Reversion to the mean

The BBC has an article based on an interesting House of Commons report on alcohol consumption… well, more accurately, it’s a very bad report on alcohol consumption with some interesting data.

The data shows that, before the global descent into miserable puritanism around World War I that led to prohibition in the US and draconian licensing rules in the UK, alcohol consumption was around its current level.

It then spiked after the war ended, fell during the Depression, rose slightly during the mid-late 1930s and WWII, fell in the austerity period, and then rose fairly consistently from 1950 onwards – accelerating slightly since 1995 due to increased wine consumption [*]. We’re now at about 9 litres of pure alcohol per head per year, compared to 11 litres in 1900.

The obvious conclusion to draw is that, with miserable busybodies out of the equation, 10ish litres per head is the natural level that Brits want to drink, that this is all well and good, and that the puritans should be deported to America on pain of pain, as we did in the good old days.

The House of Commons report instead draws the conclusion that OMG FFS AAAGH the sky is falling. Particular stupidity lies in:

Ten million adults drink more than the recommended limits and between them knock back 75% of all alcohol consumed in the country. More than two-and-a-half million adults (8% of men and 6% of women) drink above the higher-risk levels – more than double the government’s daily guidelines.

…but we know that the daily and weekly guidelines are based on *nothing at all*. And we know that, on aggregate, the only people who show a greater risk of mortality or morbidity from alcohol than teetotallers are those who drink more than 30 units a week, which is equivalent to 17 litres of pure alcohol a year. So we can crank up our national drinking by another 70% before we need to start worrying about health impacts.

Paraphrasing the report’s conclusion:

It was not inevitable that per capita alcohol consumption was only one third as high in the middle of the 20th Century as its natural level. The decline was fuelled by miserable prohibitionists lying about the data, a draconian licensing regime and massive propaganda budgets. Now that we’ve returned to normal levels, let’s bring back the 1950s. Who needs fun anyway?

[*] which, as it’s mostly consumed in moderation and with meals, mostly doesn’t have negative health effects anyway.

5 thoughts on “Reversion to the mean”

  1. "the only people who show a greater risk of mortality or morbidity from alcohol than teetotallers are those who drink more than 30 units a week"

    Admittedly this is based on a skimmed reading of the source, but doesn't it talk about 21 units being the key threshold rather than 30? E.g.

    "Among regular drinkers mortality from all causes combined increased progressively with amount drunk above 21 units a week. Among British men in middle or older age the consumption of an average of one or two units of alcohol a day is associated with significantly lower all cause mortality than is the consumption of no alcohol, or the consumption of substantial amounts. Above about three units (two American20units) of alcohol a day, progressively greater levels of consumption are associated with progressively higher all cause mortality."

  2. Never mind, have looked at the charts now. Interesting U-shaped relationship – take that, teetotallers! Still, drinking over 21 units a week still seems more dangerous thank drinking more than zero but less than 21, which seems perfectly consistent with what the guidelines say.

    Also, just because the *average* consumption is less than some threshold figure, e.g. your 30 litres a year, doesn't mean there isn't a pretty large number of people drinking above that and causing themselves and the rest of us problems. So the statement 'we can crank up our national drinking by another 70% before we need to start worrying about health impacts' is just wrong.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>