Some Extra Heft May Be Helpful, New Study Says

Stop the presses, we have another study which challenges common knowledge! This very carefully done study finds that overweight (defined as BMI 25-30) patients have lower mortality than either “normal” weight or “obese” patients.

The researchers – statisticians and epidemiologists from the National Cancer Institute and the Centers for Disease Control and Prevention – also found that increased risk of death from obesity was seen for the most part in the extremely obese, a group constituting only 8 percent of Americans.

And being very thin, even though the thinness was longstanding and unlikely to stem from disease, caused a slight increase in the risk of death, the researchers said.

The new study, considered by many independent scientists to be the most rigorous yet on the effects of weight, controlled for factors like smoking, age, race and alcohol consumption in a sophisticated analysis derived from a well-known method that has been used to predict cancer risk.

It also used the federal government’s own weight categories, which define fatness and thinness according to a “body mass index” correlating weight to height, regardless of sex. For example, 5-foot-8 people weighing less than 122 pounds are underweight. If they weighed 122 to 164 pounds, their weight would be normal. They would be overweight at 165 to 196, obese at 197 to 229, and extremely obese at 230 or over.

Researchers had a full gamut of responses to the unexpected findings, being reported today in The Journal of the American Medical Association.

Some saw the report as a long-needed reality check on what they consider the nation’s near-hysteria over fat.

“I love it,” said Dr. Steven Blair, president and chief executive of the Cooper Institute, a research and educational organization in Dallas that focuses on preventive medicine.

“There are people who have made up their minds that obesity and overweight are the biggest public health problem that we have to face,” Dr. Blair said. “These numbers show that maybe it’s not that big.”

As in all epidemiologic studies, we must examine these results very carefully, and in context of previous research. However, no one can really dispute the methodology used in this study.

Some statisticians and epidemiologists said that the study’s methods and data were exemplary and that the authors – Dr. Williamson and Dr. Katherine M. Flegal of the disease control centers, and Dr. Barry I. Graubard and Dr. Mitchell H. Gail of the cancer institute – were experienced and highly regarded scientists.

So how do we put these results into perspective? Clearly obesity remains a problem. The challenge of defining obesity remains. We have other data to suggest that waist circumference better predicts bad outcomes than BMI. Perhaps many “overweight” patients have acceptable waist circumference. So go back and read this earlier rant – The value of a tape measure. Possibly we should classify obesity not through BMI but rather circumference. Most “overweight” patients have acceptable waist circumference. Thus calling those patients overweight seems a disservice. Labeling patients overweight without supportive data to show that their weight harms them seems unwise.

Dr. Williamson, who is overweight, said that “if I had a family history – a father who had a heart attack at 52 or a brother who developed diabetes – I would actively lose weight.”

But “if my father died at 94 and my mother at 97 and I had no family history of chronic disease,” he said, “maybe I wouldn’t be as concerned.”

Dr. Barry Glassner, a sociology professor at the University of Southern California, had another perspective.

“The take-home message from this study, it seems to me, is unambiguous,” Dr. Glassner said. “What is officially deemed overweight these days is actually the optimal weight.”

Related posts:

1. Morbid obesity is very undesirable
2. Debating Medpundit again
3. Overweight and obesity linked to heart failure
4. On obesity
5. Can you lose weight through exercise?

Related posts brought to you by Yet Another Related Posts Plugin.