Friday, September 28, 2007

New Study Claims that New York State Smoking Ban Reduced Heart Attack Admissions by 8%

A study being released today concludes that the New York State smoking ban, implemented in July 2003, resulted in an 8% decline in heart attack hospital admissions statewide during the first year and a half it was in effect. The study was published online ahead of print today in the American Journal of Public Health (see: Juster HR, Loomis BR, Hinman TM, et al. Declines in hospital admissions for acute myocardial infarction in New York State after implementation of a comprehensive smoking ban. Am J Public Health 2007).

Although the study did not determine trends in heart attacks in smokers versus nonsmokers, it notes that there was not a significant decline in smoking prevalence associated with the implementation of the statewide smoking ban; thus, the decline in heart attacks is attributed to decreased secondhand smoke exposure among nonsmokers.

The study examined trends by month in age-adjusted hospital admission rates for acute myocardial infarction in all non-federal New York hospitals between 1995 and 2004. It used regression analysis to control for the effects of seasonal and secular trends as well as the presence of local smoking bans.

The paper concludes: "Rates of hospital admissions for AMI [acute myocardial infarction] were reduced by 8% after a comprehensive ban on smoking in work sites, including hospitality venues (e.g., bars and restaurants), in New York State. This is equivalent to a reduction of approximately 3800 AMI hospital admissions in 2004 and an estimated cost savings of $56 million. Our results show that enactment of clean indoor air laws was associated with an accelerated decline of hospital admissions and that a comprehensive statewide law had the largest effect."

The study results were first reported by the Associated Press here.

The Rest of the Story


While I do not dispute the conclusion that there was an 8% decline in hospital admissions in New York State during the second half of 2003 and all of 2004, I do not believe that these data support a conclusion that the observed decline in heart attack admissions is attributable to the statewide smoking ban.

Why? Because there is no control or comparison group. The article examines trends in heart attacks in New York, but it does not examine what happened to heart attack admissions anywhere else during the same time period.

Sure - there was an 8% decline in heart attack admissions in New York during 2004. But this doesn't mean anything unless there was not an 8% decline in heart attack admissions in other states, that did not implement smoking bans, in 2004. Without knowing what the heart attack trends were elsewhere, there is no basis to conclude that the observed decline in heart attack admissions in New York was attributable to the smoking ban.

This is essentially an uncontrolled study - a study without a control group. It represents a very weak study design for this type of analysis. Even the shoddy Bowling Green and Pueblo studies employed a comparison group to evaluate whether the observed changes in heart attacks in the cities with smoking bans were also occurring in comparable cities without smoking bans.

So what did happen in other states between 2003 and 2004? Were heart attack admissions stable in states without smoking bans, while dropping by 8% in New York state with its smoking ban in place?

Well, in South Carolina, heart attack admissions fell by 12.5% from 2003 to 2004. This is also in marked contrast to the existing trend in heart attacks in that state. Heart attack admissions were increasing by an average of 3.0% per year during the period 2001-2003 in South Carolina. So was the 12.5% decline in heart attack admissions during the same time period in South Carolina due to the absence of a statewide smoking ban?

In Nebraska, heart attack admissions fell by 28.5%from 2003 to 2004. This is in marked contrast to the existing trend in heart attacks in the state. Heart attack admissions were increasing by an average of 2.3% per year during the period 2001-2003 in Nebraska. Was this dramatic drop in heart attack admissions in Nebraska, which occurred during exactly the same period as the decline in New York, attributable to the absence of a smoking ban in the Cornhusker state?

In fact, in all other states for which data are available, heart attack admissions fell by 5.1% from 2003 to 2004. So does the decline in heart attacks in New York represent a dramatic effect of the smoking ban, or is it simply a change that would have been expected based on the secular trends occurring across the nation during the same time period?

The article in question does not allow us to answer that question. And that is precisely why its conclusion is unwarranted.

Based on the available data, it certainly appears that large declines in heart attack admissions were occurring across the nation in 2004. The observed decline in New York was far less than was observed in Nebraska and South Carolina - two states without smoking bans.

So does this mean that the absence of a smoking ban in those two states was the reason why their heart attack admissions dropped far more than in New York? Of course not. The point is that there are large year-to-year variations in heart attacks that have nothing to do with smoking bans and in order to conclude that a small decline (such as 8%) in heart attacks was due to a smoking ban, you absolutely have to show that the decline would not have occurred in absence of the smoking ban. And to do that, you need to look at what is happening in other states.

One lesson here is that even if a study is published, you still need to review it critically and you should not necessarily assume that its conclusions are valid and well-supported. And if this is what can happen with a published article, you can only imagine how much more difficult it is to accept the conclusions of a study that is neither published nor available (i.e., the Scottish smoking ban heart attack study).

As much as we might like to believe that reducing secondhand smoke exposure prevents thousands of heart attacks in a matter of months, the evidence is simply not there to support such a conclusion. By jumping the gun and drawing conclusions prematurely, I fear that we are hurting our overall scientific credibility. In the long run, that may harm the effort to promote smoking bans far more than spreading the belief that such bans are going to immediately prevent heart attacks is going to help enact these bans.

No comments: