Last month saw the publication of a Public Health England (PHE) report on alcohol which promoted various interventions in the market, including minimum pricing, tax rises and plain packaging. Although it purported to be an evidence review, the standard of analysis was worryingly poor. The Department of Health must decide if PHE is to be a serious public health institution or a campaign group. If it is to be the latter, it should be privatised and funded by voluntary donations. Taxpayers can ill afford to finance more one-sided propaganda from government agencies.
2016 was bookended by two rather shoddy pieces of research from the government on alcohol. At the start of the year, the Chief Medical Officer’s long awaited review of the drinking guidelines was hi-jacked by campaigners from the temperance lobby and used a computer model that bore no resemblance to the epidemiological evidence. It is difficult to ignore the suspicion that the whole process was designed to achieve three false but preordained conclusions, that there is ‘no safe level of drinking’, that the previous guidelines were too low, and that the health benefits of moderate drinking are trivial.
At the end of the year came Public Health England’s report and, once again, it was drawn up by people who have a well-documented prejudice against alcohol and an equally well-documented history of campaigning for a Big Government approach to the issue. They include Linda Bauld, Katherine Brown, Petra Meier and Gerard Hastings of the Institute of Alcohol Studies (IAS), a neo-temperance organisation that was set up, and is almost entirely funded, by the Alliance House Foundation whose stated objective is ‘to spread the principles of total abstinence from alcoholic drinks’. Also involved were the co-founders of the Alcohol Health Alliance pressure group, Ian Gilmore and Nick Sheron. Two other lobby groups, Alcohol Concern and Alcohol Action Ireland, were also represented on the expert advisory group. Leaving aside the obvious biases of many of the contributors, it is not clear what expertise they offered. Katherine Brown, for example, has no background in science whatsoever. Nick Sheron and Ian Gilmore are medics with strong views on policy but are not scientists or economists. It is also concerning that the final peer review was conducted by Mark Bellis who was also on the expert advisory group. That, clearly, is not an independent, external review.
The preconceptions of the contributors pervade the whole report. Scepticism about observational epidemiology is understandable but the authors are highly selective about how they apply their scepticism. Carrying on where the CMO’s review left off, the report goes out of its way to cast doubt on the beneficial health effects of moderate drinking but takes a far more credulous view of epidemiological studies which show the slightest risk from drinking. This requires the authors to use double standards. For example, when talking about the benefits of alcohol, they (rightly) draw attention to the well-known tendency of drinkers to under-report their consumption. However, when they discuss the risks of drinking, they make barely a passing mention of this. This is crucial because if people drink much more than they claim, the risks of drinking emerge at higher rates of consumption than epidemiological studies imply. A subtle, but telling, sign of the authors’ bias appears when they refer to the epidemiological studies which show benefits from drinking as mere ‘health surveys’ while referring to the same type of studies as ‘observational studies’ when they find evidence of harm.
The authors’ scepticism re-emerges when they discuss the government’s Responsibility Deal which Bellis, Gilmore and others resigned from several years ago in protest at the government’s refusal to bow to their demands. Despite a Department of Health review finding that the Responsibility Deal led to UK consumers consuming more than a billion fewer units of alcohol, the authors of the PHE report portray the initiative as a failure and claim – without any credible evidence – that (a) the industry would have lowered alcohol content of drinks anyway, and (b) increasing the range of low alcohol drinks can lead to people consuming more alcohol.
Time and time again, the report lies by omission when evidence could be cited. For instance, it mentions France’s near-total ban on advertising without looking at any evaluation of how it has impacted youth drinking rates (spoiler: it hasn’t). It claims that there is a lack of evidence about the effect of tax rises on the black market (there is a wealth of evidence showing a direct link between higher prices and increased illicit activity). And it selectively cites a handful of advocates for tobacco-style regulation of alcohol to imply a general consensus. For example, the view of a single psychology student based in Canada is cited as evidence that ‘expert opinion’ favours plain packaging and graphic health warnings on alcohol.
On several occasions the report makes basic errors which undermine the credibility of Public Health England. It claims, for example, that ‘real-term alcohol prices have decreased’ since 1980. This is untrue. The price of alcohol has increased 23 per cent above inflation since 1980. It further claims that minimum pricing ‘typically affects the high-strength, cheap products’. This is also untrue. Recent evidence from Scotland shows that at least half of all alcohol sold in the country would become more expensive if there was a 50p minimum price (let alone a 60p minimum price, as recommended by PHE).
The report entirely muddles the issue of external costs of drinking. The authors have either not read the literature or do not understand it. They cite a 2001 Cabinet Study report and an obscure report by the National Social Marketing Centre (they do not directly cite the latter so perhaps they really have not read it). With these two estimates as their guide, they state: ‘The economic burden of alcohol use is substantial, with estimates placing the annual cost to be between 1.3% and 2.7% of annual GDP. Few studies report costs on the magnitude of harm to people other than the drinker, so the economic burden of alcohol consumption is generally underestimated.’
This is sheer nonsense. Both studies look at the ‘magnitude of harm to people other than the drinker’. That is their entire purpose. They are studies of negative externalities. Or rather they are supposed to be. In fact, both reports inflate their estimates by including costs that principally affect the drinker, i.e. internal costs. The National Social Marketing Centre even includes money spent by drinkers on alcohol as a negative externality! The problem with both studies is not that they underestimate the external costs of alcohol, but that they greatly exaggerate it.
It would be tedious to go through every error and misrepresentation in the report but mention must be made of the way PHE released it to the press. The media were told: ‘As a nation we are drinking twice as much as we did 40 years ago’. This is totally untrue and PHE later retracted the statement. They replaced this statement on their website with the claim that: ‘Between 1980 and 2008, there was a 42% increase in the sale of alcohol.’ In so doing they replaced a claim that is totally untrue with one that is grossly misleading. It is based on HMRC clearance figures which show 537 million litres of alcohol sold in 2007/08, up from 400 million litres in the early 1980s. I have not been able to verify the latter figure – the PHE report does not cite the HMRC data, instead citing an article by Nick Sheron and Ian Gilmore which, in turn, cites the British Beer and Pub Association (BBPA) and HMRC.
But gross sales are irrelevant. What matters is sales per person. With a rapidly growing population, gross sales are bound to exaggerate consumption rates. HMRC and the BBPA quite properly provide alcohol sales per adult, but PHE chose to ignore them. The BBPA’s Statistical Handbook, which is cited by Gilmore and Sheron, show the following alcohol consumption figures per adult (aged 15+):
1980: 9.4 litres
1990: 9.8 litres
2000: 10.4 litres
2010: 10.1 litres
2013: 9.4 litres
It is quite clear that alcohol consumption is currently at exactly the same level as it was in 1980. There seems to be only one explanation for why PHE would focus on gross sales and use 2008 as the cut-off point: it provided them with the most alarming figure they could find. Naturally this was reported in the press with such headlines as ‘Britons are drinking 42% more than they were 30 years ago’ (Daily Mail). This is untrue, and PHE were complicit in this falsehood. If this was a mistake by PHE, it is such a schoolboy error that it undermines faith in the rest of the report. If it was deliberate, PHE is no better than a pressure group resorting to click-bait propaganda.
This misrepresentation has important implications for the report’s stance on pricing. By portraying 2008 as the peak in consumption, the authors are able to claim that the recent decline in alcohol consumption is the result of the alcohol duty escalator which was introduced in March 2008. In fact, when measured properly by adjusting for population growth, the peak in consumption occurred in 2004 and consumption fell thereafter despite alcohol becoming more affordable between 2004 and 2008. Admitting this fact would mean having to admit that the link between affordability and consumption is not as strong as the authors claim.
In summary, the PHE alcohol review cannot be taken seriously as a review of either science or policy. It makes no real attempt to assess the negative consequences of raising prices or prohibiting advertising and its view of the effectiveness of such policies for improving health is rose-tinted. Time and time again, it ignores real world evidence in favour of hypothetical models created by likeminded campaigners. It contains so many glaring factual errors that it raises serious questions about the competence of its authors, and it is so clearly designed to support anti-market policies that it should be viewed as part of the campaigning literature.