Editorial: Study on weight and death risk requires careful reading

Published Jan 4, 2013 at 04:00AM

The headline reads “A Few Extra Pounds Won’t Kill You — Really” in The Wall Street Journal. In The New York Times, it’s “Study Suggests Lower Mortality Risk for People Deemed to Be Overweight.”

An excuse to drop the diet plan?

Not even close.

The only way to come to that conclusion is to stop at the headline, plus maybe a few of the top paragraphs. It’s a risk of the fly-by way we do a lot of our reading, and especially ill-suited to medical studies that deal with complex subjects.

In this case, the study in the Journal of the American Medical Association explores the correlation between Body Mass Index (BMI) and the risk of death. The surprise finding is that being a little overweight appears to be associated with lower risk. Being substantially overweight, however, is found to raise the risk.

The report reviews 97 studies that involved 3 million people and 270,000 deaths, and finds similar results across ages and continents. It uses BMI, a measure that considers height and weight.

The results are the latest confirmation of what researchers call the “obesity paradox,” evidence that people with serious illnesses such as heart disease and diabetes sometimes live longer if they are a little heavier. After the provocative headlines, the stories go on to detail myriad reasons why a simple conclusion is unwarranted, including:

• BMI is an inexact measure of health that doesn’t distinguish between fat and muscle. For example, extremely fit athletes can have BMI measurements that identify them as overweight or obese.

• The measurement also doesn’t distinguish where the fat is located on the body. It turns out that not all fat is equally dangerous. Excess fat in the belly can be toxic, but does less damage if located in thighs or butts.

• Victims of traumatic injury may benefit from the extra padding of a little excess fat.

• Extra pounds may give people with serious illness a reserve if they can’t consume enough nourishment.

• Severely ill people often lose weight before they die.

• Overweight people with indicators of poor health such as high blood pressure, cholesterol and blood sugar are at greater risk than overweight people without those indicators.

• Overweight people may do better because they are getting medical treatment because of high cholesterol or diabetes.

• The study shows a correlation between weight and risk of death, but it doesn’t demonstrate causality.

It’s become a cliche that research repeatedly contradicts itself; first something is good for you and then it’s bad for you, or vice versa. But often those conclusions come from an overly simplistic reading of the studies. Drawing conclusions about appropriate actions from any single or group of studies ignores the limitations of studies and the complexity of the human body.

One thing’s for certain: Staying physically active and eating a good diet is the best path to a healthy life.