Obesity has been much touted in some quarters as being a leading cause of preventable death in the United States, second only to smoking. About a year ago, a study published by researchers at the Centers for Disease Control and Prevention (CDC) estimated the number of excess deaths attributable to obesity at 400,000, a value later corrected to 365,000. (1) Publication of these figures created quite a stir, and gave impetus to the drive by health professionals and others to encourage Americans to revise their lifestyles -- especially to eat less and move more.
These numbers have now been disputed by results of other CDC research, which substantially ratcheted down the estimate of excess weight-attributable deaths to "only" 25,814. This estimate dropped overweight/obesity from the number 2 to number 7 most-frequent preventable cause of death in the U.S. (2)
Why such a great discrepancy? First, the two groups did not employ the same data. The most recent study used data from three National Health and Nutrition Examination Surveys (NHANES) -- from 1971-75, 1976-80, and 1988-94 -- while the first did not. These are nationally representative samples that involve weighing and measuring participants -- that is, they don't just rely on self-reported data. Second, they used somewhat different statistical methods.
The second group found that while obesity per se (a Body Mass Index or BMI* score above 30) increased the risk of death compared to normal weight individuals (BMI between 18.5 and 25), the great majority of these deaths occurred in people who were excessively obese (BMI > 35), and primarily in people under seventy years old.
Perhaps the most interesting finding was that people who were considered overweight (BMI between 25 and 30) had a decreased rather than an increased risk of death. In other words, the data give the impression that overweight status confers some protection against premature mortality. This decrease partially accounted for the much smaller impact of excess weight than in other studies.
For those who were overweight but not obese, there were about 86,094 fewer deaths in 2000 than were expected (that is, relative to normal weight individuals). For people in the obese category, there were 111, 909 estimated excess deaths. Thus, for the combined category of overweight and obesity, the researchers estimated that there were 25,814 excess deaths.
A probable factor in the unexpected decrease in mortality in those considered overweight, but not obese, might be confounding by body composition -- a possibility noted by the CDC researchers. While BMI is generally used as an index of fatness, it doesn't distinguish between body fat and lean tissue (for a more detailed explanation, see http://www.acsh.org/factsfears/newsID.517/news_detail.asp). People who exercise regularly and have gained additional muscle mass could have a BMI that would put them in the overweight range, but would not be expected to have a higher risk of early death.
So, does this mean that a little extra padding is a good thing? Not necessarily. This study examined only mortality -- excess deaths. There was no evaluation of morbidity, or illnesses associated with obesity. Although overweight and somewhat obese people were less likely to die than previously thought, this doesn't mean they were healthier or that their quality of life was as good or better than that of slimmer people.
While overweight may not increase the risk of dying prematurely, based on this study we can't say that it doesn't have a negative impact on quality of life.
We also know from both this and other studies that the number of obese individuals is increasing most rapidly for the group with a BMI over 35 -- the most extreme group. And for this group, the mortality rate was indeed increased. Thus, as with a number of other factors, the effect of body fat on mortality is dose-dependent: a little more may be helpful, but a lot extra is not.
(1)Mokdad AH, Marks JS, Stroup DF, Gerberding JL. Actual causes of death in the United States, 2000. JAMA 2004;291:1238-1245. [Published correction appeared in JAMA 2005;293:298.]
(2)Flegal KM, Graubard BI, Williamson DF, Gail MH. Excess deaths associated with underweight, overweight and obesity. JAMA 2005;293:1861-1867.
*BMI = weight in pounds X 703 divided by height in inches X height in inches.
Ruth Kava, Ph.D., R.D., is Director of Nutrition at the American Council on Science and Health.