Alcohol and Good Health?

I know we've all seen or heard about these studies that say a moderate amount of drinking may be good for your health (prevent heart disease, increase "good" cholesterol, increase mineral density in bones and reduce risk of diabetes). But once again, bioethics rears its ugly, truth demanding head. New reports of conflicts of interest (alcohol companies sponsoring academic research) have been raised in this article in the New York Times. As I learn more and more about conflicts of interest through moderating this forum, I've come to the sad realization that a good amount of research is for sale. I'm not saying that moderate drinking is bad for you. I'm a moderate drinker and I like the studies that say I may be doing something healthy. But I really feel like private support for academic research is getting out of hand.

And thats not the only bioethical problem here. Readers of reports on these studies assume a direct causation between moderate drinking and improved health. These studies are much more complicated than they seem. What if the nondrinkers lead a less healthy lifestyle? What if moderate drinkings means that you also exercise and eat well? Nobody has ever tested the effects of moderate drinking in a clinically controlled study as would be required for a new drug.

Anyway, check out the article and let me know what you think, on either the increase in direct conflicts of interest calling into question the integrity of academic research or how alcohol studies are performed.

Omai

 I read this article this morning ad well.  It struck me that we researchers are in a bit of a quandry from an ethical stand point. 

First off, all of us want to continue doing research and that requires money and lots of it.  So if you are hurting for grants and InBev comes along and offers you money to do an alcohol study you defintely take that money.  Im sure all is well until the results come in and you see no effect or a god forbid a 2-fold effect that is difficult to interpret.  To get more money, you just need to report "no effect", but what does that 2-fold effect mean.  Sweep it under the table as statistcal error and you get more funding and publish your results no harm no foul. 

The real problem comes next.  After 5 tries to publish, you finally re-write the paper in the reviewers image and get it published.  Then bam the NY Times is reporting drinking alocohol is ok or even slightly beneficial. Word of mouth spreads from bar to frat party to nursing home and everyone feels good about knocking back a few.  Media hype is difficult for the researcher to comtemplate or understand.

From an ethical standpoint, I believe it is ok to take the money.  It is not ok to publish mediocre results with 2-fold effects and act like it is Nobel prize winning discovery.  Perhaps, Universities should have a set code of conduct and rules regarding publishing results from these type of studies or at least third party vetting before submission.  Any thoughts?

Rus

I am going to play the devil's advocate on this one and state flat out that there will never be a study that shows with definite proof that drinking is good for everybody's health (or bad for that matter). The reason why I say this is because the effects of alcohol (in one way or the way) are linked to so many other factors that teasing out its contribution is beyond current reasonable capabilities. 

In the end (20 years from now?), when we have taken a decent step towards personalized medicine, we will start talking about how some alcohol can be "good" for certain genotypes and lifestyles, while not "good" for others. I.e. alcohol will stimulate the following aspects of your physiology: a, g, f, r, t, and particularly d. If that is what you are looking for, then go for it. If these conflict with your health goals, don't do it. 

The health benefits of moderate alcohol consumption have long been known. One of the earliest scientific studies on the subject was published in the Journal of the American Medical Association in 1904  (Cabot, R.C. the relation of alcohol to arteriosclerosis, Journal of the American Medical Association, 1904, 43, 774-775. )

A French study found moderate drinkers to have a 75% lower risk for Alzheimer's Disease and an 80% lower risk for senile dementia. (Orogozo, J. M., et al. Wine consumption and dementia in the elderly: a prospective community study in the Bordeaux area. Revue Neurologique, 1997, 153.)

www2.potsdam.edu/hansondj/AlcoholAndHealth.html

a lot of research in this regard is conducted at NIAAA ( National Institute of Alcohol Abuse and Alcoholism ) they have  journal named Alcohol Research & Health which is is NIAAA's quarterly, peer-reviewed scientific journal (formerly Alcohol Health & Research World). and the other is ALCOHOL ALERT www.niaaa.nih.gov/Publications/AlcoholResearch/default.htm

"Perhaps, Universities should have a set code of conduct and rules regarding publishing results from these type of studies or at least third party vetting before submission" - R Bishop

I think this is an interesting idea to help curb blatant conflicts of interest. Unfortunately, the University could be just as financially tied to the positive results of a drug study as a professor since most Universities take a healthy cut off the top of grant dollars. Additionally, many large Universities are overrun with red tape, and this sort of oversight would demand more beurocracy.

An unrelated third party vetting would be interesting. The NIH actually agrees. Here is a link to a proposal from the NIH to increase disclosure of conflicts of interest and an NIH website devoted to these topics.

And check out this third party from the FDA.

Omai

I think that as long as the publication is honest about the number of subjects and the statistical analyses used then it is OK to take the money and publish the work. These days so few people (journalists, editors and ....even scientists) REALLY understand enough about statistics to be able to understand the output of such trials. This is a shame, but the solution is not to stop working or publishing, it is to improve the way maths / statistics is taught in schools.