The politics of drinking
            August 29th, 2018

Risking the credibility of safe drinking messages

It must be August. Last week the mainstream media was drunk with anti-alcohol headlines after chugging back countless units of public health research. It felt like a concerted assault. Perhaps it was.

The post-prohibitionist Institute of Alcohol Studies, in collaboration with Sheffield Alcohol Research Group (SARG), continued its ongoing campaign to remove from matters of alcohol policy those who make, distribute and sell the stuff, claiming that the drinks industry had a vested interest, to the tune of £13bn, in keeping people drinking above the recommended guidelines.

Others have exposed the flaws in this research. Here’s Rupert Millar at The Drinks Business, for instance.

That £13bn figure is inflated not only by being based on revenue rather than profit but by deploying the new alcohol guidelines that recommend consuming no more than 14 units a week for men and women alike.

It’s a good example, as I’ve argued elsewhere, that the guidelines are, in the words of one insider, more useful to “shift public discourse on alcohol and the policies that can reduce our consumption” than directly affect people’s drinking.

The guidelines question itself was taken up last week by another, very serious, paper that got hopelessly lost among the more sensationalist soundbites.

It, too, comes from SARG which, between working on commissions from other bodies, comes up with some really interesting stuff in the tea-break.

While it played a key role in providing evidence to the Chief Medical Officer in setting the latest guidelines, it was shut out of the final decision-making and, if I’m reading between the lines correctly, was not too happy at the way it turned out.

The experience seems to have driven it to a systematic critique of the business of recommending thresholds for drinking. For example, SARG argues that the framework for developing guidelines should be set out in advance and the process must be transparent.

Such transparency might avoid embarrassments like Christopher Snowdon’s allegations of “massaged evidence” and behind-the-scenes dealing that reduced the threshold for men to 14 units.

But SARG’s analysis cuts into deeper and more difficult areas. To attempt a summary of some complex statistical epidemiology, the guidelines were chosen on the basis that they represent an ‘acceptable’ 1% lifetime risk of dying from something alcohol-related – roughly the same risk as driving a car every day.

Problem is, to anyone who enjoys a drink that sounds like a bargain. They might even feel a 2% or 3% risk is worth it. (Fortunately, people don’t make those kind of calculations, since the risk curve actually gets steeper the more you drink.)

Trouble is, as SARG points out, the guidelines as they are currently construed don’t readily translate into an effective individual health message. On this occasion, a lay epidemiology that tells you some people are able to drink considerably more than others without ill-effect – and that others shouldn’t drink at all - is also scientifically correct. The 14 units is an average low-risk amount across the population, yet people drink as individuals rather than collectively (unless they’re in a round, I suppose).

It raises the spectre of some future dystopia in which people swap alcohol units in the same way that states trade carbon credits.

The simple solution might be to simply tell people to stop drinking altogether, and that seems to be the ploy of another piece of research published to great attention in the Lancet last week.

Most of the media picked up the angle urged by the accompanying editorial comment, headed ‘No level of alcohol consumption improves health’, which explicitly reinforces the UK chief medical officer’s view, as she forces another glass of wine to her lips, that there is no safe amount you can drink.

The Lancet’s assertion is based on research that says while 914 non-drinkers out of 100,000 suffer an alcohol-related health problem or injury, the number leaps to 918 for those having one drink a day.

The totally miniscule nature of the increased risk from drinking is well nailed by the statistician David Spiegelhalter. It seems to me that any rational actor would conclude that if even non-drinkers face a 1% risk of alcohol-related harm you might as well have a drink.

It’s important to note that, judging by the reaction on social media, the abstention message is going too far for many in the public health field. See, for example, this Twitter thread from SARG’s John Holmes and James Nicholls’ blog post for Alcohol Concern.

“The purpose of health guidance should not be to create unnecessary anxiety,” says the latter, yet there are some out there for whom the more extreme the message you can squeeze from a piece of alcohol research, the better. 

All this is not just winding up the drinks industry. It’s making it more difficult for those who genuinely want to address the problems of those who drink too much. The credibility of public health is at stake.