■ site map
BLOG: December 2008
Secondhand smoke, or a second-rate study?
Would you believe if someone told you that secondhand smoke is causing over 40% of all heart attacks? In other words, that nearly as many non-smokers die from ill effects of inhaling cigarette smoke on cardiovascular health, as do those who actually smoke? Probably not - it just doesn't seem logical.
Not so to the authors of a recent CDC (Centers for Disease Control) study, based on hospital records in Pueblo, Colorado. The reason they chose Pueblo is that the city had workplace smoking ban enacted in the July 2003. They found out that in the three years following the enactment, number of heart attack related admissions in Pueblo dropped by an astounding 41%.
At the same time, number of heart-attack-related hospital admissions in two nearby counties, which did not have workplace smoking ban, did not change significantly.
According to the authors - and some others commenting on it - this suggests that secondhand smoke could be "a terrible and under-recognized cause of heart attack deaths in this country". Is this conclusion really based on facts - as it seems to be, at first - or misses something important?
A closer look at the study reveals some gaping holes in how it was conducted, and conclusions drawn. The actual levels of second hand smoke - the very basic subject of the study - were not measured and compared. Only the total of hospital admissions was used, without distinguishing between smokers and non-smokers. It is very likely that the ban, if effectively enforced, has been significantly more beneficial to smokers - already much more exposed to cigarette smoke - than non-smokers.
Other factors possibly influencing the heart attack rate have not been considered. Even one as significant, and as directly affecting it as the overall rate of smoking. In this time frame, in the Pueblo county it has declined from 26 to 21%. That alone nearly cuts in half the rate presented in the study as related to the effect of secondhand smoke only.
This also indicates that those who still smoke do so less, on average, than before. And that anti-smoking activity in the county is consistent, effective, and likely to be comprehensive, not limited just to workplace smoking. In short, factors that were omitted from the study were probably more significant than that one single assumed factor - lower levels of secondhand smoke in Pueblo's workplace. We can definitely say: too significant to be omitted.
CDC's own official estimate is that less than 7% of cardiovascular deaths in the country are attributable to secondhand smoke. One would think that this huge discrepancy between 7% and 41% would have signaled to the authors that they need to re-examine their study. It didn't. For some reason, they felt that their assumption-based, fact-deprived study construction was good just as it is.
Inflicting secondhand smoke to others is something that no one should do - or be allowed to do - at home, work, public place, or anywhere else. Another thing that shouldn't be tolerated is releasing incompetent, misleading studies by a government agency - or anyone else. Obviously, the reality is different. Just another reminder not to trust "study results" without taking very close look at how it was conducted.