Monday, June 12, 2006

Helena Study Itself Reveals Implausibility of Conclusions

The study that is widely cited by anti-smoking groups as demonstrating that smoking bans in bars and restaurants immediately reduce heart attacks by 40% reveals, in its own discussion of the findings, the implausibility of the study's conclusions (see: Sargent RP, Shepard RM, Glantz SA. Reduced incidence of admissions for myocardial infarction associated with public smoking ban: before and after study. BMJ 2004).

Apparently realizing that readers of the article would likely question the plausibility of the conclusion that a mere smoking ban could result in such a drastic (40%) and immediate (within 6 months) reduction in heart attacks, the article analyzes the plausibility of its own conclusion:

"The effect associated with the smoke-free law may seem large but is consistent with the observed effects of secondhand smoke on cardiac disease. Secondhand smoke increases the risk of a myocardial infarction by about 30%; if all this effect were to occur immediately, we would expect a fall of -0.30x40.5=-12.2 in admissions during the six months the law was in effect, which is within the 95% confidence interval for the estimate of the effect (a drop of -32.2 to -0.3 admissions)."

The actual observed reduction in heart attacks in Helena was 16, so the prediction of a decrease in heart attacks by 12 is not too far off from the predicted reduction, under these assumptions.

The Rest of the Story

In essence, what the paper is arguing is that the conclusion of the Helena study is plausible because if one assumes that everyone who develops heart disease and suffers a heart attack from secondhand smoke gets heart disease and suffers that heart attack immediately (within 6 months), the observed decline in heart attacks in Helena is about what one would expect.

In other words, the paper makes the assumption that the effects of exposure to secondhand smoke on heart disease and myocardial infarction are immediate. There is no time necessary to elapse between exposure and suffering a heart attack.

What this means is that a person who is exposed to secondhand smoke acutely must immediately develop heart disease and be at risk of having a heart attack. The whole process of developing heart disease and suffering a heart attack must take place within 6 months in order for the Helena conclusions to be plausible.

The flaw here is that the paper attributes all of the observed increased risk of myocardial infarction among nonsmokers due to secondhand smoke to an immediate (within 6 months) effect of that exposure, and assumes that none of the effect requires more than 6 months of exposure. Thus, all of the effects of eliminating the increased risk in heart disease attributable to secondhand smoke would be realized within 6 months of a smoking ban.

This is clearly an impossible assumption. It cannot be true.

The evidence demonstrates that the effect of secondhand smoke on heart disease observed in the epidemiologic literature is a long-term effect, that results from chronic exposure to secondhand smoke over long periods of time.

Eliminating secondhand smoke (even if a complete ban on all smoking everywhere were implemented in Helena) would not result in an immediate 30% reduction in heart disease and myocardial infarction. It would take many years to see the effects of such an intervention - probably on the order of 10-20 years, if not more.

I am not taking issue with the assumption that based on the epidemiologic literature, a complete smoking ban would result in a 30% decline in heart disease (and heart attacks) over a 20-30 year period. But I am taking issue with the assumption that such a decline in heart attacks would be observed over a 6-month period. The epidemiologic literature simply does not support such an assumption.

Given that the assumption upon which the paper bases its defense of its plausibility is so preposterous, it basically is the case that the paper does not provide an adequate defense of its plausibility. In my view, the Helena conclusion remains a completely unfounded and implausible one, and it should not be widely disseminated by anti-smoking groups because it is not based on sound scientific reasoning (plausibility is, I think, one of the most important aspects of sound scientific reasoning).

No comments: