Tobacco Documents Online

• " ~ FTR S&T ~00~ ~1/i0 '97 14:47 ~+49 2203 30~362 INBIFO Koeln INBIFO Institutf~r bioiogische Forschung GmbH-K61n 31 October 1997 -/KHO Page Comments on the Paper: "Environmental tobacco smoke exposure and i$chaemic heart disease: an evaluation of the evidence" by M. R. Law, J. K. Morris and N, J. Waid. Draft(to be continued) The paper has the objective " To estimate the dsk of ischaemic heart disease caused by exposure to environmental tobacco smoke and to explain why the associated excess risk is almos~ half that of smoldng 20 cigarettes per day when the exposure is only about 1% that of smoking." The authors followed thls objective by conducting a mete-analysis of 19 "acceptable" published case-control studies on passive smoking and ischaemic heart disease, by the joint evaluation of 5 cohort studies on smoking and ischaemic heart disease, and by the discussion of results from studies of platelet aggregation and studies of diet according to exposure to tobacco smoke. They conclude that breathing ETS increases a person's risk of lschaemic heart disease by 25%. Methods. In the method description part, some information is given on the process of selecting s~udies of exposure to environmental tobacco smoke and ischaemic heart disease that are "relevant'. Relevance was evaluated by scanntng the reference list of each study and review article. In addition, relevance was decided after discussion with colleagues. No objective cdteria are given that make this selection process reproducible. It is stated that data on spouses of ex- smokers were excluded "where possible". Apparently no attempts were made to take into account possible misclassification bias concerning current or former smokers among subjects. The authors themselves caused selection bias by excluding two of the studies (possibly the largest and most recent). For the determination of the risk of ischaemic heart disease associated with smoking, data were used from 5 cohort studies of men recruited in the 1950s, Only the size of the studies is given as a selection argument. Apparently the number of cigarettes smoked per day in these studies from the 1950s was adjusted using data on biochemical markers of tobacco smoke intake from the 1990s, No information is given on the influence of the evolution of cigarette types and smoking behavior over the 40 years. To estimate the extent of confounding, the authors use reversibility of the excess risk of ischaemic heart disease as it is observed in the 5 cohort studies. By using this approach they ignore the fact that "a confounding factor must be associated with both the exposure under study and the disease under study to be confounding" (Rothman). This definition implies that the health effects of a confounding factor of smoking and the dsk of ischaemic heart disease will necessarily be reversed when smoking is reversed. Therefore the extent of reversibility cannot be used as an indirect estimate of the extent of confounding. To estimate the relationship between smoking and other potential risk factors for ischaemic heart disease, the authors examined published data of studies that measured the dietary intake of vadous nutrients in smokers and non-smokers and in non-smokers who lived with smokers. These investigations can provide valuable information on the excess risk of ischaemic heart disease attributable to differences in the consumption of certain nutrients as measured in these studies. However, the validity of these results may be restricted to specific societal groups and situations. Therefore data on dietary habits in the 1990s from one study cannot be used for risk estimation in another study in the 1950s without caution. Apparently that is the approach the authors followed. In addition a potential difference in the amount of stress, which is known to be a very strong risk factoF for ischaemic heart disease, is not considered in this discussion. XSLAW.DOC

• 31/10 '97 14:47 ~+49 2203 303362 INBIFO Koeln ~*~ FTR S&T ~003 INBIFO Institut fQr biologische Forschung OmbH - K01n 31 October 1997 qKHO Page 2 To investigate Stanton Glantz's proposal that platelet aggregation may account for the postulated large effect of exposure to environmental tobacco smoke on dsk of ischaemic heart disease, the authors reviewed published data of experimental studies on platelet aggregation. A corresponding review was done at INBIFO in 1994 (Attachment). Further evaluation is needed to account for the additional Information published since then and will be Included in the final version of this comment. X:~LAW.DOC