Thursday, March 11, 2010

Will booze make you skinny?

Roll out the barrel because if you believe the news reports alcohol contains magical negative calories!

Yup a recent study is making waves in the media and blogosphere and the gist of the reporting is that a few alcoholic drinks a day may help control your weight - though of course that's not the whole story.

The study's a big one. It looked at 19,220 American women aged 38.9 years or older who had a baseline normal BMI and followed them for 12.9 years and tracked alcohol consumption and self reported exit weight.

As with any long term prospective study, the authors tried valiantly to control for potentially confounding variables and they adjusted for age, race, baseline BMI, smoking status, non-alcohol energy intake, physical activity, menopausal status, postmenopausal hormone use, multivitamin use, comorbid medical conditions, and macronutrient distribution. They then stratified results into alcohol intake by grams with a 5 level subdivision.

The results?

Firstly it's important to note that statistically all groups of women gained weight. Average weight gain for the women who didn't drink at all was 3.63kg over 12.9 years and for those drinking an average of 30g or more of alcohol daily the average weight gain was 1.55kg. That's a difference of 4.6lbs over nearly 13 years.

So best case scenario is the study has indeed accounted for all variables and the association is causal and if you're a woman, drinking 2 glasses of wine daily will help you not gain roughly a third of a pound extra per year.

Worse case scenario? The study proves just how difficult it is to study nutritional variables and that it's one of those association doesn't prove causality pieces.

The lead author, Dr. Lu Wang, very kindly responded to a few email questions regarding potential confounding variables and both in the paper and in her emails made it very clear the authors did a truly bang up job trying to control for everything they could think of. At the same time Dr. Wang readily admitted that there are,

"an endless list of potential confounders."
Given we're talking about a difference of only 4.6lbs over more than a decade of time, it would therefore certainly be possible there's some subtle difference or differences other than alcohol intake to account for the results.

But does that matter?

Ultimately research on heart disease and moderate drinking in women has already suggested benefit to tempered imbibing and with this study, perhaps there's more. More to the point of this post though, the press' reporting on this story has been abysmal with some articles suggesting that alcohol will cause weight loss, others not noting the small actual absolute differences seen in the study and sadly most of the articles doing a better job of reflecting the media's need to sensationalize study results than to practice good science journalism.

So will booze make you skinny? Nope. But maybe, just maybe it'll make you ever so slightly less fat.

Wang, L., Lee, I., Manson, J., Buring, J., & Sesso, H. (2010). Alcohol Consumption, Weight Gain, and Risk of Becoming Overweight in Middle-aged and Older Women Archives of Internal Medicine, 170 (5), 453-461 DOI: 10.1001/archinternmed.2009.527

Bookmark and Share


  1. Anonymous9:55 am

    I suppose jokes about how drinking enough booze will make others look skinny are a bit too obvious here, huh?

  2. It is very unlikely that this result is causal. For simplicity, assume that there are two underlying personality types. The the moderates and the compulsives. The moderates are able to control their alcohol intake, limiting it to one or two glasses a day. The complusives cannot do this. Their either drink significant amounts or do not drink. Given this underlying personality, it is easy to see how women with the self-control to only drink one or two glasses of wine per day can also control their eating habits. Controlling for income doesn't capture the personality type. I'd guess that the underlying causal effect is self-control rather than alcohol. But hey, I'm all for attributing benefits to alcohol consumption. I just think that this study demosntrates correlation not causation.

  3. Anonymous11:44 am

    Increasing alcohol can certainly cause weight gain, look at guys in their first year of university. There is never a good enough reason to start drinking if you currently do not drink.

  4. Brandon and anonymous, it would seem that the authors of the study controlled for energy intake, so that means it had nothing to do with eating habits (unless we're talking about micronutrients, which I'm not sure they controlled for since I haven't read the full paper). As Yoni described, there are potentially many other confounders, but in this case "self-control" was not one of them.

    That being said, as I get older, I now have another reason to have the occasional glass of wine.

  5. Mavis3:26 pm

    I beleive that the other effects of alcohol such as pain relief, stress relief and the simple perfection of a great Pinot Noir might be contrubuting factors....

  6. I think the stress-relieving elements of drinking may play a role. Women who drink instead of eating more calories in a different form (i.e. eating more ice cream, perhaps not with a perfect awareness of how many calories one is taking in) may be harnessing the body's response to stress in a less weight-inducing way.

    I have observed this phenomenon -- but for me, avoiding alcohol has to do with negative associations with drinking because of growing up with an alcoholic parent. Would it be better for me to drink than eat? From a weight-based perspective? Maybe. From a holistic perspective? No. This study didn't determine overall what was "healthier" -- it could be the impact of the extra 15 pounds wouldn't be worse than the decades of drinking on other aspects of health and well-being.
    My friends who are slim and drink one or two drinks a day, I do tell them that I think it probably contributes to their staying slim. Drinking coffee appears to have some connection with delaying onset of type 2 diabetes. Does that mean that everyone should drink coffee?
    I think this type of reseearch can say, if you can manage one or two drinks a day without it becoming a problem, you don't have to stop, but you also don't have to start. I think people who regularly drink a few cups of coffee in the morning probably have some appetite suppressing going on between breakfast and lunch that is helpful for maintaining weight -- so, as long as it isn't causing other problems, drinking coffee per se isn't bad. That doesn't mean that everyone needs to start drinking coffee if they don't already.