Welsh heart attack rate rises after smoking ban

There has recently been some controversy about Jill Pell's Scottish heart attack study, a lamentable piece of epidemiology which has been debunked here and elsewhere. A few weeks ago, the British tabloid newspaper The Daily Mail told its readers - quite wrongly, as it turned out - that there had been a major fall in heart attack incidence in England.

But what of Wales? Under the Freedom of Information Act, I have been able to obtain admission data from all Welsh hospitals for the years 2006 and 2007. I have collated the data for acute myocardial infarction (AMI) (ie. heart attacks), to see if there was any unusual decline in incidence following the smoking ban's introduction on April 2 2007.

The graph below shows the year-on-year percentage change in AMI admissions from 2006 to 2007 (January to December). The arrow indicates the start of the smoking ban (April 2).

You will notice that there was a year-on-year fall in six months of 2007, and a rise in the other six months. In total, there were 4,199 heart attacks in 2006 and 4,155 in 2007. In other words, there was practically no difference between the two years. This is slightly surprising, since heart attack incidence has been falling in Wales for years, as it has in the rest of Britain. There was a 6.3% drop the previous year and a 10.3% drop the year before. Nevertheless, the fact remains that this long-term decline ground to a halt in the year of the smoking ban.

Furthermore, there was a sustained rise in heart attack admissions in the first five months of the smoking ban compared to the same period in the previous year. If the Helena hypothesis is correct there should have been a noticeable drop straight after the ban. The data from Wales - seen here for the first time - provides strong evidence that smoking bans have no effect whatsoever on heart attack incidence. If we were inclined to use the logic of Pell, Glantz et al, we might even say that the smoking ban 'caused' a rise in heart attacks in Wales. That, of course, would be nonsense, just as the reverse hypothesis is nonsense.

Considering that 5 out of the 9 post-ban months showed a rise in heart attack admissions, you would expect the anti-smoking lobby to maintain a discrete silence about these figures and hope that nobody notices them. Astonishingly, and shamelessly, they have instead told the public that Wales experienced another heart attack miracle. This has been done in the most underhand way.

This is how the statistics were relayed to the public on June 30 this year:

"The number of patients suffering a heart attack in Wales has fallen dramatically following the ban on smoking. New figures unveiled today appear to support claims that the smoking ban would improve the health of the nation. And it could signal that Wales has taken the first steps towards shedding its 'sick man of Europe' title.

Across the nation as a whole there was a 12.5% fall in the number of patients admitted to hospital with a heart attack between October and December last year, compared to the same period in 2006, before the ban on smoking in enclosed public spaces was introduced in Wales."

"The Welsh figures, which are released today by Plaid Cymru, reveal that between October and December 2007 the number of patients admitted to hospital following a heart attack fell 22% in South East Wales, 6% in North Wales and 5.4% in Mid and West Wales.

It would appear that the smoking ban has coincided with an acceleration in the decline of heart attacks."

Chris Franks, the Plaid Cymru AM for South Wales Central*, who obtained the information, said: "While there are clearly several factors that have resulted in a reduction in treating heart attacks, I believe the smoking ban is an important element.

"These figures cover the period six months after the ban and I would hope to see the reductions in heart attacks over the months and years to come increase further."

And in The Daily Post on July 1st:

"Welsh hospitals reported a big fall in treating heart attacks in the wake of the ban on smoking in public places last year. In the three months from October to December 2007 - the date when the last full figures are available - there were 4,669 heart-related emergency admissions. This compares to 5,339 in the same period of 2006 and 5,452 in 2005, according to data released to Plaid Cymru under the Freedom of Information Act."

This is an outrageous distortion of the truth. It is plainly obvious to anyone who looks at 2007's Welsh heart attack statistics that heart attack incidence was unusually high in 5 out of the 6 months that immediately followed the ban. To exclude these months from analysis and focus solely on the last three months of the year (when incidence declined) is deeply dishonest. Two thirds of the data were ignored because - there can be no other explanation - it did not fit the preconceived hypothesis. This is as blatant a piece of statistical manipulation as one will ever encounter.

It could be argued that the anti-smoking groups did not have a hand in this charade and are therefore blameless. But while it is true that the misinformation came from a single member of the Welsh Assembly, the anti-smoking groups were more than happy to endorse it:

"Dr Mike Knapton, director of prevention and care at the British Heart Foundation, said: 'These new statistics are very significant, and indicate the smoking ban has had a beneficial effect on the number of heart attacks quicker than many people predicted.'"

"Dan Clayton, a spokesman for anti-smoking body ASH Wales, said: 'One of the major factors in ASH Wales' support for the smoking ban in all enclosed public places was the fact that it was a measure that would save lives. Second-hand smoke is extremely dangerous and can be associated with premature death, including heart attacks. Data from Ireland and Scotland has already shown a drop in reported heart attacks. That data is now coming out of Wales, showing a drop can be celebrated.'"

By associating themselves with the 'study', these groups must share culpability in what was clearly a deliberate attempt to deceive and mislead the public. As for the media, it is easy to scorn their credulity but journalists cannot be blamed for trusting the word of elected politicians and health groups. They should not have to go to the trouble of making Freedom of Information requests to verify what they are told. The reality is that journalists have neither the time nor - very often - the ability to check complicated raw data. They expect to be able to trust their sources. Increasingly, it seems that these sources are, at best, untrustworthy and, at worst, downright dishonest.

Although the Welsh heart atack miracle is not as well known as those in Helena and Scotland, it is based on the same shoddy and dishonourable techniques. The Scottish study was awful, flawed in so many ways it is hard to know where to begin. This, however, is just pathetic. Even measured by the standards of previous efforts (of which, in my view, Helena is still the worst) this is a shameful episode. It provides further evidence that sections of the anti-smoking movement believe they can now say anything and get away with it. They display contempt towards the media and insult the intelligence of the public. If inflation figures or crime statistics were misrepresented in this way, people would lose their jobs. But since the subject is the smoking ban, it is doubtful whether more than a handful of people will ever even be aware that a deception has taken place.

Christopher J. Snowdon is the author of Velvet Glove, Iron Fist: A History of Anti-Smoking

If you have any comments or corrections please send them to author@velvetgloveironfist.com

With thanks to Brian Bond for his technical expertise and to Health Solutions Wales for providing the data.

* On his website Chris Franks says he used total 'heart-related emergency admissions' for his analysis. Most of the media coverage explicitly used the term 'heart attacks' and it is heart attack data (acute myocardial infarction) that I have examined here. Total 'heart-related admissions' would cover all sorts of surgical procedures such as CABG (Coronary Artery Bypass Grafts) and PCTA (Percutaneous Transluminal Angioplasty). A reduction in the number of operations performed in hospitals is neither welcome nor relevant to this topic. The numbers Franks quotes are so high that they can only represent ALL heart-related admissions, and yet he goes on to talk solely about heart attacks. This suggests that he does not understand the difference between a heart attack and a heart-related ailment. Be that as it may, the Helena hypothesis confined itself to heart attacks and I have done the same here.