21/08/08 Heart Attack Rates Since the Ban

Since the introduction of the smoking ban in all enclosed public areas in Britain there have been a few reports claiming it has reduced heart attack admissions in hospitals.  In fact, the ban has been hailed as decreasing heart attacks to a huge extent – almost to the point where passive smoke is being accused of causing the majority of heart disease.  Is this really the case? The report goes against what I have written in Chapter 8: Smoking and Heart Disease, but that is no reason, of course, to dismiss the findings.  All reports must be viewed objectively.  Let’s have a look at what has been said about Scotland and Wales.

The rates of heart attacks in Scotland received sizeable press coverage, to the tune that the ban had made rates drop considerably. Dr Jill Pell claimed in her report that the smoking ban led to a 17% reduction in heart attacks. Evidently Dr Pell is not one to led validity and scientific credulity get in her way of a good yarn, and her study was picked up by major news companies and passed around as the gospel truth.  The fact that the study had not even been published at the time really does speak volumes about how hungry, and insatiable, the media is for anti-smoking figures – regardless of how accurate they are.  As for the study itself, the Guardian tells us that nine hospitals in Scotland, including Edinburgh, Glasgow and Dundee, were studied, making up 63% of heart attack admissins (the other 37% were from the other Scottish hospitals).  We are told that before the ban these nine hospitals reported 3,235 heart attacks, and 2,684 after it. Of these, there were 1,630 in non-smokers pre-ban, and 1,306 post-ban. Putting this all together yields a result of a 19.9% reduction in non-smokers, and 14.1% in smokers, 17% overall.

First things first, the question that comes to mind is ‘if smoking causes heart attacks, why, then, is there a greater reduction in non-smokers rather than smokers?’  Perhaps we can pass this off as an anamoly, but some things just do not shake off that easy, this next point being one of them.  The study has since been published in the New Endland Medical Journal, and has shot itself in the foot. How? Because it lets slip that the test period was two lots of ten months, rather than the standard 12 months.  At first glance this can be seen as a minor detail and not worth nothing.  However, it is far from minor and certainly worth noting, because it is a major sign that the researchers were up to no good. 10 months is essentially cherry-picking the months that directly agree with the hypothesis. Interestingly, and telling, the months neglected were the two months directly following the ban – the ban began in March and April and May were omitted from the report.  Now, the hypothesis was that a smoking ban will instantly lead to a reduction in heart attack admissions, so in actual fact the months of April and May should have been most important!  Further still, we know for a fact that the researchers had data on May because the mission statement, published in the Journal of Public Health stated “Data collection: Continuous May 2005-April 2007“.

Thus, we know for sure that data was collected for May 2005 but was omitted from the final report.  This begs the question ‘but why?’ and the answer can be found in ISD Scotland, which shows that the rate of heart attack reductions in May was exceptionally small, falling by just 1.8%.  Keeping that data in the report would have hindered their hypothesis in two ways: firstly, by showing that a smoking ban does not cause an immediate reduction in heart attacks, and secondly by showing that the true figure was not 17%.

The next problem with the study is that Dr Pell and her colleagues decided not to look at all heart attack admissions, but just acute coronary syndrome (ACS).  We know this because in 2007 Dr Pell presented it to a conference, entitled “‘Changes in myocardial infarction incidence and mortality following the Scottish smoke-free legislation“, and when it was later presented in Edinburgh the British Medical Journal reported the “Scottish evaluation, by Jill Pell and colleagues, reported a 17% reduction in admissions from acute myocardial infarction in nine hospitals in the six months after the ban“.  Acute coronary syndrome (ACS) and acute myocardial infarction (AMI) are two very different things, with AMI being heart attacks and ACS being a little harder to define, but one definition is unstable angina. This would not be a problem, had the Edinburgh presentation not been entitled “Testing the Montana hypothesis: Results from Scotland”.  The reference is to the Helena, Montana study, in which Stanton Glantz was involved and the study is described in Chapter 8 of Smoke Screens.  The Helena, Montana study looked solely at AMI, and so if Dr Pell was really trying to test that hypothesis, she, too, should have looked at AMI over ACS.  So why didn’t she? The data was readily available to her, but she did not use it.  One discrepancy can be perhaps overlooked, but when researchers make it a habit then eyebrows must be raised.  It seems, therefore, that Dr Pell put her agenda before her integrity and used ACS because the reduction rate was far greater than for AMI.

The next problem with the study is it’s sample group.  I will rephrase that: the sample group was large, covering thousands of people. The problem is Dr Pell outright lying about the size of her sample group, as she claimed it was 63% of ACS admssions, when IDS Scotland, official data, shows that in the ten months prior to the smoking ban there were 7,329 ACS admissions – meaning Dr Pell’s sample group of 3,235 was actually 48.5%, less than half.  Again, if she had been honest about it then it would be what it is. However, she was not honest and lied to make her study look more credible than it really is, which again highlights that she had an agenda and was after specific data, not accurate data. It is also worth noting that the ACS admissions have been falling year on year, and is not something that has occured only since the ban was put into effect.

It can be concluded, then, that Dr Pell et al selected the type of heart disease they would study, knowing that it would offer better results, omitted crucial and necessary months from the time period as they would have undermimed her hypothesis, and lied about her methodology and her results. The only recognition she deserves for this work is one of a woman with no scientific integrity and who puts her agenda before science. The study is archetypal junk science – at least she got one thing in common with the Montana study.

References:
http://jpubhealth.oxfordjournals.org/cgi/reprint/fdi080v1.pdf

http://content.nejm.org/cgi/content/abstract/359/5/482

http://www.isdscotland.org/isd/information-and-statistics.jsp?pContentID=2446&p_applic=CCC&p_service=Content.show&

Welsh heart attacks admissions

Leave a Reply

Your email address will not be published. Required fields are marked *