alcohol and mortality: don't let this U-shape fool you.

I went to a talk by Dr. Richard Rogers on Alcohol and Mortality hosted by the Kinder Institute for Urban Research this Wednesday. My friend was in town from D.C. so I invited him to listen to this sociology lecture I thought was so very relevant- Oktoberfest! College campuses!

Dr. Rogers began by showing us Risk vs. Exposure graphs. For example, in case of tobacco, there is a strong positive correlation between risk and exposure. Easy to understand, since there is also a causative relationship between tobacco and increased health risks.

Risk vs. Exposure graphs: positive correlation & U-shaped

However, for alcohol there seems to be a U-shaped (or J-shaped) curve between risk and exposure. The implication being that those who drink less or more have a greater mortality than those who drink moderately. Whaaat. The increased risk of those who didn't drink at all were Dr. Rogers' main focus. Why would there be a greater health risk with no drinking? Is there an overall health benefit from drinking these people are missing out on?

Actually, Dr. Rogers found that those who didn't drink at all included many different types of people. Case in point, "sick quitters" were former drinkers who had eventually quit due to health reasons. There were also those who had underlying medical conditions that kept them from drinking. Dr. Rogers noted that it is difficult to "tease out" the confounding factors that make people non-drinkers in the first place. Also, he stated that there are other social factors that are correlated with level of alcohol consumption such as culture, religion, socioeconomic status, education, marital status, and gender. An interesting continuation of the study would be looking at different types of alcohol (red wine and resveratrol!).

So I didn't get my answer to how much should I drink every week. But I found interesting how sociologists approach questions. The Q&A session afterwards focused on the methodology of his study- wording, reliable self-reporting, categories molded into one- rather than the significance and meaning of this finding.

My conclusion from this session is that I shouldn't impulsively choose how much to drink looking at these graphs. Because according to the initial risk vs. exposure graph, I should start drinking, but really, what information is that graph hiding?

Today was a long Friday! I ventured off campus to D'Amico's for a delightful lunch (Italian sausages and chicken romano) & petted a dead bird in a cardboard box, in addition to tracking down UPS and sending a gazillion emails. Anyhow, this weekend is filled with adventures with little ones. Tomorrow morning, find me on the lawns frolicking with smallest M. I actually have a canvas tote bag packed and ready to go, with diapers, fruit snacks and water bottles.