Thursday, March 25, 2010

Article Questions Conclusions About Smoking Bans and Heart Attacks, Highlights Weaknesses in IOM Committee's Conclusions

An article by Dr. Michael Marlow, a Professor at California Polytechnic State University in San Luis Obispo, questions the conclusions of many anti-smoking groups and researchers, including the Institute of Medicine (IOM) committee which examined the issue, that smoking bans lead to dramatic, immediate reductions in heart attacks. An IOM committee concluded that the scientific evidence is sufficient to conclude that smoking bans significantly reduce heart attacks, although the committee was unwilling to estimate the magnitude of the effect. The article appears in the spring issue of the Journal of American Physicians and Surgeons.

Dr. Marlow points out a number of severe weaknesses in the existing studies which were purported to show that smoking bans caused reductions in heart attacks. First, the sample sizes are quite small and only select communities are studied. In contrast, a systematic study of all communities in the U.S. found no effect of smoking bans on heart attack mortality rates. The reason for the positive findings of these small sample studies may be publication bias. Researchers may not pursue the publication of negative findings or such studies may have a more difficult time finding acceptance in journals.

Second, very few of the studies examined smokers and nonsmokers separately. Thus, they cannot possibly conclude that any observed reduction in heart attacks is attributable to decreased secondhand smoke exposure, rather than lower rates of smoking. Nevertheless, this methodologic flaw does not stop most of the researchers from drawing such a conclusion.

Third, none of the studies account for other secular changes that could explain an observed reduction in heart attacks. The most important of these variables is improved treatment of cardiovascular disease.

Fourth, the kinds of dramatic changes in heart attacks that are being attributed to smoking bans in some studies are simply implausible. Even if all secondhand smoke were to be instantly eliminated, epidemiologic calculations would not predict anything as high as a 47% reduction in heart attacks.

Dr. Marlow concludes: "Publicly led research on public health effects of smoking bans has overstated benefits by overreaching on conclusions, excluding studies that contradict predetermined conclusions, and relying on studies subject to biases outlined above. This pattern is lamentable for a number of reasons. One is that efforts claiming to improve public health appear to be driven more by social agendas than by science. The IOM released, and various media outlets promulgated overstated claims on the public benefits of smoking bans, apparently without even considering whether they met the simplest tests of believability."

The Rest of the Story

I agree that the overreaching on conclusions and cherry-picking of studies by anti-smoking groups and researchers is lamentable. Perhaps the most troubling example is the IOM committee itself, which failed to consider the considerable data which failed to show any effect of smoking bans on heart attacks.

I also agree that over-reaching on conclusions has unfortunate deleterious effects. Perhaps the greatest one is the potential for undermining the public's trust in public health. If health groups exaggerate scientific information to the public, then the public may discount very important, valid messages that these same groups offer in the future. An excellent example of this is the very low rates of immunization for H1N1 influenza, which I think is largely attributable to over-exaggeration of the risks of a worldwide pandemic. When the public saw that the dire warnings never materialized, they lost confidence in anything public health groups had to say about the H1N1 flu. This is unfortunate, because it still is a very serious epidemic and resulted in more than 10,000 deaths.

The most prized possession of public health is its credibility. Lose that, and we lose our ability to communicate vital messages to the public and have them taken seriously. Anti-smoking groups are not doing themselves any favors in the long run by risking their credibility and throwing their scientific integrity to the wind.

No comments: