A black market in booze fearmongering
Panorama has been caught out peddling dodgy alcohol stats, only the latest instance of junk-science moralising by the neo-temperance lobby.
How many deaths will be prevented by the introduction of minimum pricing for alcohol in England? It depends who you ask, it seems.
The Home Office says 900 per year, while the medical journal the Lancet reckons it will be an uncannily specific 3,393 per year. Sir Ian Gilmore of the Royal College of Physicians says it will be nearly 10,000 a year. In Scotland, which has a tenth of the population of England, the government says a 50 pence-per-unit minimum price will save a mere 60 lives a year.
Lacking in internal consistency though they are, all these figures emanate from the same source: the Sheffield Alcohol Research Group, which has been beavering away on this issue since 2008. Last month, the group came up with its most startling figure yet.
Panorama: 300 per cent wrong
In research commissioned by the BBC’s primetime current-affairs programme, Panorama, the Sheffield team announced that 50,000 people in the over-65 age group alone would be saved from the grave in the first decade of a 50 pence minimum-pricing regime.
This statistic became the centrepiece of the latest in a seemingly unending series of temperance-promoting Panorama episodes. This one, broadcast on 10 September, featured Joan Bakewell delivering the apparently shocking news that more pensioners are admitted to hospital with alcohol-related conditions than 16- to 24-year-olds. This should come as no great surprise. Not only is the 65-plus age group much larger than the 16-to-24 age group, but elderly people are much more likely to be admitted to hospital. Doubtless, most of us think of alcohol poisoning and street violence when we hear the term ‘alcohol-related hospital admissions’, but the vast majority of these admissions are for diseases of old age which have been linked to alcohol use by epidemiology. The NHS’s wacky system of ‘attributable fractions’ means that if an 80-year-old attends hospital five times for treatment of stomach cancer, for example, it counts as one alcohol-related admission, because a fifth of stomach cancers are assumed to be alcohol-related.
Even with the definition of alcohol-related harm stretched so broadly, it was astonishing to be told that minimum pricing will prevent 5,000 deaths a year in this single age bracket. Bear in mind that the total number of deaths that are ‘specifically or predominantly related to alcohol consumption’ in England each year is less than 7,000, according to National Health Service statistics. Moreover, the Sheffield team had previously estimated that the number of deaths that might be prevented by minimum pricing in the whole population was between 400 and 2,900. For the policy to suddenly reduce mortality among the oldies by 5,000 required a leap of faith, but no one, least of all the Sheffield researchers, appeared to notice, and the rest of the media seized on the figure unquestioningly, leading to such headlines as ‘Minimum price for alcohol “will save 50,000” pensioners’ (Daily Telegraph) and ‘Minimum alcohol price “could save 5,000 older people’s lives a year”’ (the Guardian).
It took a complaint from a member of the viewing public for the statistic to be reviewed. Sure enough, it has no basis in fact, and this week the Beeb quietly posted a note on its website saying: ‘The School of Health and Related Research at the University of Sheffield has confirmed to Panorama that, unfortunately, due to human error, figures they produced specifically for the programme ‘Old, Drunk and Disorderly?’ broadcast on 10 September 2012 were incorrect. The figures are in fact four to five times lower than those originally given to Panorama. The university emphasised the human error was wholly on their part and has apologised unreservedly to the BBC. The programme has been temporarily removed from iPlayer and is being re-edited to reflect the correct figures.’
The estimate has now been reduced to less than a quarter of the original (11,500 rather than 50,000), making it merely implausible rather than wholly unbelievable. The BBC’s decision to re-edit the film is welcome, but most of the programme’s original viewers will never be any the wiser.
Sliding standards and grinding axes
The details of this ‘human error’ are not known, but the mistake is beyond slapdash. The specific research appears never to have been published, and one might cynically suggest that the researchers believed that Panorama would ask fewer questions than the average peer-reviewer. If so, they were right. It is a truth universally acknowledged that Panorama is not the programme it once was. Its tabloid populism makes The One Show look like The World at War, and it now exists as an early-evening mouthpiece for moralising special interests. Bakewell’s episode about old people drinking (an epidemic!) came a few months after Alastair Campbell presented an episode about middle-aged alcoholism (a hidden epidemic!), which followed on from an episode about youthful binge-drinking (a massive epidemic!). All three programmes culminated with a thinly veiled advertisement for minimum pricing.
Whether the number of hypothetical lives ‘saved’ is 50,000, 11,500, 2,900 or 900 is almost incidental. For Panorama, as for the neo-temperance lobby, the important thing is to be able to wave around big numbers to give the illusion of cold, clinical science to a moral crusade. If there were any honesty in this debate, we would recognise that the intention of minimum pricing is to make it more difficult for the homeless and the young to get hammered on strong cider, but that might be seen as a risky, regressive and discriminatory policy that is pregnant with unintended consequences. So instead, the era of evidence-based policy demands empirical data even when the policy is inherently unpredictable and the data is patently unavailable.
Statistics for decoration
In this search for technocratic answers to political questions, no distinction is made between hard science and social science, nor between verifiable facts and wild speculation. Guesstimates from special-interest groups about what humans might or might not do in certain circumstances are afforded the same respect that would be given to predictions based on the laws of nature. For example, it is one thing to say that Halley’s comet will be visible from Earth again in July 2061, but quite another to claim that rates of obesity will double by 2030 or that a 50 pence minimum price will reduce the number of crimes committed by 45,800 and the number of people unemployed by 25,900 (as the Sheffield team’s crystal ball tells them). All of these statements can be preceded by the magic words ‘Scientists predict…’, but only the one about Halley’s comet can be described as a scientific statement.
It scarcely matters whether the evidence presented stands up to scrutiny, or even the basic test of common sense. This is the age of the disposable statistic, designed to decorate policy, not lead it. Last year, spiked reported on the British Medical Association’s (BMA’s) ‘dodgy dossier’ of lies and half-truths about smoking in cars. As with the Panorama programme, it took an outsider to point out that the BMA’s headline claim – that smoking one cigarette in a moving car with all the windows down created 23 times more smoke than would be found in a smoky bar – was patently absurd and had no basis in science. The BMA eventually retracted the claim, but by then the media had moved on; the retraction went virtually unnoticed.
Perhaps we should treat these magic numbers with the same insouciance as those who conjure them up. The public-health industry quite explicitly sees lobbying for legislation as its core activity and, like all campaign groups, judges its success by media ‘hits’. Whatever halo self-appointed ‘health advocates’ might have in the media’s eyes, these campaigners clearly don’t believe that telling the truth is their main job – rather, they are more interested in securing legislation, and the reward for coming up with an eye-catching statistic is millions of pounds worth of primetime news coverage.
The only downside is that these stat-peddlers might occasionally get rumbled by fact-checking websites such as straightstatistics.org or fullfact.org, but the damage to the health lobbyists from such debunkings is minimal. The Guardian’s resident sceptic, Ben Goldacre, tends to turn a blind eye to bad science committed in the name of ‘public health’. Even if Radio 4’s excellent More or Less gets on the case, the number of people who hear the truth will only be a fraction of the original audience. For the most part, a quiet retraction on a website nobody reads is considered sufficient atonement for the few fearmongers who are caught red-handed.
Minimum pricing, minimal evidence
So what’s 38,500 lives between friends, when the worst that can happen is a three-week-old news story has to be amended? The take-home message for the public will still be that minimum pricing is a life-saving policy, but the fact remains that no one has the foggiest idea what will happen if the EU allows this act of glorified price-fixing to take place (which seems increasingly unlikely). All the Sheffield figures are based on questionable assumptions about the relationship between price and alcohol-related harm, which are only occasionally borne out by real-world evidence.
The Sheffield team’s current estimate is based on the belief that a 50 pence minimum unit price will reduce alcohol consumption by 6.7 per cent and that significant health benefits will ensue. Perhaps they will, but alcohol consumption has already fallen by 20 per cent since 2005, which is a bigger drop than the Sheffield team would expect to see from a wallet-busting 70 pence-per-unit price. According to the team’s methodology, such a decline in consumption should lead to alcohol-related deaths falling by 29 per cent at first, and by a whopping 67 per cent after a few years.
In fact, no such miracle has occurred since alcohol consumption began its unaided descent a few years ago. If you want a realistic idea of how much difference a 50 pence-per-unit price will make to drinking and alcohol-related harm, consider the difference in the health and safety of the nation between 2005 and today. And then halve it. If that doesn’t seem a very impressive outcome, prepare to be underwhelmed and overcharged by minimum pricing.
Christopher Snowdon is the author of The Art of Suppression: Pleasure, Panic and Prohibition Since 1800 (Buy this book from Amazon(UK)).