Tobacco Documents Online

DRAFT UNITED STATES ENVIRONMENTAL PROTECTION AGENCY ENVIRONMENTAL TOBACCO SMOKE: A COMPENDIUM OF TECHNICAL INFORMATION CHAPTER ELEVEN PASSIVE SMOKING AND HEART DISEASE: EPIDEMIOLOGY, PHYSIOLOGY, AND BIOCHEMISTRY COMMENTS OF THE TOBACCO INSTITUTE April 6, 1990 Volume I TIO8872SO0

TABLE OF CONTENTS I. Tobacco Institute Overview Comments Tab A: Tab B: Tab C: Tab d: Comments Prepared by Dr. Lawrence Wexler Comments Prepared by Dr. W. Gary Flamm Comments Prepared by Dr. James A. Will Comments Prepared by Dr. Joseph M. Wu II. Copies of Additional Literature Not Cited in Chapter Eleven (Alphabetically by First Author) T!08872901

I. INTRODUCTION These comments on "Passive Smoking and Heart Disease: Epidemiology, Physiology and Biochemistry," by Stanton A. Glantz, Ph.D., and William W. Parmley, M.D., which is Chapter Eleven of the draft document entitled Environmental Tobacco Smoke: A Compendium of Technical Information, are being submitted by the Tobacco Institute in response to the invitation contained in Mr. Robert Axelrad's letter of March 13, 1990. We understand that this chapter, like the first ten chapters of the draft Compendium, is to be given a "peer review" before it is made public. We appreciate the opportunity to participate in this review process. As with the comments that we filed on February 5, 1990, with respect to Chapters i-i0 of the draft Compendium, our comments on Chapter ii are in three parts: (i) this brief overview of our responses to the chapter; (2) reviews by scientists with expertise in the relevant disciplines; and (3) copies of scientific literature that our reviewers determined was important to the issues under discussion but had not been referred to by Drs. Glantz and Parmley. Chapter Eleven of the draft Compendium has been reviewed thus far by Professor Lawrence Wexler, an epidemiologist at New York Medical College; by Dr. Gary Flamm, a consultant in toxicology and hazard evaluation who until 1988 was Director of the Office of Toxicological Sciences at the Food and Drug Administration; Professor James Will, a T108872902

specialist in comparative medicine at the University of Wisconsin at Madison; and Dr. Joseph Wu, a biochemist at New York Medical College. Because of the complexity of the material contained in Chapter Eleven and other time commitments, several additional scientists were not in a position to complete their reviews in the very short period that was provided. These additional reviews will, however, be submitted in due course. As with the first ten chapters of the draft Compendium, our reviewers have determined that Chapter Eleven manifests numerous deficiencies and contains discussion of topics unrelated to the basic purpose of the Compendium. In its present form, it would very likely confuse and mislead its intended audience. Accordingly, the chapter should be substantially revised before it is issued. Furthermore, we reiterate our position that it is critical, in the interest of fairness and objectivity, to submit the revised chapter to the Science Advisory Board for independent peer review. We will now briefly recapitulate the key points in our reviewers' statements and offer some additional observations. II. THE INTRODUCTORY PORTION OF THE CHAPTER Curiously, Drs. Glantz and Parmley begin their chapter on heart disease with a discussion of the issue of - 2 - T108872903

whether exposure to environmental tobacco smoke (ETS) is related to lung cancer. Their rhetorical purpose in doing so is quite clear: after asserting that lung cancer has been shown to be "definitively . . . caused" by ETS exposure, they go on to observe that it is therefore "not surprising" that heart disease has now been linked to ETS exposure as well. Leaving aside the illogic of this attempt to forge some sort of connection in the reader's mind between two entirely disparate types of disease, what i_~s surprising is the authors' complete lack of foundation for their assertions. First of all, the authors' claim that there is a definitive causal relationship between ETS exposure and lung cancer overstates the conclusions of the 1986 reports of the Surgeon General and the National Academy of Sciences. Moreover, as our reviewers demonstrated in our February 5, 1990, comments on Chapter Four of the draft Compendium, there is as yet no scientific basis for concluding that there is even a valid statistical association between ETS exposure and lung cancer (ar any other forms of cancer). We refer Dr. Glantz and Parmley to that discussion and will not reiterate it here. More importantly for present purposes, the authors neglect to point out that, with respect to cardiovascular disease, both the Surgeon General and the NAS determined in 1986 that it was not possible to conclude that ETS exposure was a risk factor. The Surgeon General's report concluded (at - 3 - T108872904

p. xi) that "further research . . . will be required to determine whether an association exists between ETS exposure and an increased risk of developing" heart disease. And the NAS summarized its discussion of this issue as follows (at pages 265-66): "WHAT IS KNOWN "I. No statistically significant effects of ETS exposure on heart rate or blood pressure were found in healthy men, women, and school-aged children during resting conditions. During exercise there is no difference in the cardiovascular changes for men and women between conditions of exposure to ETS and control conditions. "2. With respect to chronic cardiovascular morbidity and mortality, although biologically plausible, there is no evidence of statistically significant effects due to ETS exposure, apart from the study by Hirayama in Japan. "WHAT SCIENTIFIC INFORMATION IS MISSING "I. Experimental studies with animal models needs to be performed with ETS to determine whether the cardiovascular changes seen following exposure to whole smoke also occur following exposure to ETS. "2. Existing studies have not provided evidence of serious harm in people with heart disease. With regard to angina onset, the findings are uncertain and need to be repeated." In addition, it is worth noting that a recent article co-authored by Dr. Jonathan Samet, the author of Chapter Four of the draft Compendium, refrained from concluding that ETS exposure is a risk factor for cardiovascular disease. This article's review of the relevant - 4 - T!08872905

epidemiologic studies concluded that "[t]hese studies are preliminary, and the exposure-response relationship is not clear." Wu-Williams, A.H. and Samet, J.M. (1990). "Environmental Tobacco Smoke: Exposure-Response Relationships in Epidemiologic Studies." Risk Analysis 1011]: 39-48. It must be assumed that Drs. Glantz and Parmley are aware of the conclusions just presented. Nevertheless, they attempt to make a case for a relationship between ETS exposure and heart disease. They do so first by revisiting the epidemiologic studies in this area, most of which were considered by the Surgeon General and the NAS, in an effort to show that they are more reliable than others have found. As we show below, their effort is unavailing; the epidemiologic studies conducted thus far by no means justify a conclusion that ETS exposure increases the risk of heart disease. Perhaps aware of their weakness on this point, the authors next provide a discussion of various human and animal studies in the areas of physiology and biochemistry in an effort to show that harmful effects of ETS exposure are biologically plausible. Here, too, the authors do not achieve their goal; an objective review of the studies considered by Drs. Glantz and Parmley shows that they have seriously mischaracterized and overstated the evidence on the issue of biological plausibility. - 5 - Ti08872906

III. EFFECTS OF ACTIVE SMOKING Before turning to their discussion of the epidemiologic literature pertaining to ETS and cardiovascular disease, Drs. Glantz and Parmley state, without explaining why, that "it is worth summarizing the evidence linking active smoking with coronary heart disease" and then proceed with an attempt to do so. Suffice it to say, as we have pointed out with respect to similar efforts in our February 5, 1990 comments on Chapters 1-10 of the draft Compendium, that such a discussion in fact has no place in a treatment of the issues involving ETS. Given the numerous significant differences in the composition and behavior of ETS as opposed to mainstream tobacco smoke, such a discussion can only serve to mislead readers who are unfamiliar with the relevant scientific literature. Accordingly, this portion of the chapter should simply be deleted. IV. EPIDEMIOLOGIC STUDIES In reviewing the extant epidemiologic studies on ETS exposure and heart disease,l/ the authors acknowledge that l/ The authors refer to the existence of eleven published (footnote cont'd) - 6 - TI08872907

potential problems caused by misclassification bias and confounding variables must be considered, but they go on to suggest that misclassification has not artificially elevated the risk factors in certain of these studies, and that the studies have adequately controlled for confounders. They also imply, by focusing on the alleged statistical power of two of the studies, that a credible statistical association has been detected. Further, they claim that meta-analysis is a valid technique for pooling the studies in order to increase their statistical power. Finally, they suggest that a clear dose-response relationship has been observed in the studies, thus assertedly strengthening the hypothesis that they have detected a real rather that artefactual effect. The reviews by Drs. Wexler and Flamm show that the authors' presentation of the ETS epidemiology is a house of cards. As Dr. Wexler demonstrates in his thorough review of the available studies, each of the published studies fails to control for one or more important confounding variables, such as lifestyle, blood pressure, serum cholesterol, obesity and socioeconomic status. In addition, none of the studies (footnote cont'd) studies, but they mention only ten; in their bibliography, one (the Humble et al. study) is listed as in press and accordingly is unavailable, and another (the Martin et al. study) is listed as having been presented at a meeting in 1986 but it is not available in a publication. Another study, that by He, appeared only in Chinese with a 5-sentence English abstract. -7 - T!08872908

provides an adequate measurement of ETS exposure or demonstrates specificity of association between such exposure and cardiovascular disease. There are serious methodological problems in all of the studies, including small sample size and possible misclassification of spousal smoking status. There is no demonstration of consistency among the existing studies, and pooling them in a meta-analysis is illegitimate because it merely reinforces the artefactual associations of the individual studies. Thus, it is impossible to conclude that the five prerequisites for good epidemiologic studies -- strength of statistical association, consistency of the association, demonstration of exposure to ETS at a reasonable time before the onset of disease, specificity of the ~tatistical association to ETS, and biological plausibility -- have been met in the work reviewed by Drs. Glantz and Parmley. A~cordinglg, ~here is no reason to depart from the conclusions reached by the Surgeon General and the National Academy of Sciences in 1986: exposure to ETS has not been shown to be a risk factor for cardiovascular disease. Dr. Flamm's review of the difficulties with the epidemiologic studies echoes the points made by Dr. Wexler. In addition, he considers the issue of biological plausibility from the point of view of a toxicologist. He points out (at pages 6-7) the failure of Chapter Eleven to consider biological plausibility in dosimetric terms, given the - 8 - TI08872909