Thursday, June 25, 2009

A Few Extra Pounds Might Bring Extra Years

In my face-to-face class we read a chapter about obesity in the book The Economics of Public Issues. But the article titled A Few Extra Pounds Might Bring Extra Years discusses some research that says that being underweight might be worse than being overweight. Here is an exerpt from the article:

"Compared to normal-weight people, those who were underweight were 70 percent more likely to die and those who were extremely obese were 36 percent more likely to die..."

But "that doesn't mean that people in the normal weight range should try to put on a few pounds," according to one of the researchers. Being obese means your body mass index (BMI) was 30 and above. Here is a Body Mass Index Chart based on height and weight


Christopher B. Pugh said...

This link and a contradictory one found here stating being overweight will cause pancreatic cancer and thus shortening your lifespan were both posted in succession on the website just the other day.

I sometimes wonder how serious the medical community is when they release such findings based on small evidence which can be so greatly influenced by confounding variables.

Cyril Morong said...

Thanks for commenting again. The study about obesity followed 11,000 people over 12 years. That does not seem small. But you are right about studies. They need to apply ceteris paribus as much as possible.

Cyril Morong said...

This long New York Times article from 2007, if I recall correctly, discusses the issue with attention to confounding variables. I think it shows that studies don't always mean what we think they mean.

Do We Really Know What Makes Us Healthy?

Christopher B. Pugh said...

Just to remark, although 11,000 people may seem like a large number, when factoring in lifespans of people, 12 years is not a viable timespan to submit conducive evidence one way or another in my expert* opinion.

Furthermore, by limiting the data gathered of this study to only one small area of the world, ie. Canada is really a prohibitive effort in the same way I may state a conclusion that all birds are brightly colored because I spent 4 years living in Panama and since the climate is a tropical rain forest, indeed most all birds there are brightly colored. On a worldwide scale, these results are skewed.

I believe to improve the efficacy of the tests, statistical analysis based on a much broader population is required.

The NY Times article (book? Haha i'm on page 2 of 9 now) hit on an example of the flaw of correlation vs. causation with the noticed effects of individuals taking estrogen supplements having reduced heart attack rates, which was later in a study turned upside down, and then again perhaps proven true.

It's because these studies are conducted on small scales which leaves the results indeterminable, that I suppose I'm jaded.

* - I am in no way qualified as an expert.

Cyril Morong said...

That's great that you are reading that article. I hope you find it worthwhile. Maybe you are right about the Canadian study