Moderate drinking

may not actually be good for you.

It has generally been accepted that moderate drinking was good for you, based on the fact that epidemiological studies demonstrated decreased death rates for those who drank some alcohol compared to those who abstained. The benefit is seen largely in decreased rates of cardiovascular disease; this finding has biologic plausability as alcohol consumption raises levels of HDL, the so-called good cholesterol.

The exact definitions of moderate drinking and the minimum amount of alcohol needed to gain this protection haven't been 100% clear. 2-3 drinks/day would be the upper limit of "moderate" and some studies have suggested even a drink or two per week confers the observed benefit, although this is controversial.

A new meta-analysis of these studies suggests most have a serious flaw: they incorrectly group those who have quit drinking alcohol in with those who never partook. Because people quite for reasons including illness, aging and the need to take medicines which interact with alcohol, this may lead to bias. Specifically, buy grouping sick or old patients into the group of abstainers it may incorrectly show a benefit to alcohol consumption.

I don't have access to the new article, which was published in the journal Addiction Research and Treatment, so what I now is from this BMJ new item, which isn't free full text:

The researchers, from the University of California at San Francisco and the University of British Columbia, said that most of the 54 prospective studies were flawed as they included as abstainers people who had reduced or stopped drinking, which people often do because of ageing or ill health. Abstainers thus seemed to be less healthy than light drinkers and had a higher risk of death.

After considering 57 variables the researchers found just seven of the studies to be free of error and to include long term abstainers. Analysis of these studies showed no reduction in mortality among moderate drinkers in comparison with abstainers. Only studies containing the "abstainer error" showed protection against death with moderate drinking

So I guess my wine consumption may not be good for my health after-all and the reticence of doctors to encourage alcohol consumption by their patients may have been well-advised. At least the methodologically sound studies didn't show any harm from moderate drinking.