Tuesday, September 18, 2007

Limits of Epidemiology

It's a long slog, but the New York Times has an excellent article on the limitations of epidemiology studies in advising us on how to live and eat healthful lives.

In nutshell, epidemiology attempts to identify health or sickness in the population through statistical associations. This has a built in flaw in that human behavior and life is so full of variations that is not really possible to isolate a specific behavior to an outcome (with the exception of smoking and lung cancer or not wearing a seat belt and deaths from auto accidents).
For example, in college one of my professors had spent some time as a stock market analyst. He found that if the price of goat futures went up, there was a very good chance that the stock market would also go up. Now, he in no way believed that goats had a magical affect on the stock market yet in the realm of public health we have tried to shape policy and advice in regard to diet and exercise based on a ,comparable to goats and the stock market, flimsy cause and effect association.

Thus, in this year we have had two pillars of eating healthful fall: it was found that antioxidants like vitamins E and C and beta carotene did not prevent heart disease nor does eating a high fiber diet prevent colon cancer. In fact, so entrenched are these hopeful myths that they are still being repeated in non-technical health and life style publications.

The basic lesson to take out of the news story is that at this time there is no firm nor statistically significant association between what you eat and your getting sick or living longer.

Labels: , , ,


Post a Comment

<< Home