Thursday, October 08, 2009

Anti-Smoking Researcher Claims that Smoking Bans Reduce Heart Attacks Within Minutes of Implementation

We're all familiar with the claims by some anti-smoking advocates that smoking bans reduce heart attacks relatively quickly after implementation, such as over a period of six months or one year.

But now, an anti-smoking researcher has claimed that these bans reduce heart attacks within just minutes of implementation.

According to an article in the Lawrence Journal-World & News, Dr. David Meyers, professor of cardiology and preventive medicine at Kansas University Medical Center and lead investigator of a new study on smoking bans and heart attacks, claimed that: "Within minutes of the ban, it is going to start having an effect on heart attacks."

Meyers was quoted as supporting his claim with the following argument: "Heart attacks are caused in large part by blood clots. With 20 minutes or so of tobacco smoke exposure, people’s blood becomes hypercoagulable and sticky and clots easily, and bam, you have a heart attack."

The study being reported is a meta-analysis of 11 previously published studies that examined the effect of smoking bans on heart attack rates in specific communities (such as Helena, Pueblo, and Bowling Green) or in countries with national smoking bans (such as England, Italy, and Scotland). The meta-analysis found that: "Using 11 reports from 10 study locations, AMI risk decreased by 17% overall (IRR: 0.83, 95% CI: 0.75 to 0.92), with the greatest effect among younger individuals and nonsmokers." The study concluded: "Smoking bans in public places and workplaces are significantly associated with a reduction in AMI incidence, particularly if enforced over several years."

The Rest of the Story

What readers need to understand is that a meta-analysis is only as good as the individual studies that go into it. If the individual study conclusions are invalid, then the meta-analysis will be invalid as well. This is exactly the case with the present study.

I have previously analyzed each of the published studies on smoking bans and heart attacks and explained why the conclusions of these studies are invalid. You can't just combine the studies in a meta-analysis and argue that suddenly the conclusion becomes valid. The meta-analysis does not account for the severe flaws in these studies, including the failure to adequately rule out the possibility that the observed declines in heart attacks merely reflected a combination of random variation plus an already declining secular trend in heart attacks over time.

But the most telling fact about this meta-analysis is that it fails to incorporate any control or comparison population. In other words, it includes studies even if they did not employ a comparison group. And in the studies that did include a comparison group, it appears to throw out the data regarding the comparison group.

What this means is that the meta-analysis is not designed to estimate the effect of smoking bans on heart attacks. What it is designed to do is to determine the change in the incidence of heart attacks over the time period in these studies. What the meta-analysis demonstrates is that across all study populations, there was a significant decline in heart attack rates.

But we already knew that. For many reasons, including better treatment for heart disease (both surgical treatment and medications), heart attack rates have generally been declining, even in the absence of smoking bans. The relevant question for the meta-analysis should have been: are these observed declines due to the smoking ban? Instead, the meta-analysis simply asked the question of whether there was or was not a decline in the first place? This is useful information, as it confirms our a priori impression that rates have been declining, but it offers no evidence that the declines are due to smoking bans, as opposed to other changes, such as improved diagnosis of minor coronary events, earlier diagnosis of unstable angina, improved surgical treatment (angioplasty) for coronary artery disease, and greatly improved medical treatment (e.g., statins to bring cholesterol levels under control).

In essence, what we have here is a meta-analysis of studies without comparison groups!

I don't see how you can possibly include in this meta-analysis studies that failed to include a comparison group. There's simply no way to know whether the observed decline in heart attacks was attributable to the smoking ban or not.

In short, the meta-analysis presents the wrong analysis. It should not present the estimated change in heart attack rates before and after the smoking bans. We know that the rate is going to go down significantly because of the known secular trends in heart attacks, which are declining everywhere, even in the absence of smoking bans.

Instead, the correct analysis would have been to examine the individual estimates of the difference between the declines in heart attack rates in intervention (i.e., smoking ban) versus comparison communities or countries.

As my readers know, I strongly support workplace smoking bans. Nevertheless, I believe they should be supported based on valid scientific conclusions, not on junk science conclusions such as those in this article.

Another telling finding in the meta-analysis is that the studies which found the large declines in heart attacks (and which drive the findings of the whole meta-analysis) were the studies of the smallest communities, where there is the greatest variation in heart attack rates. The larger studies, with very large populations, failed to find substantial effects.

This suggests that the conclusions are due primarily to a few anomalous findings in small communities with very few heart attacks. When one examines the results among a large population, one fails to find the reputed effects.

The meta-analysis itself reports that there was no effect of the smoking ban in either the state of New York (the largest population studied) or the country of Italy (the second largest population studied). The results appear to be entirely driven by the findings in Helena, Monroe County, and Pueblo, which include two of the smallest populations studied (there were only 17 heart attacks in Monroe County to begin with and only about 40 in Pueblo during the post-ban period).

The most interesting aspect of this story, however, is that even if we stipulate for the sake of argument that the study conclusions are correct, what the results would show is that over a period of time -- months to years -- there is a decline in heart attacks. The study does not support the assertion that there is a decline in heart attacks within minutes of a smoking ban.

I have to take issue with the explanation for this immediate effect. As put by one of the study authors: "Heart attacks are caused in large part by blood clots. With 20 minutes or so of tobacco smoke exposure, people’s blood becomes hypercoagulable and sticky and clots easily, and bam, you have a heart attack."

If this statement were true, then we would see large numbers of previously healthy people dropping dead of heart attacks in smoky bars after 20 minutes. If it is true that your blood clots easily and bam you have a heart attack, then many healthy people going into a smoky bar would leave the bar after 20 minutes in an ambulance. We just don't see that happening (except, perhaps, for those who severely overindulge to the point of alcohol intoxication -- mostly college freshmen).

The truth is that brief secondhand smoke exposure is likely to trigger a heart attack only in people with severe existing coronary artery disease. And for such individuals, there are so fragile that any exposure which increases platelet aggregation and causes endothelial dysfunction -- is also likely to trigger a heart attack. There is no mechanism I know of by which secondhand smoke is the only exposure that can trigger a heart attack in some who is brittle enough so that a mere 20 minute exposure to tobacco smoke is going to trigger a heart attack. The same hypercoagulability and endothelial dysfunction is also caused by eating high-fat foods and even by mental stress. It doesn't follow that you are going to prevent this person from having a heart attack merely by asking them to avoid exposure to secondhand smoke. Moreover, there is simply no scientific evidence to support the assertion that by avoiding secondhand smoke exposure, we will prevent heart attacks among individuals with severe coronary artery disease.

The rest of the story is that anti-tobacco researchers and groups are making ridiculous, highly exaggerated, and scientifically unsupported claims in order to try to justify smoking bans. While I support the very smoking bans which these groups are promoting, I do not support the junky science and wildly exaggerated and misleading claims that are being made to the public.

No comments: