Tobacco was unfortunately very much in the news again recently with the BMA launching a campaign to ban smoking in cars probably as a prelude to what will then seem a more reasonable move to get it banned when children are present. I want to focus here not on the ban but on the methods being used by its advocates.
On Radio 4s Today programme Vivienne Nathanson of the BMA was questioned about her evidence:
Nathanson: Well, the evidence is, in fact, that the levels of toxins that can build up in a car do reach 23 times the levels in a smoky bar…
Interviewer: And that is—sorry to interrupt you—but that is peer-reviewed?
Nathanson: Yes, absolutely.
Interviewer: Everyone in the scientific community accepts that it’s true?
The BMA has since issued a major correction and apology with the explanation that the mistake was caused by human error. These things happen but as Head of Science and Ethics, Nathanson has a duty to check. I find her glib assertions regarding peer review and scientific consensus indefensible. It is hard to see how they could be made in error.
The same day over on Radio 5, a phone in caller queried the general evidence base for passive smoke harm and Deborah Arnott of ASH countered by emphasizing claims that heart attacks have been reduced by smoking bans:
“There’s very good evidence supported by the BMA, the Royal College of Physicians, the World Health Organisation, the Standing Committee on Tobacco & Health which reported to the Department of Health. The Coalition Government very recently conducted a review of smokefree legislation, and what we’ve seen is a significant decline in heart attacks following the implementation of the legislation. The evidence is incontrovertible.”
In response to criticism that a fall in heart attacks in England post the 2006 Health Act was part of an existing trend, she said:
“Yes, but the decline is greater than trend. And that’s in a peer-reviewed article published in a very reputable journal, and it’s been found not just in England, Scotland, but everywhere that smokefree legislation in public places has been brought in.”
The statistics on heart attack hospital admissions in the UK are freely available to all on line. Here is a link to the NHS data for Scotland.
Using the measure preferred by the tobacco control industry and the hospital admissions data from NHS statisticians (table AC5) we can calculate emergency admissions for heart attacks as follows:
12 months pre ban: 7905
0-12 months post ban: 7250 (-8.29%)
12-24 months post ban: 8913 (+12.75%)
24-36 months post ban: 7707 (-2.50%)
() = % change from pre ban baseline
The word Arnott used was incontrovertible, the claim is a 17% reduction and the intervention was presented by some supporters as certain to have an immediate major impact on public health. Taking all that into account we should see a consistent, trend independent effect in the public record but whichever table or measure we use it is impossible to objectively claim that such an effect exists.
I value sober analysis of NHS statistics more than I do articles authored by tobacco control activists, peer reviewed or otherwise. I am not alone in this view and it is unlikely that Arnott is unaware of the serious credibility issues facing all the studies that support her claim or indeed of the existence of other work that contradicts it. The evidence is very far from incontrovertible.
Arnott is a skilled professional propagandist who is all too aware that, whatever the actual truth; provided that she sounds convincing her version will be believed by enough otherwise ill-informed people to achieve her objective. An article here gives some insight into her personality and the nature of her campaign.
Arnott is correct in saying that there has been a peer reviewed publication in the BMJ that supports her claim. It claims a 2.4% decrease in heart attacks due to the ban but fails to adequately explain how the result was arrived at. This approach is sadly increasingly common in medical journals. Bearing in mind Arnott’s comment on trends, it is notable that, according to the NHS, the overall decline in heart attacks for the 12 months post ban was 4.26% compared to a 5.19% decline 2 years before the ban. The post ban decline was in fact neither large nor significant. The BMJ paper was produced by a team led by Anna Gilmore.
Arnott is also correct in saying that there has been a review of the 2006 Health Act. It was written by Linda Bauld.
The authorship of both these papers throws up some very searching questions about ethics at the Department of Health. Surely in a society that allegedly values honesty and transparency we have a right to expect government to review policy and measure its efficacy using the most qualified, objective and unimpeachable resources available. Why then did the DH directly commission two not especially qualified people both of whom have a notable history of anti-tobacco activism? The results were hardly likely to be seen as either credible or objective.
ASH is largely funded by the Department of Health and appears to be in firm control of the government’s agenda.
How can we possibly have a free and liberal society if we allow the truth to become the property of pressure groups directly funded by government departments? The principles at stake here radiate way beyond tobacco. We urgently need to reform the charity sector and I would suggest also the Department of Health.
Written by Chris Oakley.