More on those alcohol guidelines

I was going to write about the modelling conducted by Sheffield University's alcohol squad to create the new drinking guidelines, but I am relieved to find that the Stats Guy has got there first. Most of the odd aspects of the research that struck me have also struck him and his post is essential reading for anyone interested in the subject. Be sure to have a look.

As with their efforts to model the effect of minimum pricing, the Sheffield report is superficially detailed but breaks down when you try to find exactly what calculations have been made and what studies have been used as the empirical evidence. It therefore requires more trust in the authors' judgement than I am prepared to invest.

Minimum pricing is an untested policy so theoretical modelling is the only option. By contrast, the link between mortality and alcohol consumption has an extensive empirical literature. As the Stats Guy says, it doesn't require modelling at all. It requires a literature review.

It seems to me that the important question here is how does your risk of premature death depend on your alcohol consumption. That, at any rate, is what was modelled.

But there is no need to model it: we actually have empirical data. A systematic review of 34 prospective studies by Di Castelnuovo et al published in 2006 looked at the relationship between alcohol consumption and mortality. This is what it found (the lines on either side of the male and female lines are 99% confidence intervals).

This shows that the level of alcohol consumption associated with no increased mortality risk compared with non-drinkers is about 25 g/day for women and 40 g/day for men. A standard UK unit is 8 g of alcohol, so that converts to about 22 units per week for women and 35 units per week for men: not entirely dissimilar to the previous guidelines.

Quite so. It should also be noted that the protective effect is stronger for men (ie. they can drink more than women before assuming the same risk as a teetotaller). This makes a mockery of the idea that the only health benefits from drinking apply to post-menopausal women. The evidence simply does not support that.

One thing the Stats Guy doesn't mention, though it strikes me as peculiar, is that the Sheffield research focuses entirely on alcohol-related mortality and alcohol-related hospitalisations. It is a trivial tautology to say that drinkers are at more risk of drink-related problems than non-drinkers, but what we really need to know is what effect alcohol consumption has on overall mortality. The Sheffield report doesn't tell us. Fortunately we have a wealth of epidemiological evidence to show that overall mortality risk does not return to the level of a non-drinkers until alcohol consumption gets to around 20 to 40 units per week.

There are some mildly entertaining attempts to pour doubt on the health benefits of alcohol in the text of the Sheffield report which I will return to another day. For now, read the Stats Guy's post (and, if you haven't seen it, my post from Friday).