Is ‘moderate drinking’ really good for the heart?

It’s a widely recognised idea in nutrition that moderate drinking is ‘healthy’. This advice is based on evidence which links moderate drinking with improved health outcomes (notably a reduced risk of death from heart disease) compared to not drinking at all.

However, I remember more than 10 years ago sitting in a presentation given by an American nutritional physician, who made the point that such research is potentially biased by the fact that non-drinkers can be reformed drinkers, and this group may include ex-alcoholics (alcoholism may damage heart health) or individuals who have been advised not to drink or taken it upon themselves not to drink because they have had a heart attack or been diagnosed with heart disease.

Because of this, to make a better judgement of the impact alcohol has on health it would be better to compare the health of moderate drinkers, say, with those who have never drunk alcohol (lifelong teetotallers, not reformed drinkers) or at least individuals who have not drunk alcohol for a very long time.

With this in mind, I was interested to read a recent study that looked at heart disease risk and risk of death in both former drinkers and also long-term abstainers of alcohol [1]. The study was a meta-analysis, and amassed data from 38 epidemiological studies. Risk of heart disease ‘morbidity’ (e.g. a diagnosis of heart disease or heart attack) was the same in these two groups. However, for heart disease ‘mortality’ (essentially, fatal heart attacks), there were significant differences between the two groups.

In men, former drinkers were at a 25 per cent increased risk of heart disease mortality compared to long-term abstainers.

In women, former drinkers were at a 54 per cent increased risk of heart disease mortality compared to long-term abstainers.

The researchers concluded that in future analyses, former drinkers should be excluded from the non-drinker group. This way, the results will not be so biased by the fact that former drinkers tend to be at increased risk of death from heart disease compared to long-term abstainers.

Previous studies will almost certainly been subject to this bias, and will therefore have ended up with results that make moderate drinking appear to be associated with improvements in health outcome that do not exist (or are smaller) in reality.

With New Year’s eve approaching, I suppose an article highlighting the fact that the health ‘benefits’ of drinking alcohol have almost certainly been overstated will not go down so well. But, I’m also not of the mind that everything that passes our lips need be healthy. On the other hand, if we are going to drink alcohol, we may as well do it in a way that may mitigate any negative effects. I wrote a blog post about this recently, which you can find here.


1. Roerecke M, et al. Ischemic Heart Disease Mortality and Morbidity Rates in Former Drinkers: A Meta-Analysis. Am J Epidemiol 14 Dec 2010 [Epub ahead of print]

7 Responses to Is ‘moderate drinking’ really good for the heart?

  1. Chmee 31 December 2010 at 12:17 am #

    ‘But I’m also not of the mind that everything that passes our lips need be healthy.’ For once, a doctor that is not a complete killjoy. Excellent !:). As for me ( with apologies to a certain playwright, though I’m sure he would have agreed ):

    ‘Once more unto the bar dear friends, once more;
    Or close the pub up with our English dead’

    A Happy New Year to you and thanks for some informative and interesting reading.

  2. John Bowman 31 December 2010 at 7:26 pm #

    A UK TV ad paid for by the (supposedly) cash-strapped NHS warned that women who drank more than the recommended weekly allowance of alcohol were at twice the risk of death from cancer of the oesophagus.

    1. The recommended alcohol limit is an arbitrary number and has no basis in science.

    It is therefore misleading to suggest, as the ad did, that staying within this limit would not incur higher risk of Ca oesophagus, and indeed stating that drinking above the limit would.

    2. The risk from death from Ca oesophagus in women
    is extremely small, thus twice the risk is also extremely small.

    So it is a “frighten the children” ad by people who have no idea what they are about, clearly have not enough to do, except trying to control every aspect of peoples’ loves.

    Of course the justification is “to reduce the burden” on the NHS and save money which, whether people take the advice in the ad or not happen.

    But it is a telling attitude towards a service that people pay for via albeit compulsory insurance – that is the I part of NIC – that they are de facto paying for it but should avoid using it.

    A bit like a tennis club telling its members to cut down on the use of the club facilities to save money – recommended number of games per week.

    Perhaps the NHS and the drones within it should pay more attention to doing their work better and providing the service they should, instead of looking for ways to avoid it and blaming its users for its own short-comings.

  3. Ted Smythe 31 December 2010 at 9:59 pm #

    This is more proof that the science of statistics can either be manipulated, or is poorly understood.

    If one selects out a specific possible bias from a mass of data, then one needs to separate out the other sources of bias too, even if pointing in the opposite direction.

    To see a better sample of the pure effects of not drinking, one can look at religious groups like the Mormons, who mostly do not drink. Or, look at big numbers of people such as Muslim countries and check their mortality rates vs countries that traditionally drink.

    Better yet, since correlation is not causation,consider the biochemical effects of supplemental alcohol. It is suggested that this works by neutralizing the methanol (a toxin) that is imbibed as part of our diet.

    We should be able to test for this in the laboratory. Cause and effect, now we are closer to the realm of science.


  4. Reijo Laatikainen 3 January 2011 at 12:08 pm #

    Very important post! As women are increasing their consumption in many countries, it is of interest that alcohol is also one of the rare dietary factors solidly linked to breast cancer. And on top of oesophagus cancer, as was discussed by John Bowman, high alcohol consumption is a risk factor for colon cancer as well.

  5. Sophie 26 January 2011 at 6:12 pm #

    As always said, moderation is the key to a healthy living. Just like being a vegetarian, omitting fish and meat products might predispose us to health risks. So, I believe that totally omitting alcohol works the same. Just drink moderately.


  1. Tweets that mention Is 'moderate drinking' really good for the heart? - - [Hacker News FH] -- - 1 January 2011

    […] This post was mentioned on Twitter by Hacker News YC, Hacker News, News Bloom and others. News Bloom said: Is 'moderate drinking' really good for the heart? – – [Hacker News FH] […]

  2. How Much Can Drink Affect My Weight? « Carl's Get Fit Blog - 30 January 2011

    […] have a negative effect on weight loss, but also has other negative health issues, such as chance of raised heart disease. The good doctor also has some tips for drinking more responsibly, which I don’t always […]

Leave a Reply