Correspondence
The Hazards of Active and Passive Smoking
N Engl J Med 1993; 329:1580-1581November 18, 1993
- Article
To the Editor:
Chilmonczyk et al. (June 10 issue)1 demonstrated a dose-response relation between exacerbations of asthma in children and exposure to environmental tobacco smoke, and validated parental reports of exposure by measuring urine cotinine levels. There have been direct studies demonstrating that deliberate exposure to environmental tobacco smoke resulted in “shortness of breath,” “tightness in the chest,” and “wheezing” and that these symptoms were twice as frequent in subjects with asthma (25 percent) as in nonasthmatic subjects (12 percent)2,3. These direct studies add validity to the conclusions of larger retrospective investigations.
Previous research suggests a possible interaction between sex and bronchial reactivity -- namely, that the response may be stronger in boys4. Chilmonczyk et al. report that boys predominated in both the group exposed and the group not exposed to environmental tobacco smoke but do not address sex differences beyond this observation. Do their data support such an inference?
4 ReferencesR.S. Richardson, Ph.D.
University of California, San Diego, La Jolla, CA 92093-06231
Chilmonczyk BA ,Salmun LM ,Megathlin KN , et al. Association between exposure to environmental tobacco smoke and exacerbations of asthma in children. N Engl J Med 1993;328:1665-1669
Full Text | Web of Science | Medline2
Shephard RJ ,Collins R ,Silverman F . “Passive” exposure of asthmatic subjects to cigarette smoke. Environ Res 1979;20:392-402
CrossRef | Web of Science | Medline3
Shephard RJ ,Ponsford E ,LaBarre R ,Basu PK . Effect of cigarette smoke on the eyes and airway. Int Arch Occup Environ Health 1979;43:135-144
CrossRef | Web of Science | Medline4
Martinez FD ,Antognoni G ,Macri F , et al. Parental smoking enhances bronchial responsiveness in nine-year-old children. Am Rev Respir Dis 1988;138:518-523
Web of Science | Medline
To the Editor:
There are a number of erroneous statements in the editorial by Boyle (June 10 issue)1.
The author implies that the results of research on a link between parental smoking and brain tumors in children are positive; in fact, no support for this hypothesis was given by the cited work.
The selective use by the Environmental Protection Agency (EPA) of 11 American epidemiologic studies of women married to smokers does result in a relative-risk ratio of 1.19,2 but this figure is not statistically significant when meta-analytic errors and inconsistencies are corrected and when standard statistical methods (two-tailed test with 95 percent confidence limits) are used. Surprisingly, the EPA chose to use a single-tailed test with 90 percent confidence limits to assess the risk associated with being married to a smoker2. None of the 11 studies themselves are significant at the 95 percent level, and only 1 is significant at the 90 percent level. The addition to the 11 U.S. studies of 2 recent reports,3,4 for some reason excluded by the EPA, also results in a nonsignificant relative risk (for both sets of confidence limits). The statement “there was a trend toward an increase in risk with increasing exposure” is meaningless, since exposure to environmental tobacco smoke was not measured in any of the studies (instead, questionnaires were used). The trend is significant only when the control (unexposed) group is included. To arrive at its conclusion, the EPA disregarded its own guidelines for performing such work5. These guidelines state that causation can be inferred only when “there is no identified bias that could explain the association,” when “the possibility of confounding has been considered and ruled out as explaining the association,” and when “the association is unlikely to be due to chance.” In addition to using questionable techniques for assessing chance, the EPA failed to correct the results of its meta-analysis for confounding factors or for bias (other than providing an erroneously low estimate for misclassification of smokers as nonsmokers).
Sidestream smoke is not “exhaled” smoke; rather, it is the smoke produced by the burning cigarette between puffs. Environmental tobacco smoke is not a quantitative variant of mainstream smoke. Boyle's comparisons and unsupportable conclusion do not withstand even minimal scientific scrutiny.
5 ReferencesChris Coggins, Ph.D., D.A.B.T.
R.J. Reynolds Tobacco Co., Winston-Salem, NC 271021
Boyle P . The hazards of passive -- and active -- smoking. N Engl J Med 1993;328:1708-1709
Full Text | Web of Science | Medline2
Respiratory health effects of passive smoking: lung cancer and other disorders. Washington, D.C.: Environmental Protection Agency, 1992.
3
Stockwell HG ,Goldman AL ,Lyman GH , et al. Environmental tobacco smoke and lung cancer risk in nonsmoking women. J Natl Cancer Inst 1992;84:1417-1422
CrossRef | Web of Science | Medline4
Brownson RC ,Alavanja MC ,Hock ET ,Loy TS . Passive smoking and lung cancer in nonsmoking women. Am J Public Health 1992;82:1525-1530
CrossRef | Web of Science | Medline5
The risk assessment guidelines of 1986. Washington, D.C.: Environmental Protection Agency, 1987.
Author/Editor Response
The authors reply:
To the Editor: We have evaluated the possible relation between sex and bronchial reactivity in our study population of asthmatic children, as Dr. Richardson suggests. The outcome measurements available for analysis include the number of acute exacerbations of asthma during the year preceding enrollment and three measures of pulmonary function obtained at the time of enrollment. This analysis was performed by including an interaction term between sex and the measure of exposure (either a parental report or a urine cotinine measurement) in the regression model specified in Table 3 and Table 4 of our paper. Exposure to environmental tobacco smoke was associated with a slight increase in the average number of acute exacerbations in boys as compared with girls, but this difference did not approach statistical significance. No consistent differential effect of exposure attributable to sex was found for the other outcome measures.
James E. Haddow, M.D.
Glenn E. Palomaki, B.S.
Foundation for Blood Research, Scarborough, ME 04070-0190Barbara A. Chilmonczyk, M.D.
Maine Medical Center, Portland, ME 04102Author/Editor Response
Coggins raises several points that could have been resolved by a more careful reading of my editorial. He also rails against the EPA's recent report and the performance of the meta-analysis contained therein1. Those interested should read the original report.
It appears that my editorial has been the subject of detailed dissection, and Coggins has responded with the tobacco industry's objections, which are mainly irrelevant to the points I raised. For example, I stated, “Research on the hazards of smoking has turned recently to the association between exposure to environmental tobacco smoke . . . and brain tumors in children”; this statement does not imply any positive association. However, the most interesting feature of Coggins's response is its implicit acceptance of many of the points concerning tobacco-related cancer that I refer to in my editorial: the strong link between tobacco smoking and cancer; the devastating effects of tobacco smoking on cancer-related mortality and premature mortality currently being seen in Eastern Europe and building up in China; and the increasing rates of death from smoking-related lung cancer among women throughout the world. Such an implicit acceptance of the hazards of smoking by tobacco manufacturers introduces a new element into the continuing debate on many issues, including what labeling information should be included on cigarette packages; whether cigarette advertising should be banned, smoking should be restricted in public places, or cigarettes should be promoted in developing countries; and how to foot the bill for the medical costs incurred by smokers.
Coggins, who is from the R.J. Reynolds Tobacco Company (motto: “We work for smokers”), provides no facts, only bluster, to challenge my conclusion, “There is, however, no need for new evidence to justify strict measures to curb the use of this deadly substance [tobacco], which claims more lives around the world each year than war.” As the tobacco epidemic continues, governments are increasingly turning to legislation as part of their campaign to reduce the morbidity and mortality from tobacco-related diseases. The most recent survey2 indicates that over 90 countries and territories have some type of legislation to control tobacco use, and many of the recently enacted statutes are stronger and more effective than earlier laws. It is obvious that there is an ever-increasing need for activities to control the use of tobacco throughout the world and support for these activities from everyone concerned with public health.
2 ReferencesPeter Boyle, Ph.D.
European Institute of Oncology, 20141 Milan, Italy
