Thursday, December 23, 2010

Data from Scotland Show No Decline in Heart Attack Admissions Three Years After Smoking Ban

Data released by the Scottish National Health Service show that hospital admissions for acute myocardial infarction (heart attacks) in all of Scotland have failed to decline significantly after three years of the smoking ban being in effect.

The Scottish National Health Service has published monthly data on heart attack admissions in all Scottish hospitals for the one-year period prior to the smoking ban and the three year period following the ban. The smoking ban went into effect in April 2006. The graph below shows the number of hospital admissions for heart attack in Scotland during the following comparable periods:

April 2005 - March 2006: Pre-ban
April 2006 - March 2007: First year post-ban
April 2007 - March 2008: Second year post-ban
April 2008 - March 2009: Third year post-ban

Here are the data visually:

Figure 1. Heart Attack Admissions for All of Scotland By Equivalent 12-Month Periods Before and After the National Smoking Ban

Here are the actual numbers:

Pre-ban: 8308
First year post-ban: 7765
Second year post-ban: 7333
Third year post-ban: 8115

It is readily apparent from these data that although there was a small decline in heart attack admissions following the smoking ban, that decline was not sustained and heart attacks have now risen back up to the baseline level. Thus, the declines were apparently not due to the smoking ban and probably reflect random variation or secular changes in heart attacks that have nothing to do with the smoking ban.

Overall, from the year prior to the smoking ban to the 3rd year after the ban, there was a slight 2.3% decline in heart attack admissions.

The Rest of the Story

These data demonstrate that the conclusions made by Pell et al. in their study of the smoking ban's effect on heart attacks in Scotland were premature. While that study only looked at the first year after implementation of the smoking ban and only collected data for a sample of hospitals, the present data cover a full three years after the smoking ban and include admissions from all hospitals in Scotland.

Coming on the heels of the revelation that heart attacks actually increased in Isle of Man following its smoking ban and that the largest study to date more than 2 million heart attack occurrences throughout the United States found no significant relationship between smoking bans and short-term changes in heart attack admissions, these data add further support to my assertion that there simply is not evidence to conclude that smoking bans cause an immediate reduction in heart attacks, as is being claimed by many anti-smoking groups.

This shows the dangers of disseminating conclusions to the public that are not based on rigorous scientific methods. If you end up being wrong, you are going to look bad and your scientific reputation is going to be shot. Better to be more careful in the first place and to be beyond reproach in your scientific integrity.

Special thanks to Christopher Snowdon over at Velvet Glove, Iron Fist, who first reported and analyzed these data. Snowdon provides more detailed data, including a month-by-month analysis, data for both heart attacks and acute coronary syndrome, and excellent graphs which nicely illustrate the true trends in heart attack admissions in Scotland.

No comments: