Thursday, Mar 5, 2009

A Glass of Veneno a Day

A glass of wine a day, according to conventional wisdom, is good for you.

Not so fast. In their "Memo to the Media," the Journal of the National Cancer Institute (JNCI) has some grim news for the light drinker: Million Women Study Shows Even Moderate Alcohol Consumption Associated with Increased Cancer Risk. With a cohort of over a million women, the study harnesses some considerable statistical power; the odds ratio is arguably a modest figure but still impressive: researchers found that "low to moderate alcohol consumption among women is associated with a statistically significant increase in cancer risk and may account for nearly 13 percent of the cancers of the breast, liver, rectum, and upper aero-digestive tract combined." Low to moderate alcohol consumption means ~1 drink per day; the type of alcohol (wine vs. spirits, for example) made no difference in the final outcome.

Any statement like this should, of course, put you on guard. It seems that every week there is another piece of breaking news on diet and health, and what's worse, these supposed scientific findings often contradict one another, leading the public to become confused and disillusioned with health research generally. Most of the blame, in my opinion, goes to overzealous reporters unversed in heavily nuanced scientific wording and statistical language. But the scientists get it wrong too. Public—and personal—health is a complex beast, and epidemiology studies often face a lot of challenges in proper study design and confounding variables.

Last month there was also a an editorial in JNCI that addresses the study, and it is worth reading [1]. The authors are cautious in their assessment and point out some limitations of the study design. But they are clearly impressed, and they agree that the evidence is compelling—especially in relation to more questionable cardiovascular studies that show beneficial effects of alcohol. (They also point out that cancer, not cardiovascular disease, is the leading cause of death among middle aged women.) In the end, they caution that "the only reasonable recommendation we can make is that there is no clear evidence that alcohol has medical benefits."

There will, of course, continue to be people advocating the use of alcohol as a health supplement. As long as there is some health benefit, no matter how marginal, people will continue to choose to hear what they want to hear [2]. (And even today there is a top emailed article in the Times' health section.)

I will say this: I do hope that people keep their head and use some common sense. While people may continue to drink responsibly if they want (I don't think there is anything intrinsically wrong about alcohol consumption; after all, even Jesus Christ drank wine), our society should stop justifying alcohol consumption for health reasons [3]. If you feel the benefits of drinking are worth the risks, then that is your decision to make—but do know that there are long-term medical risks.

Regarding the final verdict from a public health perspective: the jury will likely be out until a critical mass of large scale epidemiology studies address all aspects of alcohol, human health, and disease (in other words, never).

In the meantime, I think we should probably put one bit of conventional wisdom to rest.

[1] By coincidence, the commentary was co-written by my former epidemiology teacher. I took his night class last year (through NIH's FAES), where we spent an hour every week dissecting media and peer-reviwed articles, looking for erroneous reporting and problematic study design.

[2] Just because something helps one part of the body doesn't mean that it's overall a good thing. (Just because tapeworms may help with weight loss, doesn't mean its a good idea to ingest them.)

[3] For members of the LDS Church, this is perhaps a great "told-you-so" moment, as we have abstained from alcohol as a church body since 1921, when President Grant made the Word of Wisdom's recommendations against alcohol mandatory for all members. Lest we get too arrogant, however, yesterday I found an analysis (albeit not without flaws) that ranks Utah as the #1 state of online porn subscriptions per broadband user. Ouch.