The graph above charts the rate of heart attack admissions (acute myocardial infarction) in England over the past few years. The arrow points to July 1, 2007, when England implemented its smoking ban in pubs and restaurants. To the untrained eye it may appear that the ban had no discernible impact on the rate of heart attacks. Ah, but to the trained eye of an anti-tobacco activist this graph demonstrates that smoking bans save lives. Hence the headlines in the British press this week declaring the ban a roaring success: "At least 1,200 heart attacks were prevented in England in the year after the ban's introduction," reports The Daily Mail. Critics of the nanny state "get their comeuppance with research showing that nannying works," says the Independent. Even the usually savvy Economist lauds the study as proof that owners of private places should have no right to allow smoking on their premises.
The dubious credit for turning this unremarkable decline into front page news goes to Dr. Anna Gilmore at the University of Bath. Her own data show no apparent impact from the smoking ban. The number of heart attack admissions in each year is copied here along with rates of decline calculated by journalist Christopher Snowdon:
2003/04: 60,680 (a fall of 1.33%)
2004/05: 58,803 (a fall of 3.1%)
2005/06: 55,752 (a fall of 5.19%)
2006/07: 53,964 (a fall of 3.21%)
2007/08: 51,664 (a fall of 4.26%)
The decline in 2007/08, the first year of the smoking ban, is right in line with previous years. In fact the decline in 2005/06 was significantly higher. So how does Dr. Gilmore conclude that the smoking ban reduced heart attack admissions by 2.4%? In short, by manipulating the data in every way imaginable. She adjusts them for air temperature, flu seasons, holidays, and population. At the end of these calculations we are supposed to believe that these factors account for all other variations in heart attack rates and that absent the smoking ban the ongoing decline in these rates would have inexplicably slowed in 2007/08. That's a lot to swallow, especially without a control population to use for comparison. As tobacco researcher Dr. Michael Siegel writes in criticism:
Studies of the effects of interventions that do not include a comparison group of any kind generally represent an extremely weak study design. With an huge effect size, you can sometimes get away without a comparison group because it is implausible that an effect that large represents a common secular trend. However, in this case, the effect size is miniscule (2.4%) and there is simply no way one can credibly attribute that small an effect to the smoking ban without ruling out the possibility that it is, instead, simply a reflection of a decline in heart attacks that is occurring anyway due to secular trends.
It gets worse. As the study admits, the data do not include any information on the smoking status of heart attack patients. Thus even if the decline is real it may be caused more by smokers having fewer cigarettes than by any change in exposure to non-smokers, which is who the law was intended to protect.
To conclude that England's smoking ban caused a precise decline in AMI rates of 2.4% or prevented 1,200 heart attacks is pretty much pure speculation. The real story here is how far the anti-smoking movement has fallen in its claims linking bans and heart health. In 2004 Stanton Glantz made the astonishing claim that a ban in Helena, Montana caused a 40% drop in heart attacks in just six months. A 2007 study of tiny Bowling Green, Ohio claimed a 47% drop in three years. Last year an updated study of Pueblo, Colorado concluded that its ban reduced rates by 27% in the first 18 months of implementation and 41% over three years.
Critics of these so-called "heart miracles" argued that these dramatic numbers were the result of studying small towns with large year-to-year fluctuations in the data. Indeed, as researchers turn their attention to larger populations the conclusions get more conservative: an 11% reduction in Piedmont, Italy, then 17% in Scotland, which researchers sheepishly corrected to 8% a few weeks later.
And now we come to England with its population of 49 million people. Here, armed with a battery of statistical techniques and a lead author who serves on the board of UK Action on Smoking and Health, the best researchers can dig up is a decline of 2.4%. If the anti-smoking movement has any integrity we will hear no more about miraculous declines in heart attack admissions or false promises of lives and money saved by smoking bans. If science and health reporters take their jobs seriously they will treat the next press release boasting dramatic results with skepticism. We can move forward knowing that the most robust study so far conducted by anti-smoking researchers finds at best a small effect.
Alas, I expect we'll be back to more of the same in no time.
[Image courtesy of Christopher Snowdon, whose book Velvet Glove, Iron Fist is highly recommended.]
The right to a trial by jury is among the most fundamental in the Constitution. Juries are guaranteed three times in the Bill of Rights: grand juries in the Fifth Amendment, criminal juries in the Sixth, and civil juries in the Seventh.
This year’s health care reform bill included a measure to require chain restaurants and vending machines to post calorie information for the food they sell, taking up space on menus and menu boards.