of these unpublished data from several large studies not only suggest that published data overestimate the association of spousal smoking and coronary disease but also show no increased risk of disease with ETS exposure. 77

.. rw arr r.. .rr r. ,r,.. r. ;rr ` r ar .. ts ..M r. 's.r r TABLE I RELIABILITY OF THE EPIDEMIOLOGIC DATA ETS AND CARDIOVASCULAR DISEASE* REFERENCES V pp 0\ p~p 0~ ~ 00 ~ O . 00 Ch O, Q` T W O, , . . . y ~p p c~ '3 L u Cr'i v ~ 0` 'r3 u ~ a ~ rn ~ ~ .-, a ~ u nf cd y a y Rr ;y ~ - QUALITY CRITERIA ~d ~ , ~ ~ ~ O; v ~ r., ~ • ~ ~ ~ ~ .n x o ~ v -a tr, W a x x x " v i x L1 a x Seatistically significant elevated risks? O O • 0 0 1 0 10 • • O 0 • 0 0 Major confounding considered? 0 0 0 0 O O O 0 O' O O • • Important biases addressed? Selection bias • 0 Inlormation bias O O O O 0 O O O O O O O Are the data internally consistent? O O • 0 0 0 0 • • • Criteria were applied to studies of home exposure, unless data from home exposures were not separately presented. " The He(sing et al. (1988)and Sandler et al. (1989) studies reported data from the same cohort. The studies were assessed in this report independently because of inconsistencies in the authors' reporting of the data. Key: 0 = no • = yes blank space - cannot ascertain '. 8IG';r"..,cgeaSoz

I I I The differences between epidemiologic data and experimental data have led to discussions as to how the statistical significance of epidemiologic data is best assessed. It is now customaty and preferred in epidemiological and observational research to estimate the summary statistic (whether odds ratio or relative risk) and to provide a confidence limit for a possible range of values around this statistic with stated confidence. When, and if, however, significance tests are used instead, the appropriate methodology is to conduct a two-tailed test. Two-tailed significance tests are formally equivalent to confidence limits. As stated by Joseph Fleiss. "If ... the investigator intends to report the results [of the test of significance] to professional colleagues, he is ethically bound to perform a two-tailed test.... Even if ... a large accumulation of published data suggests that the difference being studied should be in one direction and not the other, the investigator should nevertheless guard against the unexpected by performing a two-tailed test. Especialk• in such cases, the scientific importance of a difference in the unexpected direction may be greater than yet another confirmation of the difference being in the expected direction." (Fleiss 1981, p. 28). If an investigator believes that a one-tailed test of significance is appropriate given prior knowledge of outcome or expectations of results, the use of the test must be specified before the data are analyzed. As Selvin states, "The decision to use a one- or two-sided test must be made in advance of the data analysis. Basing the decision on information from the collected data incurs test-direction bias" (Selvin 1991, p. 44). Dr. Brown states that, "The justification for [the use of the one-tailed testj is based on the a priori hypothesis (from the plausibility of a lung cancer effect documented in Chapters 3 and 4) that a positive association exists between exposure to ETS and lung cancer." This approach is not consistent with the recommended use of a one-tailed test since Dr. Brown sets out to test the statistical significance of a null hypothesis of no effect which he has already rejected. He has clearly reached a conclusion about the significance of the data before deciding on a test of significance. Dr. Brown's decision to use a one-tailed test allowed him to estimate a 90% confidence interval rather than the conventional 95% confidence interval. This ultimately allowed him to more easily obtain an apparently significant outcome. As recognized by Bjorn Ander~en: "The advantage of a one-tailed test is that a significant outcome is easier to obtain. ... If one-tailed tests are to be used.at all, the essential requirement is that the decision is made independent of the data. Choosing a one-tailed test in order to obtain a significant result, once the direction of the 3ifference is evident from observations, is a kind of 'data dredging' approaching scientific misconduct..."(Andersen 1991, p. 235). Dr. Brown's desire to apply a 90% confidence interval is all the more surprising given the knowledge that a meta-analysis will re-inforce any systematic biases in the individual studies. As observed by LeVois and Layard (1994), the spousal smoking design is subject to positive bias and confounding, and therefore, "Given the large number of studies, all using the [same] flawed spousal smoking design, a[meta-analysisJ ... will with high probability detect the influence of 11 I

Garland et al. 1985 This cohort study did not demonstrate a statistically significant elevated risk, did not adequately consider major confounding, and did not adequately address information . bias. Potential selection bias could not be ascertained In addition, the statistical methodology employed was not appropriate (one-sided tests of significance). In my opinion- these data are not reliable for assessing cardiovascular disease risk. Dr. Brown comments that "the study evidence indicates only a tenuous association." I Lee etal. 1986 This case-control study did not demonstrate a statistically significant elevated risk, did not adequately consider major confounding, did not adequately address information bias, and the data were not internally consistent. Potential selection bias could not be ascertained. In my opinion, these data are not reliable for assessing cardiovascular disease risk. Dr. Brown comments that numerous opportunities for misclassification of disease and exposure status were present in this study, and that a method of analysis that could consider confounding and risk modifying factors was not used. He concludes that study results for ischemic heart disease and stroke are "questionable, particularly so for exposure in all-places from all-sources." Martin et al. 1986 This abstract contained only limited study details and although a statistically significant elevated risk was reported, very sparse information was available on consideration of confounding or bias. In my opinion, the data presented are not reliable for assessing cardiovascular disease risk. Dr. Brown does not provide an assessment of this study. Svendsen et al. 1987 This. cohort study did not demonstrate a statistically significant elevated risk and did not adequately consider major confounding. Additionally, it could not be - ascertained if other important systematic biases were addressed. In my opinion, these data are not reliable for assessing cardiovascular disease risk. Dr. Brown comments that this study is limited by small size id short follow-up period and is "not without some residual sources of uncertainty „ Butler et al. 1988 This cohort study adequately attempted to control for selection bias but did not demonstrate a statistically significant elevated risk, did not adequately consider major confounding, and did not adequately address information bias. Further, the data were not internally consistent. In my opinion, these data are not reliable for assessing cardiovascular disease risk. Dr. Brown's opinion of the study is that none of the known risk factors for heart disease were taken into account either in the selection of the cohorts or for adjustment in the statistical analysis. He concluded that the "study's results are at most suggestive." Palmer et al. 1988 This abstract provided very few study details from which to assess reliability, and the statistical methods employed were not appropriate. Dr. Brown does not provide an assessment of this study. Helsing et al. 1988 This cohort study demonstrated a statistically significant elevated risk but did not adequately consider major confounding and did not adequately address information bias. In addition, potential selection bias could not be ruled out. The data were found to be internally 19 I

Dr. Brown has not included the Palmer et al. 1988 and Martin et al. 1986 abstracts in his assessment of risk of cardiovascular dis,!ase and has not revised Appendix A to include LaVecchia et al. 1993. I 1 j 10. On p. 4-1 of deliverable 3, factors that may affect risk of heart disease, Dr. Brown did not include the LaVecchia study in this analysis. 11. On p: 4-2 of deliverable 3, factors that may affect risk of heart disease, the report states that "In case-control studies, matching is often employed to help control risk modifiers but it is still necessary to adjust for n>k modifiers statistically ..." This is wrong if he means that adjustment should re aone for those risk modifiers used in the matching. Matching can be "tight" or "loose" which may permit some adjustment, but according to Breslow and Day (1983) "the unconditional [i.e., unmatched] analysis of matched pair data results in an estimate of the odds ratio which is the square of the correct, conditional {on matching] one; a relative risk of 2 will tend to be estimated as 4 by this approach...'' (p. 250)• 12. In deliverable 5, tables and figures on heart disease, Dr. Brown is not consistent in his use of calculating, reporting, or rounding values out to significant decimal places. For example in Table 3, he rounds the percentage of cases and controls with PTS exposure for He et al. 1994 to the whole number. In Table 6, he rounds the adjusted RR for MI (males) from 1.03 to 1.0. INDEPENDENT ASSESSMENT OF CVD STUDIES A determination of hazard due to an exposure must be preceded by convincing evidence that a reliable association exists between the exposure and disease. A reliable association is one that is independent of chance, bias, and confounding. The question of whether a quantifiable risk of he-rt disease or death due to heart disease exists from exposure to environmental tobacco smoke requires not only a relative risk or odds ratio greater than 1.0 at a statistically significant level (p<0.05 for a two-sided test) but also evidence that the increased risks are real and not artifactual or spurious. Determination of whether a relationship is real is made by evaluating the degree of control for confounding and whether methodological or design flaws that may introduce systematic errors or bias have been adequately eliminated. This evaluation is particularly crucial when evaluating risks for heart disease that are low or"weak," that is, risk ratios in the range of 2 or less. Control for confounding is especially important in the investigation of cardiovascular disease. Many researchers have reported a number of established risk factors for heart disease, including older age, male gender, family history of heart disease, elevated blood pressure, elevated cholesterol and other lipoprotein patterns, diabetes or glucose intolerance, certain electrocardiographic abnormalities, active cigarette smoking, and, more controversially, obesity 15 1

Subjective Tier Classifications I t I I 1 I I I The tier classification system and weightings developed by Dr. Brown to rank studies by utility in answering questions with regard to the effects of ETS exposure is highly subjective. It is very difficult to adequately quantify the qualitative differences among studies. On page 5-1 of. the fourth deliverable, which describes his tier classification scheme, he admits that "Different analysts would be apt to disagree about elements of any such approach and the appropriate weights for those elements in assigning studies to tiers" and that "the scheme for assigning penalty points to studies... is an unrefined instrument." However, Dr. Brown bases much of his analysis and conclusions on these tier rankings. To illustrate the subjective nature of this quantitative ranking, Dr. Brown, himself, is inconsistent in the weighting elements he considers to evaluate the cancer and cardiovascular disease studies. For example, "exposure status unverified" and "smoking-related disease in the controls° are included in ranking only one, not both sets of studies (the studies on spousal exposure to environmental tobacco smoke and lung cancer, Table A-I of the first deliverable, and the studies on spousal exposure to environmental tobacco smoke and cardiovascular disease, Table 9 of the fourth and fifth deliverables). In the fourth and fifth deliverables evaluating cardiovasular disease studies, Dr. Brown is also inconsistent in the elements used in the revised Table 9 (of the fifth deliverable) and in his weightings between text descriptions in the first deliverable (pages 5-5, 5-6) and revised Table 9. The weights described in the text and shown in Table 9 for the elements "unblinded interviews," "unadjusted for cofactors (set 1)," "unadjusted for cofactors (set 2)," "cause of death unverified," and "problems with statistical methods" were changed "` between the fourth and fifth deliverables, resulting in different tier rankings for half of the studies in the group. These revised element weightings were also different from those used to evaluate the lung cancer s.adies, providing further evidence of the arbitrary nature of this technique for rating studies. In addition, any reasonably informed cardiovascular epidemiologist would consider diabetes to be a strong risk factor for disease (although of low prevalence in the population) and would place it.in "set..l'.'..of.cofactots and not in "set 2" as Dr. Brown has done. Further, the importance of such factors as occupation and marital status described as "set 2" oofactors is questionable. Errors, Discrepancies, and Misinterpretations I recognize that these submissions are preliminary drafts, and as such are likely to be revised considerably at a later time for style and conteat. However, I would like to take this opportunity to note some errors or misrepresentations I have observed in the deliverables. Although noticeably minor, they reflect the preliminary nature and inaccuracies of Dr. Brown's presentations. 1. In Table 5-1 of the first deliverable, a revision of Chapter 5 from the EPA report, the reference for the Sobue study is incorrectly reported as Shimizu et al. 1988. 13 C3I GO ~ 2'V . ~ tR

Potential selection bias could not be ascertained. Further. the data were not intemally consistent. In my opinion, these data are not reliable for assessing cardiovascular disease risk. Among Dr. Brown's concerns are limited exposure assessment, limited control for confounding, and bias, leading him to conclude that "this study is more suggestive than conclusive." Dobson et al. 1991 This case-control study demonstrated a statistically significant elevated risk but did not adequately address systematic bias. Further, the data were not internally consistent. In my opinion, these data are not reliable for assessing cardiovascular disease risk. Dr. Brown identified potential sources of bias and confounding in this study leading him to conclude that "a causal link with PTS exposure jis] uncertain." LaVecchia et aL 1993 This study, whose data are reported in a letter, adequately considered major confounding and the data were internally consistent, but the study did not demonstrate a statistically significant elevated risk or adequa,e:y address information bias. Potential selection bias could not be ascertained. In my opinion, these data are not reliable for assessing cardiovascular disease risk. These data were not reviewed by Dr. Brown in this report. He et al. 1994 This study adequately considered major confounding but did not demonstrate a statistically significant elevated risk or adequately address information bias. In addition, although the data were internally consistent, selection bias could not be ascertained. It is my opinion that these data are not reliable for assessing cardiovascular risk. Dr. Brown did not comment on the quality of these data since at the time of the renort, they were available only in an abstract. CONCLUSIONS An interim review of the draft Meridian report raises important questions about the reliability of Dr. Brown's analyses with regard to hazard. Dr. Brown misused epidemiologic data, improperly used statistics, and reached inappropriate and unsupported conclusions based on the data he used. Dr. Brown failed to adequately rule out bias, confounding, and chance in evaluating the relative risks reported in the studies he reviewed. In the area of cardiovascular diseasee risk, Dr. Brown acknowledges the inadequacies of the data available, yet provides tier rankings that are inconsistent with his qualitative assessments; and that fail to recognize many of these studies' liabilities. My assessment leads me to conclude:that studies showing an elevated risk of cardiovascular disease due to envirotunental tobacco smoke exposures generally are of insufficient quality to determine that the claimed associations are real. Since my oral testimony (November 1994), and the submission of Dr. Brown's report (July 1994), there have been three new reports on the relationship between ETS and cardiovascular disease (Muscat and Wynder 1995, Tunstall-Pedoe et al. 1995, and LeVois and Layard 1995). The study of Muscat and Wynder, a hospital-based case-control study of myocardial infarction in four U.S. cities, found an elevated but not statistically significant odds ratio of 1.5, 95% CI=0.9- 2.6 (adjusted only for age, education, and hypertension) associated with adult exposure to ETS 21

Hirayama. T. 1984. Lung cancer in Japan: Effects of nutrition and passive smoking. Hirayama. T. 1990. Passive smoking. i;Z Med. J. 103:54. Hole, D.J.. Gillis, C.R., Chopra, C., and Hawthorne. V.M. 1989. Passive smoking and cardiorespiratory health in general population in the west of Scotland. BMJ 299:423-427. I I I Humble, C.. Croft, J., Gerber, A., Casper, M., Hames, C.G., and Tyroler, H.A. 1990. Passive smoking and 20-year cardiovascular disease mortality among nonsmoking wives. Evans County, Georgia. Am. J. Public Health 80:599-601. _ Jackson, R. 1989. The Aukland Heart Study: A Case Control Study of Coronary Heart Disease. Doctoral dissertation, University of Aukland, Aukland, New Zealand. La Vecchia, C., D'Avanzo, B:, Franzosi, M.G., and Tognoni, G. 1993. Passive smoking and the risk of acute myocardial infarction. Lancet 341:505-506. Lee, P.N., Chamberlain, J., and Alderson, M.R. 1986. Relationship of passive smoking to risk of lung cancer and other smoking-associated diseases. Br. J. Cancer 54:97-105. LeVois, M.E. and Layard, M.W. 1994. Inconsistency between workplace and spousal studies of environmental tobacco smoke and lung cancer. Reg. Toxicol. Pharmacol. 19:309-316. LeVois, M.E. and Layard, M.W. 1995. Publication bias in the environmental tobacco smoke/coronary heart disease epidemiologic literature. Reg. Toxicol. Pharmacol. 21:184-191. Martin, M.J., Hunt, S.C., and Williams, R.R. 1986. Increased incidence of heart attacks in nonsmoking women married to smokers. Presented October I, 1986 at the I 14th Annual Meeting of the American Public Health Association. Abstract. Muscat J.E. and Wynder, E.L. 1995. Exposure to environmental tobacco smoke ^nd the risk of heart attack. Int. J. Epidemiol. 24(4):715-719. Palmer, J.R., Rosenberg, L., and Shapiro, S. 1988. Passive smoking and myocardial infarction in women. CVD Epidemiol. News. 43:29. Abstract. Sandler, D.P., Comstock, G.W., Helsing, K.J., and Shore, D.L. 1989. Deaths from all causes in non-smokers who lived with smokers. AJPH 79:163-167. Selvin, S. 199 L. Statistical Analysis of Fpidemioloeic Data. New York: Oxford University Press. 24

REFERENCES Agresti, A. 1990. Categorical Data Analysis. New York: Wiley & Sons. I I I I I i I Andersen, B. 1991. Methodoloeical Errors in Medical Research. Oxford: Blackwell. Breslow, N.E. and Day, N.E. 1983. Statistical Methods in Cancer Research. Volume I The Analysis of Case-Control Studies. Lyon, France. IARC.' Breslow, N.E. and Day, N.E. 1987. Statistical Methods in Cancer Research. Volume 2. The Design and Analysis of Cohort Studies. Lyon, France: IARC. Butler, T.L. 1988. The Relationship of Passive Smoking to Various Health Outcomes Among Seventh-Day Adventists in California. Doctoral dissertation, University of California, Los Angeles. Dobson, A.J., Alexander, H.M., Heller, R.F., and Lloyd, D.M. 1991. Passive smoking and the risk of heart attack or coronary death. h4ed. J. Aust, 154:793-797. Fleiss, J.L. 1981. Statistical Methods for Rates and Prooortions. Second edition. New York: Wiley. Garland, C., Barrett-Connor, E., Suarez, L., Criqui, M.H., and Wingard, D.L. 1985. Effects of passive smoking on ischemic heart disease mortality of nonsmokers. Ant. J. Epidemiol. 121(5):645-650. He, Y., Li, L.X., Fong, C.C., Li, L.S., Chang, X.L., and Qua, Q.L. 1989. Passive smoking in females and coronary heart disease. Chin. Prev. Med. 23:19-22. He, Y., Lam, T.H., Li, L.S., Li, L.S., Du, R.Y., Jia, G.L., Huang, J.Y., and Zheng, J.S. 1994. Passive smoking at work as a risk factor for coronary heart disease in Chinese women who have never smoked. BMJ 308:380-384. Helsing, K.J., Sandler, D.P., Comstock, G. W., and Chee, E. 1988. Heart disease mortality in nonsmokers living with smokers. Am. J. Epidemiol. 127:915-922. Hirayama, T. 1981. Non-smoking wives of heavy smokers have a higher risk of lung cancer: A study from Japan. BMJ 282:183-185. Hirayama, T. 1984. Cancer mortality in nonsmoking women with smoking husbands based on a I large-scale cohort study in Japan. Prev. Med. 13:680-690. ~ , 23 Go W  C4 O 4~h

consistent, but certain statistical methods employed were not appropriate and may have underestimated the variability of the risk estimates. In my opinion, these data are not reliable for assessing cardiovascular disease risk. Dr. Brown comments that there are potential sources of bias in this study and that some factors not considered remain potential confounders. , . He et al. 1989 This case-control study is translated from Chinese and provided little information on study details. The study demonstrated a statistically significant elevated risk but insufficient information was available to assess sources of systemattc bias. In my opinion, the data are not reliable for assessing cardiovascular disease risk. Dr. Brown similarly comments on the absence of certain methodological information precluding assessment of the potential for bias, and concluded that the findings leave "room for ambiguity or misunderstanding." Hole et al. 1989 This cohort study demonstrated a statistically significant elevated risk but did not adequately comsider major confounding and did not adequately address information bias. In addition, selection bias was not addressed. Further, the data were not internally consistent. In my opinion, these data are not zeliable for assessing cardiovascular disease risk. Among the limitations cited by Dr. Brown are differential misclassification of smokers as never smokers and inadequate examination of gender differences in response. Jackson 1989 This case-control study did not adequately consider major confounding and did not adequately address information bias; chance and potential selection bias could not be ascertained. Futther, the data were not t intemally consistent. In my opinion, these data are not reliable for assessing cardiovascular disease risk. Dr. Brown's concerns include inadequate - identification of potential sources of bias and potential for confounding leading Dr. Brown to conclude that "it is unlikely that firmly supported conclusions can be drawn." Sandier et al. 1989 This cohort study examined various causes of disease or mortality in the cohort examined by Helsing et al. 1988. As with the earlier study, a statistically significant elevated risk was reported but confounding and information bias were not adequately considered. Potential selection bias could not be ascertained. In addition, statistical methods were not appropriate as in the Helsing study. In my opinion, these data are not reliable for assessing cardiovascular disease risk. Hirayama et al. 1990 This cohort study, whose relevant data are presented in a letter, did not demonstrate a statistically significant elevated risk and did not adequately consider major confounding. Other potential systematic biases could not be ascertained. In addition, statistical methods were not appropriate (one-sided tests of significance). In my opinion, these data are not reliable for assessing cardiovascular disease risk. Dr. Brown similarly expressed concerns relating to potential bias and/or confounding and determination of exposure to environmental tobacco smoke. Humble et al. 1990 This study did not demonstrate a statistically significant elevated risk, did not adequately consider major confounding, and did not adequately address information bias. 20

2057837324

I I I I I I I I Martin et al. 1986 Page 2 Other Potential Limitations (Recognized by authors): Statistical Methods: Study Results: Dose-Response: Ascertainment of Temporal Relationship: None. Proportional hazards model used to control for known risk factors. Compared to women married to never-smokers, the women married to current smokers were 4.4 (p<0.01) times as likely to have had a heart attack. When a proportional hazards model was used to control for known risk factors, the relative risk was 3.4 (p<0.01). Investigators reported an increased risk with an increased length of exposure: women married to former smokers had less of an increased risk (RR = 1.9) than women married to current smokers (RR = 4.4). Historical study; no information on presence of disease at commencemetlt of the study. I

1 I I I I ' I I I I I CRITERIA BY WHICH TO ASSESS STUDY QUALITY Reference: Lee et al. 1986 ... QUALITY CRITERIA Yes No Cannot Ascertain RELiT3VC.RISK-. . . .. .. ... ... . • Is crude relative risk elevated° . . . . . . . . . . V • Is adjusied relative risk elevated? V • Is adjusted (or crude i(adjusted not available) statistically significant by two-sided test, p<.05? . ... V CCMIFOLFNDINC . . . : :._ . .. . . .. _ ... .. . . . • Have age and gender been adequately cohtrolled? . . . V • Have az least 4 of the 5 following risk factors been controlled for: family history, blood pressure, cholesterol, diabetes/glucose intolerance, left venmcular hypertrophy on EKG, or any specific EKO abnormality? V • Has the study asccrtained which of the above risk factors ara potential confounders and appropriately controlled for them? . . .. . . . .. . . . . t/ SFGNII+(C"AHT 13T.SS • Has the choice of the control group been adequate? . r •Nonresponse ' - a) Was the nonresponse rate unacceptably high? - -- - - -- V b) Was it different in cases and eontrols" r c) Were refusals among controls more likely to be smokets or ETS exposed? - --- - V • Dropouts (cohon study only) N/A . . a) W as the dropout ratc unacceptably high? b) Did the dropout rate differ by ETS exposure status? ........ ........ ....... ......... . .... ... .. . - • Proxy response .. .. a) W as proxy response rate greater in cascs than controls? V b) Was thetg greater than 10% proxy response? -- - - - - - - V • Was there adequaaze validation of ective smoking and ETS exposum? V • Were the fntervitWers blinded to case/control status? . V RYi`ERNAL CONSISI'E:`7CY OF DATA • Was thete crude evidence of dose-response? . . . V • Wavthcrestatistieallysigniftcantevidenceofdose-)esponse? - - V • Were there similar results in subgroups studied? (Results that go in the same direction; iPnotf do the differences makesense?) . V • Was temporal relationship clearly demonstrated? - V ATPttQPR4ATBNES6 OF STATISTIGE METI7OD3MIVAf.4'SFS . . . • Was there use of two-sided tests of significance? V • Were statistical tests or confidence intervals reported? .- " -. V • Wete statistical methods and adjustments appropriate? --- -- V I

ABSTRACT FORM EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE , Reference: Garland, C., Barrett-Connor, E., Suarez, L., Criqui, M.H., and Wingard, D.L. 1985. Effects of passive smoking on ischemic heart disease mortality of nonsmokers. Am. J. Epidemiol. 121(5):645-650. (USAI Type of Study: Prospective Source of Study Population: Participants were culled from a predominantly white, upper-middle-class suburb of California during 1972 and 1974. I I I I I I 1 Definition of Exposure: Exposed were 695 married women aged 50-79 years who had never smoked cigarettes and had no prior history of heart disease or stroke; they were classified according to their husband's self-reported smoking status. at entry into the study -- former or current smoker. Never-smokers were considered the unexposed control group. Length of Follow-up: 10 years Participation Rates: Vital status determined by an annual mailing for an average of 10 years had an overall ascertainment rate of 99.6%. Percentage of Proxy Respondents: N/A Definitions of Outcome(s): Ischemic heart disease (ICDA 410.0-414.9) as coded from death certificates by a certified nosologist. ~ Procedures used to Validate -- Outcome(s): . p g Death certificate diagnosis of ischemic heart disease was validated by interview with next of kin, physicians, and/or hospital records in 85% of a~ subsample of the cohort. ~ W ort blinding by exposttre status : The authors do not re Blindin

ABSTRACT FORM EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE I I I I I I I Reference: Martin, M.J., Hunt, S.C., and Williams, R.R. 1986. Increased incidence of heart attacks in nonsmoking women married to smokers. Presented October 1, 1986 at the 114th Annual Meeting of the American Public Health Association. [Abstract] [USA] Type of Study: Nonconcurrent cohort Source of Study Population: Definition of Exposure: Length of follow- up: Participation Rates: 18,344 parents (9,172 spouse pairs) of Utah high school students; there were 7,115 never-smoking women who were between the ages of 30 and 59 and for whom there was information on the husband's smoking status. Exposure based on husband's smoking status -- current, former, and never. Not specified. Not reported. Percentage of Proxy Respondents: Not reported; it is assumed no proxy respondents were contacted. Definitions of Outcome(s): Heart attack as reported by subject. Procedures used to Validate Outcome(s): None. Blinding: Not specified. Confounders: ° Family history of CHD, hypertension, diabetes, weight, alcohol intake, and amount of exercise.

CRITERIA BY WHICH TO ASSESS STUDY QUALITY . I , I I I I I Reference: Garland et at. 1985 QUALITY CRITERIA - Yes Vo Cannot Ascerrain RELiTIVERiSM. --. ._, .. ..,.~ . • Is crude relative risk elevated? V • Is adjusted relative risk elevated? V • Is adjusted (or crude if adjusted not available) statisticallv significani by two-sided test, pz.05? t/ CoNFOUNatNa .: .. .. .. . . ... .. . .. .. . .. ...- . .. • Have age and gender beenn adequately controlled? d • Have at Icast 4 of the 5 following risk factors been controlled for: family history, blood pressurq cholesterol. diabetes/giucose intolerance, left ventricular fiypertrophy on EKG, or any specific EKG abnonnality? - . +e • Has the study ascertained which of the above risk factors are potenual confounders and appropriately controlled for them? r S[GNII+[CANI"BtAB . . . .. . . . . • Has the choice of the control group been adequate? NiA • Nonresponse . . .. . . .. . . . .. . . . . .. . . . . . . . . . a) Was the nonresponse rate unacceptably high? . ... _ . . . . . . . . . . . s~ b) Was it different in cases and controls? N/A c) Were refusals among controls more likely to be smokers or ETS exposed? N/A •Dropuuts(cohurtstudyonly) -'-- a) Was the dropout rate ~ ~aceaptably high? . - ... . . .. .. . . _ r b) Did the dropout rate differ by ETS exposure status? • Proxy response N/A a) Was proxy response rare greater.in cases than controls? ... .... , . . . b) W as there greater than 109% prozy respoirse? . -+.. .. - .- • W ac there adequate validation of aotive smoking and ETS exposun:° --- - --- -- - rl •WeretheinterviewersblindedtocaseJ:, .rulstatus? N/A . fSkTERNAL EQNSISTENCY OR D.4T A • W as there crude evidence of dose-response? .. --. .. . .. . . . . . . .: ~. d • Wastherestatisticallysignificanievidenceofdose•response? .. ... .... . . . .. ..... . .. . v • Were there similar resultt in subgroups studied? (Results that go in the same direc8on: if not, do the . difTerencesmakeYense?) N/A - • Was temporal relationship clearly demonstrated? V ATPRUPRUTENEl50FSFA•[ib`TI£ALMETRODSfA1ALYSLS . ; . . ... . • W as there use of two-sided tests of significance? V • Were statistical tesis or confidence intervals reported? d • Were statistical methods and adjustments a ro riate' .:. .. - .... .. . - .. J i

I I f I CRITERIA BY WHICH TO ASSESS STUDY QUALITY Reference: Martin et al. 1986 (abstract) ' . . . .. ... . . ...... .... .. ... ..... ... .... ...... . .. .. QCALITS'CRITERIA - --- - --Yes Nb Cannot Ascert.in REE-ATIVERISIf..' . . . . ... . .. . . . • Is crude relative risk elevated? . . y •Isadjustedrelativerrskelevatcd'--. -- - y • is adjusted (or crude if adtusted not available) stztistically significant by two-sided test, p<.05? s~ TORULEUUTCU7\FOUl'FDINCi. . . . . . ... .., . . . ... .. • Have age and gender been adequateiy controlled? •- -. y • Have at least 4 of the 5 following risk factors been controlled for: family history, blood pressurb, cholesterol, diabetcstglucoae intolcrance, left ventricular hypertrophy on EKG, or any specific EKG abnormality? d • Have at least 3 out of 5 other important risk factors been conuolied? or, d • Has the study ascenained which ofthe above risk factors are potential confounders and appropriately • crontrollcd for them? N - TORULEQI)TSiC.N1FZCA.FP1•BLt.S . ., . . . ., ... . .. .. • Has the choice of the control group been adequate? . I y •Nonresponse - a) Was the nonresponse rate unacceptably high? V b) Was it different in cases aiid controts? . . .... -.- -- _, - y c) W ere refusals among controls more likely to be smokers or ETS exposed? y • Dropouts (cohort sro,.y only) " .. . ... a) Was the dropout rxte unacceptably high? y bl Did the dropout rate differ by ETS exposure status? . . y • Proxy response . a) W as proxy response rlfe greater in cases than cqntrols^ ~ r bl Wastheregreatcrthan IOY.proxyrcsponse? ---.- - ..... ._ . . . . t/ • Was there adequate validation ofactive smoking and ETS exposure^ r • Were the interviewers blinded to caulcontrol status? -~ . .. • .. y TOASC=P.ii'fAIN IFTfiE DATA ARE CO'15757E-N"P • Wat there crude evidence of dose-rvsponse? - r/ .... ......... • Was there sta[istic-il(y significani cvidence af dou•response^ s~ • Were there similar rcsults in subgroups studied? (Results that go, in the same ditection; If not, do the, diffcrencesmakesePse?) .~ • Was temporal relationship clearly demonstrated? v-- p TOA5CE:ATAINiFSFATISTICALifErfEOD5'AN.4LYSLSAREAPYROPRIAT€ . • Was there use of two-sided tests of significance? y • Were sta[istsal tesEcorconfsdence intervals reported? d- • Were stetistical methods and adjustmentsappropriates r/

Garland et al. 1985 Page 2 Confounders: Cumulative mortality rates and relative mortality risks were adjusted for age, systolic blood pressure, plasma cholesterol, obesity index, and duration of marriage to current spouse using Cox's proportional hazards model. Other Potential Limitations (Recognized by authors): Small sample size Results confined to pass~va smoking exposures in the marriage in effect at the time of entry only. I I I I I I Study Results: Crude and age-adjusted mortality rates (percent), 1974-1983 in non- smoking women according to husband's cigarette smoking status at entry, 1972-1974 Husband's smoking status Crude p-Vaiue' Adjusted p-Value' Never 1.0 < 0.05 1.2 5 0.10 Former 3.8 3.6 Current 2.1 2.7 Dose Response: No dose-response data are provided for ischemic heart disease. The authors conclude that a dose-response relationship exists between quantity of cigarettes smoked by the husband and the age-adjusted mortality rate (all causes) of the wife. Ascertainment of Temporal Relationship: Diseased were excluded from exposure group at outset of study. wives of current and former smokers. ~ ~ fta ~ Statistical significance assessed at one-sided p levels of <_ 0.05 and _< 0. 10. Lower than combined rate for X ca -11 ~

CRITERIA BY WHICH TO ASSESS STUDY QUALITY Reference: Svendsen et al. 1987 QUALITY CRITERIA yes No Cannor .Ascertain RELATIVER7SK .:. . ._. . .~ .. - . ... . ... • Is crude relative nsk elevated? r • Is adjusted relative risk elevated? . . . d • Is adjusted (or crude if adjusted not available) statistically significant by two-sided test, p405? V TO RULE OUT CONFOUNDING . . . • Have age and gender been adequately conuolled? . - d • Have at least 4 of the 5 following risk factors been controlled for: family history, blood pressure, cholesterol, diabetes/glucose intolerance, left ventricular hypertrophy on EKG, or any specific EKG abnormality? . . . V • Hac the study ascertained which ofthe above risk factors are potential confounders and appropriately controlled for them? d . TO RULE OIlPS[GNIIRCANf18I65 ... . .. I . . . . . . . .. . .. . . . . • Has the choice of the control group been adequate9 N/A • Nonresponse - - a) Was the nonresponse ratc unacceptably high? J b) Was it different in cases and controls? N/A c) Were refusals among controls more likely to be smokers or ETS exposed? N/A • Dropouts (enhort study only) - -- - - al Was the dropout rate unacceptably high? .~ b) Did the dropout rate differ by ETS exposure status? .~ • Proxy response N/A - a) Was proxy response rate greater in cases than controls? b) Was there greater than 10'/% proxy response? • Was there adequate validation of active smoking and ETS exposme? d • Were the interviewers blinded to ease/conrrol stetus? N/A TOA3CERT.IINIFTREDATAARECOP181STENf . . • W as there crude evidence of dose-response? V • Wastherestatistieallysignificantevidenceofdose-msponse? N • Were there similat results in subgroups studied? (Results tAarn go in the same direction; if not, do the differences make sense?) N/A • Was temporal relationship clearly demonstrated? . se T/)A5C'.EkTAINIFSTA'tl$'j1CAf.311;fRQUSrANALYSISA REAPPROPRIATE . . . . .. . . •Wasthereuseoftwo-sidedtestsofsignificana? . r • Were statistical tests or confidence intervals reported? .~ • Were statistical methods and adjustments appropriate? J

I i I I I I Svendsen et al. 1987 Page 3 Study Results: Mean thiocyanate levels at baseline and the average of baseline and averaged over all annual follow-up visits were similar. Expired air carbon monoxide was statisticallv different for the men whose wives smoked compared to the men whose wives did not smoke at the third annual examination. Relative risk estimates, wife who smoked compared to wife who did not smoke, and their 95% confidence intervals for never smoking men RR (95%CI) Death from any cause Unadjusted 1.96 (0.93-4.11) Adjusted 1.94 (0.91-4.09) Coronary heart disease death Unadjusted 2.11 (0.69-6.46) Adjusted 2:L3 (0.72-6.92) Fatal or nonfatal coronary heart disease Unadjusted 1.48 (0.89-2.47) Adjusted 1.61 (0.96-2.71) Although workplace data were limited, the relative risk for coronary heart disease death was 2.6 (95%CI, 0.5-12.7); for fatal or nonfatal coronary heart disease, the relative risk was 1.4 (95°/aCI, 0.8-2.5). Dose-Response: No results are reported by level of smoking. Ascertainment of Temporal Relationship: Presumably, those who had overt cardiovascular disease were excluded from the study, so it is reasonable to assume that exposed and unexposed were free of disease at entry into the study. I

,.f Butler 1988 Page 3 Other Potential Limitations (Recognized by authors): Statistical Methods: Study Results: Possibility that under-ascertainment of cases occurred. Potential misclassification of passive smoking exposure. No information on year subject started or stopped smoking, number of cigarettes regularly smoked in the home, and detailed marital history. Mantel-Haenszel approach for stratified person time incidence rate data used to estimate summary rate ratios and 95% confidence intervals. For outcomes will small numbers of cases, the maxiumum likelihood estimator and an exact method for calculation of confidence intervals is used. For mortality outcomes, Cox's proportional hazards model is used. Spouse Pairs Cohort: Risk ratios and 95% confidence limits of ischemic heart disease mortality (1976-1982) for nonsmoking females Husband Smoking Status Crude RR Adjusted RR Never 1.00 Past 1.42 0.96 (0.55, 1.66) Current AHSMOC Cohort: 1.15 1.40 (0.51, 3.84) Relative risk ratios and 95% confidence intervals of ischemic heart disease mortality (1976-1982) for selected ezposure factors (nonsmoking females) Factor Crude RR Adjusted RR Yrs. lived with smoker ~ None 1 00 1 00 A . . G11 1-10 yrs. 1.06 1.46 (0.70,3.08) ~ i 1+ vrs, 1.32 1.53 (0.92, 2.54) ~ W W W l~

I I I Svendsen et al. 1987 Page 2 Definitions of Outcome(s): Procedures used to Validate Outcome(s): Blinding: Confounders: Other Potential Limitations (Recognized by authors): Statistical Methods: Classification of cause of death was performed by a committee of three cardiologists unaware oftreatment assignment or passive smoking status. Coronary heart disease deaths were classified as (1) documented myocardial infarction, (2) sudden death within 60 minutes, or between I and 24 hours of symptom onset, without documented myocardial infarction, (3) congestive heart failure due to coronary heart disease, or (4) death associated with surgery for coronary heart disease. Results are also presented for fatal or nonfatal coronary heart disease. Hospital records, physicians' reports, next-of-kin interviews, death certificates, and autopsy reports were used when available. The authors state that endpoints were assessed without knowledge of passive smoking status. Adjusted for age, baseline blood pressure, cholesterol, weight, education (as a titeasute of SES), and drinks per week. The men were not randomized to wives who smoked and to those who did not smoke. A man who did not smoke married to a woman who smoked may have had other tuuneasured health behaviors that increased morbidity and mortality. Student's t-test (two-sided) or 2x2 Chi squre test to test differences in baseline characteristics and differences in the means between the two rroups for thiocyanate and expired air carbon monoxide. Tests for a dose effect of smoking exposure were performed using regression models with number of cigarettes smoked per day reported by wife as an independent variable. Relative risk estimates for men whose wives smoked compared with men whose wives did not smoke, for the endpoints death from any cause, coronary heart disease death, and fatal or nonfatal coronary heart disease event were calculated using Cox proportional hazards model with Breslow's approximation. I

I I I 1 I 1 I ABSTRACT FORM EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE Reference: Lee, P.N., Chamberlain, J., and Alderson, M.R. 1986. Relationship of passive smoking to risk of lung cancer and other smoking-associated diseases. Br. J. Cancer 54:97-105. [UK] Type of Study: Case-control Source of Study Population: Cases were patients with ischemic heart disease from medical (including chest medicine), thoracic surgery, and radiotherapy wards of a large hospital. There were a total of 507 married hospital inpatients with ischemic heart disease (286 males and 221 females) who completed passive smoking questionnaires. Unclear how many of these were included in a subsequent follow-up of spouses of non-smokers. Controls were patients from the same hospital, without ischemic heart disease, individually matched to cases on sex, age, hospital region, and, when possible, hospital ward and time of interview. Participation Rates: Authors present figures on refusals, but for the total lung cancer group of cases; no figures are presented for patients with ischemic heart disease. Percentage of Proxy Respondents: Unclear; since this was an in-patient study probably all data were derived from the cases themselves. Definitions of Exposure: In-hospital administered questionnaires were used to gather data on exposure to passive smoking. Questions ascertained length of marriage of case; number of manufactured cigarettes per day smoked by the spouse both during the last 12 months of marriage and also at the period of maximum smoking during the marciage; and whether the spouse ever regularly smoked hand-rolled cigarettes, cigars or a pipe during the marriage. Patients were also asked to quantify, according to a four-point scale, the extent to which they were regularly exposed to tobacco smoke from other people prior to coming into the hospital in four different situations: home, work, during daily travel, and during leisure time.

I Lee et al. 1986 Page 2 Procedures used to Validate Exposures: Blinding: Confounders: Other Potential Limitations (Recognized by authors): Statistical Methods: Stult, Results: Dose-Response: Ascertainment of Temporal Relationship: -• The follow-up study was designed to compare information on spouses' smoking habits obtained first-hand with that obtained second-hand during the in-patient interviews. Not specified. Confounders were not specified, although the authors mention that information on confounders was collected in the main questionnaire. None. The data were examined as a 2xKxS table. Results presented are for the combined strata; relative risk assessed using the Mantel-Haenszel estimate with the significance of its difference from a base level (risk 1.0) and/or the dose-related trend. Nu significant relationship of any index of passive smoking was observed with ischemic heart disease. For the sexes combined, the relative risk in relation to the spouse smoking during the whole of the marriage was 1.03 (95%CI, 0.65-1.62) for ischemic heart disease. These data were standardized for age, for spouse smoking, and whether the marriage was ongoing or ended. [Note that scant details are presented on ischemic heart disease; this was principally a lung disease study]. No trends were noted by combined exposure index, and risks were under 1. No mention of ascertainment of temporal relationship.

Butler 1988 Page 4 ! ~ 1 Yrs. worked with smoker ..._..._ . None 1.00 1-10 yrs. 0.71 11+ yrs. 0.88 1.00 1.85 (1.00. 3.44) 1.86 (0.99. 3.48) Relative risk ratios and 95% confidence intervals of ischemic heart disease mortality (1976-1982) for selected exposure factors (nonsmoking males) i Factor . Crude RR Adjusted RR Yrs. lived with smoker None 1,00_ 1.00 1-10 yrs. 0.24 0.41 (0.13, 1.30) 11+ vrs. 0.45 0.61 (0.31, 1.19) ~ Yrs,, worked with smoker ._ None 1.00 1.00 1-10 yrs. 0.62 1.26 (0.68, 2.33) 11+ yrs. 0.56 0.76 (0.37, 1.55) Dose-Response: No calculation for test off trend. Data on dose-response may be derived from the RRs for work and home exposures. 'i Ascertainment of Temporal Relationship: Individuals were free of the disease outcomes of interest at the commencement of the study and prior to the period of follrw-up for diagnosis of the disease.

I I I I I I I Butler 1988 Page 2 Participation Rates: The response rate to the hospital history form was 86.% to 94.7%. 1.2% of the total incidence population was not found and was classified as lost to follow-up at the time of their last response to the form. Percentage of Proxy Respondents: The Census Questionnaire was completed by the head of each household, and therefore, was a proxy response for other members of the family. The reliability of these responses were checked with a face-to-face interview of a random sample of households; discrepancy was less than 6% for items other than educational leN el and age of baptism. Definitions of Outcome(s): Ischemic heart disease mortality coded by nosologist according to ICD 9th Revision. Procedure3 used to Validate Outcome(s): Mortality was ascertained using computerized record linkage with the California Death Certificate file, computerized record linkage with the ~ National Death Index, and notification of death from Seventh-Day Adventist church records by the local church clerks. In addition, mail responses to the annual Hospital History Form or attempts to contact non- respondents for this hospital history form revealed some deaths that were not ascertained by the other methods. Blinding: Not specified. Confounders: RRs were adjusted for age. Spouse Pairs Cohort: History of hypertension and diabetes related to risk of ischemic heart disease mortality. AHSMOG Cohort: For females, history of hypertension and diabetes related to risk of ischemic heart disease mortality. For males, history of hypertension, diabetes, and a lower educational level related to risk of ischemic heart disease mortality.

'. CRITERIA BY WHICH TO ASSESS STUDY QUALITY Reference: Butler 1988 I I I I I I 1 I I I QUALITY CRITERIA 1'es No Cannot Ascertain RELATFYERLSK • Is crude relative risk elevated? Y • Is adiusted relative risk clevated7 r • Is adjusted (or crude if adjusted not available) statistically significant by two-sided test, p<.05? r TOR[fLE.OiFTCONFOI7NDING. • Have age and gender been adequately controlled? . d • Have at least 4 of the 5 fol lowing risk factors been conuol led for: family history, blood pressure, cholesrerol. diabetes/glucose intolerance, left ventricular hypertrophy on EKG, or any specific EKG abnotmality^. .~ • Has the study ascertained which of the above risk factors are potential confounders and appropriately contmlled for them? J .TORU6t!OtFTSICNiRFCAiVTBtAS • Has the choice of the control group been adequate? N/A • Nonttsponsc . . a) Was the nomesponse rate unacceptably high? t/ b) Was it different in cases and controls? . ..... . . N/A c) Were refusals among controls more likely to be smokers or ETS exposed? N/A • Dropouts (cohort study only) N/A .) Was the dropout rate unacceptably high? . . .~ b) Did the dropout rate differ by ETS exposure status? r • Proxy response N/A .) W as proxy response rate greater in cases than controls? b) W at there greater than 10'/% proxy response? • Was dtem adequate validation of active smoking and ETS exposure? J • W ere the interviewers blinded to case/control status? N/A .. . TOABCEATSINIFTfIEITATAARECO?*SI3fEhT • Was there crude evidence of dose-respottu? (years lived with smoker-women) Y • Wasthereststistipallysignificantevidenceofdose-responses d • Were them similaYrosults in subgroups studied? (Results that go in the same direction; if not, do the differences make sense?) (mcnvs..wpmen) d • Was temporal relationship clearly demonstrated? r TDASCERTAINtFST4T1Sf(CA{,ME"ITInUS~ANALySF.SAREAPPRqPR1AT$ . . • Was there use of two-sided tests of significenee? r • Were statistical tests or confidence intervals reported? . d * Were statistical methods and adjustments approprlate? l~ s I

ABSTRACT FORM EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE Reference: Svendsen, K.H., Kuller, L.H., Martin, M.J., and Ockene. J.K. 1987. Effects of passive smoking in the Multiple Risk Factor Intervention Trial. Am. J. Epidemiol. 126(5):783-795. Type of Study: Prospective Source of Study Population: Multiple Risk Factor Intervention Trial (MRFIT). Male participants. 35- 57 years old and in the upp:r 10-15% risk score for heart disease, who reported at entry that they had never smoked tobacco products were classified according to the smoking status of their wives. Men were randomly allocated into special intervention group or to usual care. Definition of Exposure: Analyses were restricted to 1,245 married nonsmoking men which included never smokers and ex-smokers who quit prior to entry into the study. Exposure was defined based on smoking histories obtained for each of the participants before randomiration. The history included not ~ only their own smoking history but also that of their wives, family members, and coworkers. Detailed histories were also collected on those participants who smoked. The smoking status of the wife was used as an index of passive smoking exposure for the men who did not smoke. Altogether, 23% of the men who did not smoke were exposed at home to the environmental tobacco smoke of their wives. Serum thiocyanate was measured during screening and at each anraal visit. At the third and sixth annual examinations, expired air carbon monoxide was measured. Length of Follow-up• Average of 7 years. Participation Rates: -- The authors state that follow-up was complete for all MRFIT men. Percentage of Proxy Respondents: Unspecified, although the authors state that proxies were used.

Reference: Type of Study: Source of Study Population: Participation Rates: ABSTRACT FORM EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE Palmer, J.R, Rosenberg, L., and Shapiro, S. 1988. Passive smoking and myocardial infarction in women. CVD Epidemiol. News. 43:29. [Abstract] [USAj Case-control Hospital-based study of past oral contraceptive use and myocardial infarction in women aged 20 to 64. There were 336 married cases and 799 married controls. No data on refusals or participation rates are provided. Percentage of Proxy Respondents: No data are provided. Definitions of Exposure: Procedures used to Validate Exposures: Blinding: Confounders: Other Potential Limitations (Recognized by authors): Exposure is determined by husband's smoking habits. None reported. No information on blinding with regard to case/control status is provided. None reported; investigators report that the observed trend was not accounted for by the known risk factors for myocardial infarction. None reported. Statistical Methods: Not reported; no confidence intervals are reported. I .I

CRITERIA BY WHICH TO ASSESS STUDY QUALITY I i I I I I Reference: Palmer et al. L988 (abstract) CRITERIA aoaoi Cannot Yes No Ascertain REFATIVERFSK::., • Is crude relative risk elevated? ... .. _ V • Is adjusted relative risk clcvated? .. . . . V • Is adjusted (or crude if adjusted not available) statistically significant by nvo-sided test, p<05? V TORULEOUTCONFOUNDIFIG..... • Have age and gender been adequately controlled? .- . . V • Have at least 4 of the 5 following risk factors been controlled for: family history, blood pressure, cholesterol, diabetesiglucose intolerancx, left ventricular hypertrophy on EKG, or any specific EKG abnonnality? V • Has the study ascertained which of the above risk factors are potential confounders and appropriately controlled for them? V TORULEOt1TS{GIYIFIC*.YCBIAS . . ... ..._ ,,.. . . • Has the choice of the control group been adequate? V • Nonmsponse a) Was the nonresponse rate unacceptably high? V b) Was it different in cases and controls? . V c) Were refusals among controls more likely to be smokers or ETS exposed? . V • Dropouts (cohort study only) N/A ~ a) Was the dropout rate unacceptxb)y high? b) Did the dropout rate differ by ETS exposure status? . . I • Proxy response a) Was proxy response rate greater in cases than controls? . V b) W as there greater than 10% proxy response? V • Was there adequate validation of active smoking and ETS exposure? V • Were the interviewers blinded to casNcontrol status? V 'TO.SSCERTAINIF'f'HEDATAA0.ECO':S78TEVT . .. . • Was them crude evidence ofdtne-response? V • Wastheresratisticallysigni8cantevidenceofdose-response? , a • Were there similar.resutts in subgroups studied? (Results that go in the same direction; ifhot, do the differences make sense?) N/A • Was temporal relationship clearly demonstrated? V :TO:ASCF.R'PAINiFSTATI-STiCALSiET4iOOS/ANALY:StSAREAPFROPRLITE •Wasthereuseoftwo-sidedtestsofsignificance? V • Were statistical tests or confidence intervals reported? V • Wcre statistical methods and adjustments appropriate? V 6

' 1 ,. 1 I I i Palmer et al. 1988 Page 2 Study Results: With a reference category of nonsmoking women married to nonsmoking men, the relative risk estimate for nonsmoking women whose husband smoked was 1.2; for women who smoked less than 25 cigarettes per day. the estimates were 2.9 (nonsmoking husbands) and 3.9 (smoking husbands); and for women who were heavy smokers, the estimates were 6.3 (nonsmoking husbands) and 8.3 (smoking husbands). Dose-Response: The observed trend was not accounted for by the known risk factors for myocardial infarction. Ascertainment or Temporal Relationship: Insufficient data reported.

ABSTRACT FORM EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE _I Reference: Butler, T.L. 1988. The Relationship of Passive Smoking to Various Health Outcomes Among Seventh-Day Adventists in California. Doctoral dissertation, University of California, Los Angeles. (USAI Type of Study: Prospective Source of Study Population: The study had two objectives: (1) to determine whether nonsmoking women in the Adventist Health Study cohort who are married to current or former cigarette smokers have an increased risk of ischemic heart disease and (2) to determine whether long-term exposure to passive smoking in the home and work environments from subjects of the Adventist Health Smog Study (AHSMOG Study) leads to an increased risk of ischemic heart disease. The spouse pairs cohort was composed of 11,060 pairs, married and living together in 1976. The AHSMOG population was composed of 6,467 subjects who competed the AHSMOG questionnaire in 1977. I I I I I I I I I I Definition of Exposure: Passive smoking exposure for the "spouse pairs" was based on the husband's smoking status in marriage. Those classified as not smoking during marriage included neversmokers and past smokers when the age of marriage was equal to or greater than the age of baptism. Those classified as smoking during marriage included current smokers, past smokers when the age of baptism was greater than the age of marriage, and past smokers who are not Adventist baptized. For the AHSMOG cohort, ETS exposure was based on number of years lived with and the number of years worked with a smoker. Length of follow- up: For 6 years Hospital History Forms were mailed to every member of the incidence population.

I I I I I I I CRITERIA BY WHICH TO ASSESS STUDY QUALITY Reference: Helsing et aI. 1988 QUALITY CRITERIA S'es No Cannat Auertaio RELATIYERL4K • Is cmde relative risk elcvated? V • Is adjusted relative risk clevated? V • Is adjusted (or crude if adjusted not available) statisncally significant by two-sided test p<,05? V TO RULEUt[YCtlNFOUNBiNiG ... . . .... . . . . ... ;:~. -. • Have age and gender been adcauately conbolled? V • Have at least 4 of the 5 following risk factors been controlled for: family history, blood pressure, cholesterol, diabetes/glueose intolerance, left ventricular hypenrophy on EKG, or any specific EKG abnotmality? V • Has the study ascertained which of the above risk factors are potential confounders and appropriately controlled for them? . . . . V TURULEOICTSIGABIFICANTSiAS • Has the choice of the control group been adequate? N/A • Nonresponse a) Was the nonresponse rete unacceptably high? V b) Was it different in cases and controls? N/A _ cl Were refusals among controls mom likely to be smokers or ETS exposed? N/A • Dropouts (cohort study only) . _ a) Was the dropout rate unacceptably high? r bl Did the dropout rate differ by ETS exposure status? .' • Proxy response a) W as proxy response rate greater in cases than controls? N/A b) Was therc grcater t~ n 10Y. proxy response? V • Was therc adequate validation of active smoking and ETS exposure? V • Were the interviewers blinded to cacdcontmi status? N/A .t'QASCS,RTAIN.FET71EItAT.4.ARE:COHSISTEN7 .. ...: . .. • Was there crude evidence of dose-response? V - • Was there statistically significantevidence of dose-response? (women only) V • Were there similaftcsutts in subgroups studied? (Results that go in the same direction; if not, do the differences make sense?) (men vs. women) - V • Wa5 temporal relationship clearly demunsttated? V TD'ASC6BTAFNIFS[ATIS17C+lLMETN6DSYANALkSIS.UiEAP#ROPRLATE .... :. : .. .. •Wastheteuseoftwo-sidedtestsofsfgnifcance? V • Were statistical tests or confidence intervals reported? V • Were statistical methods and adjustments appropriate? (may have underestimated variability of rates) V 7

CRITERIA BY WHICH TO ASSESS STUDY QUALITY I I I I I I , I Reference: Hole et al. 1989 Ql.'ALITY CRITERIA yes No Cannot Ascertain R£F;AT1V£RFSK , _ -. . .. . . -.-' . . . ... • Is crude relative risk clcvatcd? .1 • Is adjusted relative risk elevated? J • Is adjusted (or crude if adjusted not available) statistically significant by two-sidcd test, p<.05? .~ TO RULE O[YfOON9OUNDING- ~ - . . .. . ... ... • Have age and gender been adequately controlled? .. . . ., • Have at leatt 4 of the 5 following risk factors been controlled for: family history, blood pressure, cholesterol, diabetes/glucose intoierance, left ventncular hypertrophy on EKG, or any specific EKG abnonnalitv?. r • Has the study a5oeruuned which of the above risk factors are potential confounders and appropriately controlled for them? N TO RitLEOUTSIGNIFiCANf $3A5 . ... . . . . . . • Has the choice of the control group been adequate? N/A • Nonresponse , ... a) Was the nonresponse rate unacceptably high? s, b) Was it different in cases and controls? , . . . . . NIA c) Were refusals among controls more likely to be smokers or ETS exposed? *1/A • Dropouts (cohort study only) . . . a) Was the dropout rate unacceptably high? O b) Did the dropout rate differ by ETS exposure status? . .~ • Proxy response N/A - a) Was prozy response rase greater in cases than controls? b) Was them greater than IoY. proxy response? . . • Was them adequate validation ofaqtive smoking and ETS exposure? V • Were the interviewers blinded to case/control status? . N/A TCYASCL'R-,FAINIFTIiEDATAAR€CONSIB'[EhT • Was there crude evidence ofdose-responsa? . . .. r • Wasthemstatistieallysignificantevidenceofdose-response? . . . . - r • Were them similarresults in subgroups studied? (Results that go in the same direction; if not, do the differcnces make sense?) (Inconsistency of risks for active and passive smoking) .~ • Was temporal relationship clearly demonstrated? d TOASCEtYEAINIFSTATLSTIGIE.ME741ODS/ANALY'SISARE.APPRGPRIATE. ' • Was them use of two-sided tests of significance? o/ • W ere statistical tests or confidence inteivals reported? d • Were statistical methods and adjustments appropnaa? N 9

CRITERIA BY WHICH TO ASSESS STUDY QUALITY I I I I I I Reference: Humble et al. 1990 QUALITY CRITERIA Yes No Cannot Ascertain RELiTlV£R}$K -. ._.... . .c _.::. ,,. . • Is crude relative risk elevated? . . . . . . N • Is adjusted relative risk elevated? d • Is adjusted (or crude if adjusted not available) statistically signiicant by two-sided test, p<_05? . d TORULEOU7CONFOUNDINC - ' - • Have age and gender been adequately controlled? . d • Have at least 4 of the 5 following risk factors been controlled for fartiily history, blood p¢ssure, cholesterol, diabetes/glucose intolerance, left ventricular hypertrophy on EKG, or any specific EKG abnormality? .e • Has the study ascertained which ofthe above risk factors are potential confounders and appropristely controlled for them? t/ TORULEOUTSIGN(FICA1'FHLLS : . . ..: . -.... . • Has the choice of the control group been adequate? N/A •Nonresponse . . - . -~ a) Was the nonresponse rate unacceptably high? r b) Was it different in cases and controls? N/A _ c) Were refusals among controls more likely to be smokers or ETS exposed? N/A • Dropouts (cuhortstudyonly) .. . . .q Was the dropout rate unacceptably high? +~ b) Did the dropout rate differ by ETS cxposure stuus? J • Proxy response N/A . _) W as proxy response rate greater in eases than controls? b) Was thetn greater than 10'/% proxy response? • Was there adequate validation ofactive smoking and ETS exposure? . . - a~ • Were the interviewers blinded to case/cOntrol statas? N/A TOABCERT.41lYIFTILEDATAIRECOYSYSCEiT . • Was there crude evidence ofdose-response? • Wastherestatisticallysignificantevidenceofdose-responses d • Were there similar'results in subgroups studied? (Results that gq in the same direction; if not, do the diffcrences make sense?) (racial/SES diffemnces) V • Was temporal relationship clearly demOnstnted? s/ 't'OASC£kTA1NIFSTATI$TiCAL fstETHOD5EANALY5F5.AAE APPROPRIATE ... • Was them use of two-sided tests of significanca? J • Were statistical tests or confidence intervals reported? N • Were statistical methods and adjustments zppropriate? J 13 I

I ';.1 I ! I I I I I Helsine et al. 1988 Page 2 Procedures used to Validate Outcome(s): Blinding: Confounders: Other Potential Limitations (Recognized by authors): Statistical Methods: Death certificates of Washington County residents who died between July 1963 and July 1975 were matched against the census. Not specified. Age, housing quality, marital status, years of schooling. No smoking data obtained in the 1963 census, so no provision can be made for changes in smoking habits as a result of publicity about health effects of smoking. No data on changes in the household composition which may have occurred prior to or after 1963. Did not control for diet and exercise which might differ in families with and without smokers and such differences could have influenced the findings. Death rates were calculated as deaths in 12 years per 1,000 midpoint population, adjusted for age, housing quality, marital status, and years of schooling by the binary variable multiple regression procedure.

He et ai. 1989 Page 3 I Crude associations between husband's average daily cigarette consumption, passive smoke exposure years, cumulative passive smoke amount index, and risk of getting CHD I Husbands' daily cigarette consumption Crude OR 0 1.0 I <20 2.3 >20 6.9 (s.s. x2)  Passive smoke exposure years 0 1.0 s10 1,9 ! s20 3.1 >20 5.5 (s.s. X2) ~ Cumulative passive smoke index (years) ~ 0 1.0 1-199 1.5 200-399 2.3 ~ 400-599 5.1 (s.s. X2) 600+ 12.7(ss. x2) Ascertainment of i i !I Temporal Relationship: Authors did not ascertain that exposure preceded disease. No indication of latency. 1 I

1 r I I I I I CRITERIA BY WHICH TO ASSESS STUDY QUALITY Reference: Dobson et al. 1991 QUALITYCRITERIA Yes No Cannot Asceruin RELATIYERkSK_.., ... . . . _.: ._; . . .. .. :.:. . .. r ._ . • Is crude relative nsk elevated? V. • Is adjusted relative risk elevated? Iwomen only) V • Is adjusted (or crude if adjusted not available) statistically significant by two-sided test, p<,05? (womenonly) V TORULEOUTL`ONF~OUA`DENt'i. • Have age and gcader been adequately contmlled? V • Have at least 4 of the 5 following risk factors been controlled for: family history, blood pressure, _ cholesteroY, diabetes/glucose intolerance, left ventricular hype.trcphy on EKG, or any specific EKG abnormaiity? V • Has the study ascertained which of the above risk factors aro potential confounders and apprupriately controlled for them? V TO RUl.E.OUTSICNIFIC:SF?I"H1A8 . .. .. . . .. .... : . . • Has the choice of the control group been adequate? (but undcr-represented smokers) V • Nonrasponse . a) Was the nonresponsc rate unacceptably high? V b) Was it different in cases and controls? . . .. . V ~ c) Were refusals among controls more likely to be smokers or ETS exposed? V • Dropouts (cohort study only) N7A . . . .. . a) Was the dropout rate unacceptably high? b) Did the dropout rate differ by ETS exposure status? • Proxy response . a) Was proxy response rete greater in cases than controls? V b) Was there greater than IOY% proxy response? . . V • Was these adequate validation of active smoking and ETS exposure? V • Were the interviewers blinded to case/control status? V TOASCERT6INIFTHEDATA,ARECONSfStE"iT • Was there ciude evidence of dose-response? V • Was there statistically significant evidence afdose-response? - . r • Were there similar results in subgroups studied? (Results that go in the same direction; if not, do the differences make senu?) (males vs. females; home vs. work) V • Was temporal relationship clearly demonstrated? . V TOA3C$R7A1N iF'STA'17.5'f1Ctf, vtET'(IOWANAI:YSI.SARE.APpROPRLATE . • Was thera use of two-sided tests ofsigniFlcance? . V • Were statistical tests or confidence intervals reported? V • Were statistical methods and adjustments appropriateP V 14

Helsing et al. 1988 Page 3 Study Results: Deaths from arteriosclerotic heart disease among nonsmokers exposed or not exposed to tobacco smoke in the home, adjusted relative risks and 95% confidence intervals (all ages) _ Passive Smoking Score Adj. RR 95%CI Men 0 1.00 1+ - 1.31 (1.1, 1.6) 1-5 1.38 (1.1, 1.8) 6+ 1.25 (1.0, 1.6) Women 0 1.00 1+ 1.24 (1.1, 1.4) 1-5 1.20 (1.0, 1.4) 6+ 1.27 (1.1, 1.5) Dose-Response: For men, the trend with increasing exposure is negligible. For women, the trend of increasing mortality with increasing levels of exposure in the home (Cochran Chi-square) is statistically significant. Ascertainment of Temporal Relationship: No information on health status of population at baseline reported.

CRITERIA BY WHICH TO ASSESS STUDY QUALITY ~~ Reference: Jackson 1989 , 1 .1 I QUALITY CRITERIA Yes No Cannot Ascenain RE7.ATIVERLSR.-. . . . . .. ~... .. • Is crude relative risk elevated? V . • Is adjusted relative risk elevated? . . . . . . . . . . . V • Is adjusted (or cmde ifadjusted not available) statistically significant by two-sided test, p<A5? V TORULEOUTCONFOt31VDiHG , . .. .. . . . ... . • Have age and gender been adequately controlled? V • Have at least 4 of rhe 5 following risk factors been controlled foc family history, blood pressure, cholesterol, diabetes/glucose intolerance, left venvicular hypertrophy on EKG, or any specific EKG abnonnaliry? .. . . .. . . V • Has the study ascertained which of the above risk factors are potential confounders and appropriately controlled for them? - - V TORULEOUTSiGtY1B[C.AH°fH1aS • Has the choice of the control group been adequate? V • Nonresponsc : ...:. . . . . . a) Was the nonresponse rare unacceptably high? _ . V b) Was it different in cases and controls? . . _ . . s/ c) Were refusals among controls more likely to be smokers ar ETS exposed? V • Dropouts (cohort study only) N/A - a) Was the dropout rate unacceptably high? b) Did the dropout rate differ by ETS exposure status? '_ • Proxy response a) Was proxy response rate greater in cases than conarols? V b) Was there greater r.han 10'/% proxy responsc? V • Was therc adequate validation of active smoking and ETS exposure? V • Wem the intervmwers blinded to cax/cantrol status? V .TOASCERT.AINfFTffEDATAeIRE.CO7KSLSTEhT • Was there crude evidence of dose-response? (women only) V . • Wastherestatistiwllysigniflcantevidenceofdose-response^ V • Were there similar.results in subgroups studied? (Results that go in the same direction; if not, do the differences make sense?) (males vs. females) V • Was temporal relationship clearly demonstrsted? . . . V TOASC.ER'CAIV(f~fh'C4S1'tC.AL.tifTliOIIS!ANALYSiSAREAPPRO)'RL4TE •Wasthereuseoftwo-sidedtesrsofsignificance? . V • Were statistical tests or confidence intervals reported? V • Were statistical methods and adjustments appropriate? V 10

CRITERIA BY WHICH TO ASSESS STUDY QUALITY I I ar Reference: He et al. 1989 QUALITYCRITERIA }'es tio Cannor Ascertain R£F«ATLY£RISK-.. '- . __..- . - ... . . • Is drudc relative risk elevated? y • Is adjusted relative nsk elevated? - si • Is adjusted (or crude if adjusted not avaiiablcl statistically signifmant by two-sided test, p<05? s, TORULE.OUTCONFOUNDIN6 - - ~ • Have age and gender been adequately controffed? ., • Have at least 4 of the 5 following risk factors been controlled for: family history, blood pressum, chofestcrol. diabetes/glucose intolerance, le8 ventricular hypertrophy on EKG, or any specific EKG abnormality? O • Has the study ascertained which ofthe above risk factors are potential confounders and appropriately controlled for them? N TO Rl7i:E OUT SIGNiBfC-!NT atAS • Has the choice ofthe control group been adequate? d • Nonresponse . a) Was the nonresponse rate unacceptably high? r/ b) Was it different in cases and controls? r c) Were refusals among controls more likely to be smokers or ETS exposed? r •Dropouts(cohortstudyonly) N/A a) W as the dropout rate unacceptably high? b) Did the dropout rate differ by ETS exposure status? • Proxy response - a) Was proxy response rate greater in cases than controls? . sI b) W at therc greater than 10°/6 proxy rcsponse? t/ • Was there adequate validation ofacNve smoking and ETS exposure? .1 • Were the interviewers blinded to catehontrol stauu? d TO.KfiCERTAIN IF Tt{E DATA AA£ tQNa,iSTEh7 :: . • Was there crude evidence of dose-Rsponse? aI • Wastherestatistieallysignifrcantevidenceofdose-response? .1 • Were theresimilar t4sults in subgroups studied? (Results that go in the same direction; if not, do the differences make sense?) r • W as temporal relationship cleariy demonstnted? y TOASCERTAIiY1Y5TATI5TICAJ.ME•{'}{QO5iANALY$ISAREAPPROPR14jE .. .. •Wasthercuseoftwo-sidedtestsofsignificancc? ------- Y • Were statistical tests or confidence intervals reported? y • Were statistical methods and adjustments appropriate? (inconsistencies on paper) rI S

CRSTERSA BY WH1CH TO ASSESS STUDY QUALITY I I I ! I I I I I I I I I Reference: Hirayama et aL 1984 (updated 1990) QUALITY CRITERIA Yes No Cannot Ascertain RELATSYERiSR • Is crude relative risk elevated? r • Is adjusted relative risk elevatedP y • Is adjusted (or crude if adjusted not available) statistically significant by two-sided test p',05? s~ TO RULE OUTCONFOUNDING ~ - .. • Have age and gender been adequately controlled? -- - 61 • Have at least 4 of the 5 following risk factors been controlled for family history, blood pmssure, choiesterol• diabetes/glucose intolerance, left ventricular hypeuophy on EKG, or any specific EKG abnormality? . d • Has the study ascertained which of the above risk factors are potential confounders and appropriately controlled for them? V TO RULE OUT SiGND?aCAIYT BIAS . .. .. : .. .. .. .. ..... . . ... . - ; • Has the choice of the control group been adequate? N/A • Nonrcsponse . . a) Was the nonresponse rate unacceptably high? J b) Was it different in cases and controls? . N/A _ c) Were refusals among controls more likely to be smokers or ETS exposed? N/A • Dropouts (cohort study only) a) Was the dropout rnte unacceptably high? r/ bl Did the dropout rate differ by ETS exposure staws? .. . -. r/ • Proxy resp(nse . N/A .. a) Was proxy response rete greater in ca<es than controls? b) Was there greater than 10•/% proxy response? . • Was them adequate validation of active smoking and ETS exposurc? . . se •WeretheinterviewersblindedtocattJcontrnlsbtus? N/A '.TOAS!CBRPA[N [FTHEDATAARF COISSISTE.e'1T . . . • Was there cmde evidence ofdose-response? se • Wasthercstatisticallysignificantevidenceofdoseaesponse? - V • Were there similar iesulta in subgroups studied? (Results that go in the same direction; if not, do the differences make sense?) (does not separafe out ex-smokcrs and 1-19 cigs/day) N • Was temporal relationship clearly demonstrated? V TO ASCERTACNTF'..Si1TFSPiCAi. \tE1HOFStAPtALYSLCr1RE AYpROPRLATE . , .. . • Was there use oftwo-sided tests ofsignificance? .~ • Were statistical tests or cnnftdence intervals reported? 1~ • Were statistical methods and adjustments appropriate? J 12

CRSTERlA BY Wt-tICH TO ASSESS STUDY QUALITY I I I , - 1 I I I I I ' Reference: Sandler et al. 1989 QUALITY CRITERIA Yes No Cannot Ascerrain .. .._.. .-........ ........... .. ..... ...... ..,...__ _... .. ... .. . RELATIVERISX • Is crude relative nsk elevated7 - sr • Is adjusted relative risk elevated? s~ • Is adjusted (or crude if adjusted not available) statistically significant by two-sided test, p<05? t/ TO6ULEQDTCIINFRUNDDVG: • Have age and gender been adequately controlied? Y • Have at least 4 of the 5 following risk factors been controlled far: family history, blood pressure, cholesterol, diabetes/glucose intolerance, left ventricuiar hypertruphy on EKG, or any specific EKG abnormality? d • Has the study ascertained which of the above risk factorsase'potential confounders and appropriately controlled forthem? .. .. . . . . 1/ .. ,.._ .. _. . . ~'i'URULEi3UTSIGCFIFFCADFf'HL48 • Has the choice of the control group been adequate? N/A • Nonresponse .' a) Was the nonresponse rate unacceptably high? J b) Was it diffen:nt in cases and controls? N/A _ c) Were rcfusals among controls mom likely to be smokers or ETS exposed? N/A • Dropouts (cohart study only) a) Was the dropout rate unacceptably high? . . - - r bl Did the dropout rate differ by ETS exposure status? I v • Proxy response . a) Was proxy response rate greater in cases than controls? N/A b) Wastheregreaterthan10Xproxyrasponse? .~ • Was there adequate validation of active smoking and ETS exposure? . . t/ • Were the interviewers blinded to csse/eonaol status? . N/A -TpaSCeRTAL'VIF'sHEDATAA3tBCOMLSTENT . ... .. . .. .. • Wasdiere crude evidence ofdose-responsef r • Wasthemstatisticallysignificantevidenaofdose-response? ' d • Were there similar rzsults in subgroups studied? (Results that go in the same direction; if not, do the differenus make sense?) (men vs. women) . . se • Was temporal relationship clearly demonstrated? . d TOASE'-ER.TAICY IF STATISTICAL METiff3D8tAtfALI'SLS :lRE APPEOPRIATE . ... . .. • Was there use af two-sided tests of significance? r • Were statistical tests or confidence intervals reporred? d • Wete statistical methods and adjustments appropriate? (see },elsing) J 11

1 1 ABSTRACT FORM EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE Reference: He, Y., Li, L.X., Fong, C.C., Li, L.S., Chang, X.L., and Qua, Q.L. 1989. Passive smoking in females and coronary heart disease. Chin. Prev. Med. 23:19-22. [China] Type of Study: Case-control Source of 5tudy Population: Cases were 34 women with coronary heart disease, 22 with diagnosed coronary heart disease. (C.1D) and 12 with myocardial infarction, hospitalized between 1985-1987. Controls were 34 hospitalized women with endocrine problems but no symptoms of CHD and 34 randomly selected women from the general population matched for race, occupation, residence, and age (+1- 5 years). Two controls were thus selected for each case. Participation Rates: No data on refusals or participation rates are provided. Percentage of Proxy Respondents: No data are provided. Definitions of Exposure: Active Smoking: 1. Smoking at least one cigarette per day for a period of at least 1 year. 2. Spouse was defined as an ex-smoker if he had not smoked for at least 5 years at the time of interview. Passive Smoking: 1. Non-smoking wife who lived with a smoking husband for at least 5 years. 2. If the husband was a smoker before marriage, the wife's exposure was considered to begin at the time of marriage. Alternatively, the wife could become exposed after marriage if the husband started smoking after marriage. 3. Total exposure time determined by divorce, death of husband, or CD time when husband quit smoking. ~

Jackson 1989 Page 4 I I ,,I 1 I I I I 1 I I exposed to more cigarettes (i.e., from their husbands) than were men (i.e., from their wives). Crude relative risk and 95% confidence intervals of CHD associated with passive smoking at home by degree of exposure and combined disease categories Exposure*** Low High *a Men Women 1.3 2.1 (0.4, 4.2) (0.4, 11.0) 0.9 7.5 (0.4, 4.3) (1.8, 30.5) High exposure was defined as more than one cohabitant smoking or exposure to more than 7 cigarettes per day at home. The remaining people were classified as low exposure. This classification was arbitrarily chosen to ensure similar numbers irr each category. Asr -rtainment of Temporal Relationship: The analyses were limited to people with no history of myocardial infarction or angina.

1 I I I , I I I I I I I CRITERIA BY WHICH TO ASSESS STUDY QUALITY Reference: La Vecchla et al. 1993 (letter to the editor) QUALITY CRITERIA - Yes No Cannot Ascertain RELATIVERLSK,. . .. ... .. . . . . . .. ..: • Is crude relative nsk elevated? d •Isadjustedreiativeriskelevated? RR=1.2t t/ • Is adjusted (or crude ifadjusted not available) statistically significant by tsvo-sided test, p<.05^ r TORULEOBTCONFOUNDING~ `%_ . . . .. .... . . . . . . • Have age and gender been adequately controlled? t/ • Have at Ieast 4 of the 5 following risk facmrs been controlled for: family history, blood pressure, cholesterol, diabetes/glucose intolerance, left ventricular hypenrophy on EKG, or any specific EKG abnormality? . r • Has the study ascertained which of the above risk factors are potential confounders and appropriately controlled for them? " .e . ... . : . .. . . . . . TORULEOtT1'SiGNIRICANT BIAS • Has the choice of the control group been adequate? .1 • Nonrcsponse . .. . a) Was the nonresponse rare unacceptably high? V b) Was it different in eaaes and controls? . rC . c) Were refusals among controls more likely to be smokers or ETS exposed? r • bropouts (cohort study only) N/A . - a) Was the dropout rate unacceptably high? . . . b) Did the dropout rate differ by ET$ exposure status? • Proxy response . a) Was proxy response rate gmater in cases than controls? - d b) Was therc grear.• _.,an 10% proxy response? y • Was there adequate validation of active smoking and ETS exposure? V • Were the interviewers blinded to case'conuol status? Y ,TOA5CERTAIN[rTHEDATAAIIHCOrSISIB1Vi • Was there crude evidence of dose-rtsponse? . r • Was there statistically significant evidence of dose-response? J • Were there simila7results in subgroups studied? (Results that go in the same disecrion; if not, do the differences make sense?) (males vs. females; calculated OR > 1) r • Was temporal relationship clearly demonstrated? r TOASi;ERTAINI,R&TATfSTICALSfETiiODS1ANALYSISAREAPPROPRLATE •Wasthereuseoftwo-sidedtestsofsignificance? d • Were statistical tests or confidence intervals reported? . d • Were statistical methods and adjustments appropriam? W 15

CRITERIA BY WNICIi TO ASSESS STUDY QUALITY I I I I I I Reference: He et al. 1994 QUALITY CRITERIA Yes No Cannot Ascertain R£EATIV£RISK . . ' .. . - . . • Is crude relative risk elevated? . . . . . . . . . .~ • Is adjusted relative risk elevated? J • Is adjusted (or crude if adjusted not available) statistically significant by two-sided rest, p<.05? . r TO RULE.OIFi' CONFOUNDING . .... . .. . • Have age and gender been adequately controlled? .1 • Have at Icast 4 of the 5 following risk factors been controlled for: family history. blood pressure, cholesterol, diabetes/glucase intolerance, left venuicular hypenrophy on EKG, or any specific EKG abnormaiity? r . • Has the study aseenained which of the above risk factors an: potential confounders and appropriately controlled for them? . . Y TOR(fLE.OIfPBIGNllRCA14CBIAS . .. . . . .. ... . '. . ... • Has the choice of the control group been adequate? a~ •Nonrcsponse . a) Was the nonresponse mte unacceptably high? r b) Was it different in cases and controls? .. r/» ct Were refusais among controls more likely to be smokers or ETS exposed? r • Dropouts (cohort study only) N!A . ' a) Was the dropout rate unacceptably high? b) Did the drdpout rate differ by ETS exposure status? I •Proxyrespunse (probably no proxies used) . - a) Was proxy response rate greater in cases than controls? . . r b)Wasthercgreate.han10Y.proxyresponse? . ae • Was there adequate validation of active smoking and ETS exposure? (some but not enough) r • Were the interviewers blinded to casecontrol status? (only 30Y6 re-interview) r TO;ASCERTiMN WTH£DATA AaE CON3I.i'IEN€ . . .,, ,. .. . . • Wastherecrudeevideneeofdose-responus r • Wasthercstatisticallysigni5cantevidenceofdose-rtsponse?(Notaftercovariableadjustment) s~ • W ere there similar results in subgroups studied? (Results that go in the same direction; if not, do the differences make sense?) (home vs. work) r • Was temporal relationship clearly demonstrated? +e TOASCBR~FAINIF5CA'FLS"31CAE.METffODS!ANALYSIS.ABEAPPRqPR1ATE .... . ... • Was therc use of two-sided tests ofsignificancef r • Were statistical tests or confidence intervals reported? J • Were statistical methods and adjustments appropriate? r 16 I

' Hole et al. 1989 Page 2 Participation Rates: etween 1972 and 1976, 15,399 residents completed a standardized self- administered questionnaire. This constituted an 80% response rate. No fiuther details are presented on subsequent loss to follow up. ~ Percentage of Proxy Respondents: N/A a 1 Definitions of Outcome(s): Outcomes included symptoms of respiratory and cardiovascular disease, mortality, and incidence cf cancer. Mortality data were obtained from the National Health Service central register and the General Register Office for Scotland. Incidence of cancer was obtained from the cancer registry system and used to verify that the classification on the death certificate was the same as that received by the registry, Procedures used to Validate Outcome(s): See above. Blinding: The authors did not specify whether interviewers were blinded to exposure status. Confounders: Mortality was standardized for age and sex using life tables to estimate survival at 1 l years of follow-up. Estimates of relative risk and 95% confidence intervals for passive smokers compared with controls were adjusted for age, sex, socia'' class, diastolic blood pressure, serum cholesterol concentration, and body mass index using the logistic regression tnodel for cardiorespiratory symptoms. I t Other Potential Limitations (Recognized by authors): Sample size in this study did not provide sufficient statistical power to detect risks of the magnitude expected. Risk of 2.0 for ischemic heart disease seems large in comparison to that found for active smokers, and the possibility that chance has inflated this risk cannot be excluded.

Hiravama 1990 Page 3 Ascertainment of Temporal Relationship: Not specified. I ''i I I I I I I , G°t . -11 ! Cr ~ CA .F I

ABSTRACT FORM EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE Reference: Hole, D.J., Gillis, C.R., Chopra, C., and Hawthorne, V.M. 1989. Passive smoking and cardiorespiratory health in general population in the west of Scotland. BMJ 299:423-427. [Scotland] Type of Study: Prospective Source of Study Population: Study population was obtained from all men and women aged 45-64 years resident in two towns in Scotland between 1972 and 1976. Definition of Exposure: Four exposure groups were defined based on an index case aged 45-64 at the time of the survey (3,960 men and 4,037 women) and on the cohabitees (a respondent sharing the same household environment and examined at the same time in the survey as the index case) ever or never having smoked: 1) Control: the index case had never smoked and lived at the same address as another subject who had never smoked. No one else in the household who attended for screening was a smoker or ex-smoker. I 2) Passive smoking: the index case had never smoked and lived at the same address as a subject who had. 3) Single smoking: the index case was a smoker or ex-smoker and lived at the same address as a subject who had never smoked. No one else in the household who attended for screening was a smoker or ex-smoker. 4) Double smoking: the index case was a smoker or ex-smoker who lived at the same address as a subject who was also a smoker or ex-smoker. If the index cases were ex-smokers, they were classified as single smokers or double smokers depending on whether the cohabi:ees had never smoked " or ever smoked. If the cohabitees were ex-smokers, the index cases were classified as passive smokers if they had never smoked or as double smokers if they had ever smoked. Thus, the controls represented a group whose passive exposure was as low as possible within the constraints of the study design. Length of Follow-up: 11.5 years on the average.

aBSTRACT FORM EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE I I I i I Reference: Helsing, K.J., Sandier, D.P., Comstock, G.W., and Chee, E. 1988. Heart disease mortality in nonsmokers living with smokers. Am. J. Epidemiol. 127:915-922. [USAJ Type of Study: Prospective Source of Study Population: The cohort was derived from households in Washington County, Maryland; tobacco smoking habits were recorded for each household member over the age of 25 in 1963. 3,454 nonsmoking men and 12,345 nonsmoking women were included in the analysis. Definition of Exposure: All adults assigned smoking contribution scores ranging from 0 to 12 based on their reported smoking histories-never smoked, present or ex- smoker of cigarettes, cigars, or pipe, and amount smoked. A household exposure score was calculated as the sum of the contributions of all persons living in that household, and each person's passive smoke exposure score is the household score minus his or her own contribution to it. Length of follow-up: 12 years. Participation Rates: Not specified. Percentage of Proxy Respondents: Some family members asked about prior smoking habits, but numbers not specified. Definitions of Outcome(s): Causes of death were classified using the ICD 7th Revision; only deaths with underlying causes of death classified as arteriosclerotic heart disease including coronary disease (ICD 420) and other myocardial degeneration (ICD 422) were used. Investigators also analyzed deaths for which arteriosclerotic heart disease was listed on the death certificate but not coded as the underlying cause of death to confirm that similar associations were observed.

He et al. 1989 Page 2 Procedures used to Validate Exposures: A single female was considered to be equivalent to a female without a smoking spouse. Procedures to verify the accuracy of the data collected by the structured interview included tape recording and randomized re-interview. No data are presented, however, as to the results obtained by these procedures. ,l. No information on blinding with regard to case/control status is provided. Blinding: t Confounders: Other Potential Limitations (Re )gnized by authors): Statistical Metaods: Study Results: History of hypertension, family history of hypertension, family history of CHD, history ofpassive smoke exposure, history of drinking, exercise performance test, and history of hypercholesterolemia were included in logistic regression and were found to contribute to CHD risk. Passive smoke was found to contribute a statistically significant risk of CHD (OR=1.5). Confidence limits provided only in He et al. (1994) are 1.28, 1.77. It is difficult to ascertain &om this paper, however, if passive smoking ORs are adjusted for other risk factors. Authors suggest some of the analyses are based on small samples. Multiple regression analysis was performed. Crude odds ratio of passive smoking status on "disease" was 3.0 (95%CI: 1.3,7.2). See "adjusted" odds ratio, above. I I Dose-Response: Crude odds ratios of passive smoking status on "disease" by clinical diagnosis were: Angina pectoris: 4.7 (p<0.05) Myorcardial infarction: 2.6 (p>0.05) A dose-response relationship was reported between the number of passive smoke exposure years and the increase in OR for CHD. I

I I I I I I Hole et al. 1989 Page 3 There is no direct measure available to prove that the passive smokers received a higher environmental dose of tobacco smoke than the controls. Statistical Methods: Estimates of relative risk and 95 percent confidence intervals for passive smokers compared with controls were adjusted for the confounders noted above using the logistic regression model for cardiorespiratory symptoms and and Cox's proportional hazards model for mortality. Levels of significance were derived from the partial likelihood function. The biomedical data piocessing programs (BMDP) package was used to compute estimates of risk and levels of probability. Study Results: Relative risks associated with passive smoking. RRs were adjusted for age, sex, social class, and for cardiovascular variables (diastolic blood pressure, serum cholesterol concentration, and body mass index). Passive smokers are compared to controls. RR (95%CI) Cardiovascular symptoms: Angina 1:11 (0.73, 1.70) Major abnormalities on EKG 1.27 (0.48, 3.35) Ischemic heart disease mortality 2.01 (1.21, 3.35) Dose-Response: No estimate of dose-response is provided for passive smokers; however, investigators report tk.at a dose-response relation was seen based on data from active or ex-smoking groups. Ascertainment of Temporal . Relationship: Unclear whether those with pre-existing heart conditions were excluded at time of exposure assessment.

.1 1 I I I I I I I I I Sandler et al. 1989 Page 2 Confounders: Potential Limitations (Recognized by authors): Statistical Methods: Study Results: Dose-Response: Ascertainment of Temporal Relationship: Age, marital status, housing quality, and education. Factors such as alcohol consumption or dietary habits which are correlated with both smoking and risk for some.diseases may be alternative explanations for the findings. Because of low relative risks, and a weak dose-response reported for total mortality, other biases may be responsible for findings. Misclassification of smoking status also possible. Relative risks were calculated as the average annual death rates among smokers or passive smokers divided by the death rates among nonsmokers without household smoke exposure. All relative risks were adjusted for differences in age, marital status, housing quality, and education using Poisson regression. - Relative risks and 95% confidence intervals of deaths from circulatory disease in nansmoking men and women exposed to E fS Adj. RR 95%CI Men Circulatory disease 1.19 (0.97, 1.46) Arter. heart disease 1.31 (1.05, 1.64) Other 0.65 (0.36, 1.16) Women Circulatory disease 1.17 (1.05, 1.32) Arter. heart disease 1.19 (1.04, 1.36) Other 1.14 (0.89, 1.29) Not addressed for heart disease specifically. No information on health status of population at baseline. I

I I I 1 Jackson 1989 Page 2 Percentage of Proxy Respondents: 100% proxy respondents for coronary death cases; to reduce the extent of recall bias in the coronary death case-control comparisons, information was collected from a close friend or relative of both the coronar_v death cases and their living controls. Definitions of Exposure: Exposure to passive smoking in the home was defined by the presence and number of smoking members of the household as well as by the number of cigarettes smoked by the person in the household who smokes the most in the subject's presence per day. Workplace exposure to tobacco smoke was defined as the presence of smoking workers in the same room where the subject works. Procedures used to Validate Exposures: Not performed or reported. Blinding: After numerous pilot trials, it was not considered appropriate to blind the interviewers to the subjects' status. - Confounders: Alcohol, exercise, respiratory infection, and dietary variables were examined for active smoking. With the exception of dietary variables, one-tailed tests were used. These variables had little effect on relative risk estimates for active smoking and were presumed by authors not to impact passive smoking as well. Crude RRs were adjusted for age and social class. Other Potential Limitations (Recognized by authors): Number of people included in the analyses were relatively small resulting in imprecise relative risk estimates. Possibility that the increase in CHD risk related to passive exposure due to confounding. Statistical Methods: 2x2 tables were used to calculate crude odds ratios. Test-based 95% confidence intervals using the method described by Miettinen were also calculated. Where appropriate, stratified analyses were undertaken to control for confounding. Initially crude odds ratios were calculated for I

'1 `1 I , 1 I I I i I I Jackson 1989 Page 3 each stratum followed.by Mantel-Haenszel summary odds ratios and test- based 95"/u confidence intervals. Adjustment for age and social class by unconditional multiple logistic regression. Test for linear trends was an extension of the Mantel-Haenszel test for homogeneity. Study Results: Crude relative risk and 95% confidence interval of CHD associated with passive smoking by exclusive exposure categories [reference category is the unexposed group] Men Women Exposure MI CD MI CD Home only 0.6 2.4 23.5 (0.96, 5.9) (0.4, 14.1) (2.8-199) Work only 1.5 1.2 0.9 5.2 (0.6, 3.8) (0.4, 3.6) (0.09, 8) (0.9, 30.1) Home+Work 2.7 1.6 4.4 (0.6, 12.1) (0.3, 9.6) (0.4, 49.5) Relative risk and 95% confidence interval of CHD associated with " passive smoking by combined exposure categories [reference category is the unexposed group] Men Women Exposure MI CD MI CD Home±Work 1.1* 1.0 2.8 7.8 (0.3, 4.3) (0.2, 4.5) (0.6, 13.6) (1.3, 48) 1.03** 1.1 2.7 5.8 Work±Home 1.7 1.3 1.4 3.6 (0.7, 4) (0.5, 3.6) (0.3, 8.2) (0.7, 20.1) 1.8 1.8 1.1 2.2 * Crude relative risks ** Relative risks estimated using logistic regression and adjusted for age and social class. I Dose-Response: A dose-response relationship for home exposure was reported for women. Investigators also note that cases were exposed to approximately 50% more cigarettes at home than controls and that the effect of passive smoking exposure at home was greatest in women who were in general

ABSTRACT FORM EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE Reference: Hirayama, T. 1990. Passive smoking. NZ Med. J. 103:54. [Japan] Hirayama, T. 1984. Cancer mortality in nonsmoking women with smoking husbands based on a large-scale cohort study in Japan. Prev. Med. 13:680-690. Hirayama, T. 1984. Lung cancer in Japan: Effects of nutrition and passive smoking. Hirayama, T. 1981. Non-smoking wives of heavy smokers have a higher risk of lung cancer: A study from Japan. BMJ 282:183-185. Type of Study: Prospective Source of Study Population: Total population represented 91,540 married women, ages 40 years and above residing in 29 Health Center Districts in Japan. Original interviews were conducted in October-December 1965, and the population was tracked by establishing a record linkage system between the risk factor records and death certificates. Definition of Exposure: The exposed population was comprised of 69,645 nonsmoking women who were classified according to their husband's smoking habit at entry into the study -- exsmokers or smokers of 1-19 cigarettes daily and smokers of more than 20 cigarettes daily. The unexposed pop»?,tion was comprised of 21, 895 nonsmoking women whose husbands were also nonsmokers. Length of Follow-up: 16 years. Participation Rates: •• Not specified. Percentage of Proxy Respondents: N/A Definitions of Outcome(s): Not specified.

Hiravama 1990 Page 2 Procedures used to Validate Outcome(s): Blinding: Confounders: Other Potential Limitations (Recognized by authors): Statistical Methods: Study Results: Dose-Response: Not specified. Not specified. . Age, occupation, wife's consumption of green-yellow vegetables. Consumption of green-yellow vegetables had a risk inhibitory effect. Not specified. Mantel-Haenszel Chi, one-tail p-value. Overall results were not significant at 14 years follow up but were significant at 16 years follow up. Relative risk and 90% confidence limit for ischemic heart disease mortality in non-smoking wives living with husbands who were ex- smokers or who smoked 1-19 cigarettes per day (1966-1981) compared to non-smokers = 1.08 (0.9, 1.3). Relative risk and 90% confidence limit for ischemic heart disease mortality in non-smoking wives living with husbands who smoked more than 20 cigarettes per day (1966-1981) compared to non-smokers = 1.30 (1.06, 1.60). lnvestigators report that risk of ischemic heart disease increased with the number of cigarettes smoked by the husband. Number of Cigarettes Smoked by Husband OR (90% CI) Nonsmoker 1.00 Exsmoker, 1-19/daily 1.08 (0.90, 1.30) 20+/daily 1.30 (1.06, 1.60) .1

ABSTRACT FOR,M EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE Reference: Sandler, D.P., Comstock, G.W., Helsing, KJ., and Shore, D.L. 1989. Deaths from all causes in non-smokers who lived with smokers. AJPH 79:163-167. [USA] Type of Study: Cohort Source of Study Population: The cohort was derived from households in Washington County, Maryland; tobacco smoicing habits were recorded for each household member over the age of 25 in 1963. 4,162 men and 14,873 women were included in the analysis. Definition of Exposure: A household smoke exposure score was calculated to serve as a measure of passive smoking. To create this score, each adult was assigned a smoking contribution score ranging from 0 to 12 based on their personal smoking history. A household smoking total was calculated as the sum of the smoking contribution scores of all persons living in that household, - and each individual's household_smoke exposure score was the household total minus his or her own contribution to it. Length of follow- up: 12 years. Participation Rates: Not specified. Percentage of Proxy Respondents: Not specified. Deflnitions of Outcome(s): Procedures used to Validate Outcome(s): Causes of death coded according to the 7th revision of the ICD. Death certificates of Washington County residents who died between July 1963 and July 1975 were matched against the census. ~ ~ Blinding: Not specified. (,!I . ~ Lb ~ . . CJ W

Dobson et al. 1991 Page 4 Dose-Response: No data by categories of exposure; only subgroup studied was ex-smokers. Ascertainment of Temporal Relationship: Authors did not ascertain that exposure preceded disease. No indication of latency. ';I I

I I I I I I I I ABSTRACT FORM EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE Reference: Jackson, R. 1989. The Aukland Heart Study: A Case Control Study of Coronary Heart Disease. Doctoral dissertation, University of Aukland, Aukland, New Zealand. (New Zealand] Type of Study: Case-control Source of Study Population: Cases were identified from a CHD register, the ARCOS study, which included all people aged 25-64 years normally resident in the Central Aukland Statistical area. Subjects were registered on one of the 22 general electoral rolls in the study area during the period of the project. For the passive smoking portion of the study, cases included 28 male and 11 female subjects with nonfatal myocardial infarction and 21 male and 9 female coronary death cases identified from the ARCOS CHD register between March 1987 and February 1988 (excluding current smokers and ex-smokers). The ARCOS Study identifies hospitalized definite myocardial infarction cases and both hospitalized and non-hospitalized _ coronary death cases; for this study, cases were defined as nonfatal if they were alive at the time they were interviewed. The criteria for nonfatal myocardial infaiction are definite ECG criteria classified according to the Minnesota coding method; or typical or atypical symptoms together with a probable ECG and abnormal enzymes; or typical symptoms and abnormal enzymes with nonevolving ECG or noncodable ECG or no ECG. The criteria for coronary death were fatal cases who do not meet the criteria for nonfatal myocardial infarction and where there is no good evidence for another cause of death. Controls were identified from the same 22 electoral roles in the Central Aukland Statistical Area. For the passive smoking portion of the study, myocardial infarction controls included 123 males and 112 females and coronary death controls included 61 males and 62 females (excluding current smokers and ex-smokets), Participation Rates: Response rates not provided for the passive smoking portion of the study. Data on passive smoking was only collected in the second year of the Aukland Heart Study when reports on CHD and passive smoking in the medical literature stimulated the investigators' interest in this issue. I

i I Humble et al. 1990 Page 2 Procedures used to Validate Outcome(s): Blinding: Confounders: Other Potential Limitations (Recognized by authors): Statistical Methods: Not specified. N/A Analyses of white women stratified,by social status because of its inverse relationship with smoking status and CVD mortality in this cohort. Age, systolic blood pressure, serum cholesterol, and body mass index. Data are lacking to examine whether exposure status changed during follow-up due to remarriage. Power to test for small differences in effect of passive smoking by race or social standing was lacking as were data to evaluate alcohol use of physical activity. Mean baseline characteristics by passive smoke exposure were compared using t-tests. Cox proportional hazards models were used to estimate the association of passive smoking with time to all CVD, smoking-related CVD, and all-cause mortality in this population while adjusting for the confounders noted above. Relative risks and 95 percent confidence intervals were calculated using the SAS proportional hazards modeling procedures, and the statistical significance of the trends was tested using a method proposed by Rothman. Constancy of the relative risks over time was verified before,the proportional hazards were modeled.

) He et al. 1994 Page 5 Ascertainment of Temporal Relationship: Authors did not ascertain that exposure preceded disease. No indication of latency.

La Vecchia et al. 1993 Page 3 Ascertainment of Temporal Relationship: Not possible to determine whether exposure preceded disease.

.1 I ''I I I I I I I I I ABSTRACT FORM EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE Reference: Humble, C., Croft, J., Gerber, A., Casper, M., Hames, C.G., and Tyroler, H.A. 1990. Passive smoking and 20-year cardiovascular disease mortality among nonsmoking wives, Evans County, Georgia. Am. J. Public Health 80:599-601. [USA] Type of Study: Prospective Source of Study Population: Cohort composed of 513 rural, married Black and White women from the Evans County, Georgia study, who were disease-free and self-described as never-smokers at baseline in 1960, married to male examinees who reported they had either never smoked or were current smokers at baseline. Women married to ex-smokers were excluded from the analysis. Definition of Exposure: Exposure to passive smoking was defined by husband's smoking status (current, never) at the time of the baseline interview. Stability of smoking status was assessed by a second survey of study subjects in 1967. This comparison showed that 98% of wives again reported themselves as never having smoked in 1967. Similarly, 98% of never smoking husbar.ds maintained their reported status in 1967 while 25% of husbands who smoked in 1960 described themselves as non-smokers in 1967. Length of follow- up: 20 years (vital status determined as of May 1, 1980). Participation Rates: Unclear. Authors mention, however, that three subjects who did not have follow-up information were excluded. Percentage of Proxy Respondents: N/A Definitions of " Outcome(s): All CVD mortality according to ICD 8th Revision.

Humble et al. 1990 Page 3 Study Results: Relative Risks of Smoking-Related CVD mortality (ICD8 codes 410- 456) by two types of adjustment Whites All . Blacks HSS' LSS' I 1. Age-adj. RR 1.29 1.57 1.67 0.61 (95% CI) (0.79, 2.10) (0.73, 3.37) (0.64, 4.36) (0.25,1.47) Adj. Hazards Rates 1.54 1.68 1.97 0.82 (95% CI) (0.93, 2.55) (0.76, 3.71) (0.72, 5.34) (0.31, 2.15) Dose-Response: A trend in risk ove.r level of husband's smoking as reported in 1960 was only seen among high social status Whites. Relative risks for both total and smoking related CVD mortality among wives whose husband's smoked <10,10-20, and >20 cigarettes per day as compared to wives of non-smokers were 1.02, 2.11, and 2.55 respectively (p for trend <0.06). A marginally significant (p<0.09) trend in risk for all CVD and smoking- related CVD over crude levels of duration of exposure was also apparent only among high social status White women. [No tables are presented for dose-response data]. As. _rtainment of Temporal Relationship: Vital status ascertained 20 years after baseline exposure. Ascertainment that subjects were disease free at baseline; because the endpoint of this study was mortality reasonable certainty exists that exposure preceded the endpoint. ' High social status 2 Low social status

Dobson et al. 1991 Page 2 Response data: Smoking prevalence in control group: Males I Main resp.4 24% current. 40% ex-, 35% non- . Brief questionnaire : 28% cuaent, 424io ex-. 30% non- 'I Interviewed at home}: 29% current, 42% ex-, 29% non- Females Main resp.t: 14% current, 19% ex-, 67% non- Brief questionnaire2: 21 °/u current, 13% ex-, 66% non- Interviewed at home': 31% current, 16% ex-, 53% non- I Percentage of Proxy Respondents: Unclear, although authors state next-of-kin/medical records review were utilized. However, information on smoking behavior was not obtainable for 34% of fatal cases and 4% of non-fatal cases; data on passive smoking were missing for about 15% of all cases. Definitions of Exposure: The way in which ETS exposure in cases and controls was assessed is not clear. The authors mention that smoking history was obtained by - interview in hospitalized cases and by interview and questionnaire in controls, but no definition of ETS home exposure is given. No definitions are provided for what constitutes a non-smoker, but data are presented separately for non- and ex-smokers suggesting that the category "non- smoker" may represent never smokers. Current smoker cases and controls were not queried about ETS exposure. Procedures used to Validate Exposures: None. Blinding: No blinding apparent with regard to case/control status. Confounders: Age, sex, and a prior history of heart disease. ' This was the group for which information on passive smoking was obtained 2 This group was part of the follow up of non-respondents 3 This group constituted further follow up of non-respondents

ABSTRACT FORM EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE I I ~ I I I ~ Reference: I Type of Study: Source of Study Population: Participation Rates: Dobson, A.J., Alexander, H.M., Heller, R.F., and Lloyd, D.M. 1991. Passive smoking and the risk of heart attack or coronary death. Med. J. Aust. 154:793-797. [Australia] Case-control Cases were residents of the Hunter Region of New South Wales aged 35- 69 years during the stuay period July 1988 through October 1989. This population was part of the MONICA, WHO Project, a study designed to monitor trends and determinants of cardiovascular disease in well-defined populations over 10 years. Cases were defined as those who during the study period had a fatal or non-fatal definite or possible MI or coronary death (with insufficient information for more specific classification). There were 183 male nonsmoker cases and 160 female nonsmoker cases. Controls were randomly selected participants in the community-based _ risk factor prevalence study conducted as part of the WHO MONICA Project. The risk factor study was conducted in June-December 1988 and June-November 1989.. Bias associatedh with using electoral role as source of controls is considered by the authors because of potential under enrollment. There were 293 male nonsmoker controls and ~32 female nonsmoker controls. Cases: Uncertain, but authors state that they had "z:.nost complete ascertainment of all cases of heart attack and coronary death in the study population" (p.795). Controls: Initial response was 63%. Some people who were unable to attend a study center for physical measurements and blood samples did complete a brief questionnaire covering demographic characteristics, smoking behavior, and medical hisiory. Others were interviewed at home to obtain this information. Final participation was 80%. However, passive smoking information was obtained only from those C who participated fully (63%) and those who were not ~ current smokers. ~ m . ~ ~

':I ABSTRACT FORM EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE I Reference: La Vecchia, C., D'Avanzo, B., Franzosi, M.G., and Tognoni, G. 1993. I Passive smoking and the risk of acute myocardial infarction. Lancet 341:505-506. [Italy] Type of Study: Case-control I Source of Study Population: Study conducted in 1988-1989 within the framework of the GISSI-2 study (a randomized clinical trial of alteplase v. streptokinase and heparin v. no I heparin in 12,490 case of AMI). Cases were 113 currently married never smoker subjects (44 women and 69 men) with first episodes of AMI. Controls were 225 currently married never smoker subjects (60 women and 125 men) in hospitals for acute diseases not related to any known or potential cardiovascular risk factors. I Participation Rates: No data on refusals or participation rates are provided. I Percentage of Proxy Respondents: No data are provided. Definitions of Exposure: Exposure is investigated through questions on the spouse's habits, including smoking status (never-smoker, current, ex-smoker), number of cigarettes smoked per day, number of years the couple had lived together, and, if ex-smoker, time since quitting. Procedures used to Validate Exposures: None reported. Blinding: No information on blinding with regard to case/control status is provided. Confounders: ` Sex, age, education, coffee consumption, body mass index, serum cholesterol, hypertension, diabetes, family history of AMI. CA

I I I He et al. 1994 Page 3 Validation with cotinine was not possible; thus the doses of exposure in this study cannot be used for risk assessment for unit doses of exposure. Small probability of bias due to exposure misclassification. Statistical Methods: Data were managed and analyzed using computer packages of Epi-Info and SPSS-PC. Statistical procedures used included x, t-test, Chi square, Chi square test for trend, and standard multivariate techniques for unmatched case-control studies (stratified analysis and multiple logistic regression analysis). Study Results: Passive smoking from husband and at work and crude ORs of CHD From husband From work By combination of exposure No husband/no work Yes husbandIno work No husband/yes work Yes husband/yes work Any exposure (husband, work, both) OR (95%CI) 2.12 (1.06, 4.25) 2.45 (1.23, 4.88) 1.0 2.07 (0.78, 5.55) 2.53 (0.82, 7.83) 4.18 (1.63, 10.92) 2.87 (1.28, 6.55) Adjusted OR for passive smoking from husband was 1.24 (CI, 0.56, 2.72), and at work was 185 (CI, 0.86, 4.00). When passive smoking from husband was semoved from the logistic models, the ORs for passive smoking at work were slightly higher. Dose-Response: For passive smoking from husband, the crude OR showed significant linear trends with mnount smoked daily by husband's duration of exposure and by cumulative exposure (amount daily multiplied by duration), but the trend became non-significant after passive smoking at work and the other five risk factors were adjusted for in the final model (no data or tables are presented).

He et al. 1994 Page 4 Passive smoking at work and ORs of CHD by levels of exposure Crude OR Adj. OR P P(lin) (95%oCI) (95%oCI)' Cigs/d by co-workers 0-5 (actually 0) 1.00 6-10 1.28 (0.50, 3.24) 0.87 (0.30, 2.53) 0.02 0.02 11-20 3.42 (1.35, 8.74) 2.95 (1.05, 8.28) z20 6.38 (1.57,27.84) 3.56 (0.81, 15.58) stat. sig. trend 1 i I I Duration of exp. (yrs.) 0-5 (actually 0) 1.00 6-15 2.13 (0.70, 6:40) 3.08 (0.90, 10.58) 0.2 0.1 z16 2.57 (1.23, 5.42) 1.56 (0.67, 3.64) stat. sig. trend No. Smokers 0 1.00 1-2 1.42 (0.64, 3.15) 1.16 (0.48, 2.82) 0.02 0.02 3 6.38 (1.97, 21.51) 5.06 (1.42, 18.02) a4 15.96 (1.67,378.10) 4.11 (0.39, 43.68) stat. sig. trend Exposure time daily (hrs.) 0 1.00 1-2 0.85 (0.31, 2.25) 0.62 (0.22, 1.80) 0.001 0.002 3-4 4.35 (1.164, 11.73) 4.03 (1.33, 12.25) i5 15.96 (2.98, 113.32) 21.32 (2.71, 168.00) stat. sig. trend Cumulative exp. 0 1.00 1-2000 1.30 (0.55, 3.03) 1.00 (0,39, 2.57) 0.005 0.003 2001-4000 2.00 (0.51, 7.55) 2.05 (0.47, 8.87) z4001 15.96 (3.89, 75.27) 9.23 (2.01, 42.25) stat. sig. trend I 1 Adjusted for age, history of hypertension, personality type, high density lipoprotein cholesterol, and passive smoking from husband. . .

Dobson et al. 1991 Page 3 Other Potential Limitations (Recognized by authors): Information on. smoking unobtainable for many case subjects who died. Residual confounding (e.g., SES). Low statistical power (small sample size). Statistical Methods: Study Results: Differences in methods of data collection (interviews for cases and questionnaires for controls) may have contributed to misclassification of smoking status. . . The statistical program GLIM was used to calculate adjusted odds ratios and approximate confidence intervals by logistic regression. Terms for age (5-year age groups from 35-39 to 64-69 years) and history (previous MI or history of other ischaemic heart disease versus no history) and interaction between these two factors were included in the model as well_ as tenns for the smoking variables. Results for non-smokers and ex- smokers were calculated separately. Epi-Info was used to calculate exact confidence intervals and tests for trend for crude odds ratios. Passive smoking at home and risk of heart attack or coronary death compared to never-smokers (ex-smokers excluded) Ctude OR (95%CI) Adj. OR' (95%CI) Males 1.04 (0.56, 1.91) 0.97 (0.50, 1.86) Females 1.61 (1.04, 2.47) 2.46 (1.47, 4.13) Passive smoking at work and risk of heart attack or coronary death compared to never-smokers (ex-smokers excluded) Crude OR (95%CI) Adj OR (95%CI) ';I Males 0.90 (0.50, 1.60) 0.95 (0.51, 1.78) Females 0.71 (0.19, 2.27) 0.66 (0.17, 2.62) ~ ° Adjustment forage and heart disease.

La Vecchia et al. 1993 Page 2 Other Potential Limitations (Recognized by authors): Investigators report that the interpretation of the study remains inconclusive because of ;ts lack of statistical significance, limited exposure assessment, and potential misclassification of the smoking status of subjects interviewed and their spouses. Statistical Methods: RRs with 95% confidence intervals calculated using Mantel-Haenszel estimates adjusted for sex and decade of age and estimates from multiple regression equations, including terms for sex, age, years of education, coffee consumption, body mass index, serum cholesterol level, hypertension, diabetes, and family history of AMI. Study Results: Adjusted RRs with 95% confidence intervals for never smoking cases (maies and females) of AMI according to smoking habits of spouse, comparison with never smoker spouses I I I Spouse Smoking Habit Adj: RR 95%CI Never smoker 1.00 Ex-smoker 1.19* (0.33, 2.10) 0.91** (0.36, 2.28) Current smoker 1.31 * (0.65, 2.64) 1:21** (0.57, 2.52) * Adjusted by Mantel-Haenszel procedure ** Adjusted by multiple regression Dose-Response: No analysis for trends; some data presented suggestive of increased risk with increased exposure, but RRs not statistically significant. Spouse Smoking Habit Adj. RR 95%CI Never smoker 1.00 <15 cigarettes/day 1.24* (0.52, 2.96) 1.13** (0.45, 2.82) ..CII >_15 cigarettes/day 1.39* (0.54, 3.63) 130** (0 50 3 40) W . , . C.3 41 CJ 01

I I I I I I I I I I I I I I I t I I He et al. 1994 Page 2 Definitions of Exposure: Passive smoking from husband was defined as living with a smoking husband for over 5 years. Single women would have been considered as not exposed, but all women were found to be married. Procedures used to Validate Exposures: Passive smoking at work was defined as working with smoking coworkers in the same office or factory unit for over 5 years. Periods away from the work environment were considered as no exposure in the workplace. All subjects were found to be either not exposed or exposed for over 5 years. Quality controls for interviewing included tape-recording of the interviews for 10% of hospital subjects; interviewing the husband to validate data from the wife for 33% of hospital subjects; and single blind re-interview by a second interviewer, who was not aware of the case-control status of the subjects, for 30% of hospital subjects. Test-retest methodology showed agreement for passive smoking at work was 74.3% and for passive smoking from the husband was 91.4%. Subjects were not interviewed again on quantity of exposure, but it is expected that this agreement would be lower. A total of 26 patients who had initially been diagnosed as having CHD but were subsequently confirmed as normal were accepted as controls. Their exposure to passive smoking was compared with that of controls and cases who had not had coronary arteriography to check for subjective bias due to interviewers. Blinding: Single blind re-interview by a second interviewer, who was not aware of the case-control status of the subjects, for 30% of hospital subjects. Confounders: Five factors were included in logistic regression, although others (not specifiad) were considered. The potential confounders adjusted for were age, history of hypertension, type A personality, total cholesterol, and high density lipoprotein cholesterol. Other Potential Limitations (Recognized by authors): There may be prevalence-incidence bias from this being a study of incident cases of CHD. I

ABSTRACT FORM EPIDEMIOLOGIC STUDIES ETS AND CARDIOVASCULAR DISEASE Reference: He, Y., Lam, T.H., Li, L.S., Li, L.S., Du, RY., Jia, G.L., Huang, J.Y., and Zheng, J.S. 1994. _Passive smoking at work as a risk factor for coronary heart disease in Chinese women who have never smoked. BMJ 308:380384. [Chinal Type of Study: Case-control Source of Study Population: Cases were 59 women, ages 37-67, with CHD (non-fatal, incident cases) from three large teaching hospitals and two medical universities between December 1989 and November 1992. Final diagnosis was myocardial infarction according to WHO criteria or coronary stenosis confirmed by coronary arteriography (34 confirmed by arteriography and 25 with myocardial infarction). Controls were 126 women, 42-66 years, derived from three sources: patients admitted because of suspected or diagnosed coronary heart disease but confirmed to be normal after coronary arteriography (n=61); ' other medical outpatients attending cardiology departments (patients with nsychosomatic symptoms, menopausal syndrome, dysrhythm~s, or non- cardiac chest pain) (n=28); and a random sample of healthy subjects from a community screening program for coronary heart disease (n=37). The latter two groups were confirmed to be free of coronary heart disease by WHO cnteria and normal exercise electrocardiography. Cases and controls were full-time working ethnic Chinese women who had never smoked cigarettes (lifelong non-smokers). Full-time housewives, peasants, and those who were retired for 5 years or more were excluded. Participation . .. Rates: Almost 70% of all eligible cases of coronary heart disease treated in the three hospitals were included. Percentage of Proxy Respondents: No data are provided.