SALT LAKE CITY — Losing weight is invariably on the top 10 list of New Year's resolutions. People want to get healthy and improve their lives. Certainly, getting healthy is a good thing, but shedding pounds may not be the right answer, at least in terms of mortality rates.

A study published this week in the Journal of the American Medical Association shows a slightly lower risk of death from all causes among those who are slightly overweight or even among the slightly obese, according to body mass index.

Researchers from the Centers for Disease Control looked data produced by 97 other studies that related BMI and death risk, most from the U.S. and Europe. That included data for some 2.88 million people and over 270,000 deaths.

They found that those who were overweight, with a BMI between 25 and 30, were 6 percent less likely to die of any cause than those who had a normal BMI of between 18.5 and 25.

Even more interesting was the finding that those who were in the lowest category of obesity, with a BMI of 30 to 35, were also at a 5 percent lower risk of death compared to a person with a normal BMI. Nevertheless, those in the two highest categories of obesity were much more likely to die: Those with a BMI greater than 35 were 29 percent more likely to die.

Related:

Half of Utah will be overweight by 2030, report says

Utah's obesity rate has doubled in the past 20 years and is set to double yet again in the next 20 years if the collective body mass index doesn't shrink.

BMI measures the ratio of weight over height squared. It is used in the medical industry as a simple means of gauging the appropriateness of one's weight. The findings shouldn't be so surprising, given that lead researcher Katherine Flegal has published similar findings in the past, and other studies appear to have confirmed them.

"I think there's a lot of under reporting of this finding ... and so people are sort of repeatedly surprised by it," Flegal told CNN.

Exactly what causes these results is not totally clear yet, but the authors suggest that heavier patients may be getting better medical care and that doctors might screen them for disease more frequently due to their weight. They even suggested that there might be a "cardioprotective" effect to having higher body fat.

Roughly one third of Americans are overweight and another third are considered obese,

Some doctors not associated with the study have been stressing that the results aren't all there is to the story. While there may be all sorts of reasons why being overweight could be associated with decreased mortality, the study does not measure things like quality of life.

"We have recent evidence — from the Lancet's 'Global Burden of Disease' study — that we are living longer, but sicker," Yale University Prevention Research Center Director David Katz told ABC. "It may be that overweight does, indeed, contribute to type 2 diabetes and heart disease, but not to premature death.

Flegal also stressed that BMI can't tell you how healthy you are.

"These are not health categories, these are weight categories," Flegal told ABC.

The BMI itself might be overused as a health indicator, rather than a more limited expression of body fat. An associated editorial published in the JAMA said that other factors like blood pressure and cholesterol and blood glucose were necessary to really determine overall health of overweight people.

Editorialists, Steven Heymsfieldand William Cefalu, also worried that Flegal might have included too many skinny people in the "normal" BMI category because "persons with a BMI between 18.5 and 22 have higher mortality." They suggested that using 22 to 25 might have been a better range, thought Flegal's range is endorsed by the National Heart, Lung and Blood institute.

Related topics