by Gail Marchessault, RD, PHEc, PhD (Cand)
“The number of estimated annual deaths attributed to obesity among US adults is approximately 280,000, based on hazard ratios from all subjects and 325,000 based on hazard ratios from only non-smokers and never-smokers.”
So conclude David Allison and his colleagues in their report, “Annual deaths attributable to obesity in the United States,” in the Oct. 27, 1999 Journal of the American Medical Association (1). This estimate is based on a mammoth compilation of statistics, including data from six US studies, the national distribution of adult body mass indexes (BMIs), and estimates of population size and total deaths from 1991.
They are aiming to correct previous estimates of 300,000 deaths per year in the US, wrongly attributed to obesity. They suggest this figure may have been adapted from either Amler and Eddin’s analysis of precursors of premature death in the US for 1980 attributing 289,502 deaths to “overnutrition,” or from McGinnis and Goege’s estimate of 309,000 to 582,000 deaths associated with unhealthy diet and exercise. They clarify their purpose is to estimate the total societal burden in terms of mortality.
While “attributable risk” has a formal statistical meaning, most people will interpret the title to mean that obesity causes 280,000 deaths. “Attribute” is also defined as “regard as characteristic of” (2), possibly closer to the authors’ intended meaning, since this statistic is based on correlations.
In the article, the estimate is defined with an example: “Of the people who were alive at the beginning of 1991, how many fewer would have died by the end of that year if all of the obese people alive at the beginning of the year had not been obese and if those people had the hazard of those with a BMI just below the cutoff for obesity (e.g. BMI 23–25 kg/m2)?”
The value of the estimate depends on how well controlled the original studies were. The effect of variables that accompany obesity and also affect death rates must be removed to get a true estimate of deaths associated with obesity.
The authors write that the calculations assume (after controlling for age, sex, and smoking) that all excess death in obese people is due to obesity. They acknowledge that it has not been established that if currently obese people were to lose weight or never become obese in the first place, they would have a lower death rate. They suggest studying this by determining the impact of intentional weight loss on mortality.
Oddly, the investigators do not comment on the striking similarity of their estimate of deaths attributable to obesity with the prior estimates of deaths due to “overnutrition,” or unhealthy diet and exercise patterns. How many of the deaths they have attributed to obesity are really due to lifestyle factors? Most of the original studies do not control for these factors, and this is a major limitation of this statistical exercise.
The authors of the report attempt to account for these influences by reporting that according to the Nurses Health Study (one of the studies included in the analysis), controlling for alcohol intake, saturated fat intake, and physical activity gave results that were “virtually identical” to those adjusted for age and smoking alone.
Other studies, however, find strong evidence to the contrary (3,4). After a review of the literature, Roger Hammer concluded that “exercise appears to eradicate the relationship between BMI and morbidity or mortality” (5).
Many see exercise and eating habits as more appropriate concerns than weight.
References
1. Allison DB, Fontaine KR, Manson JE, et al. Annual deaths attributable to obesity in the United States. JAMA 1999; 282:1530–1538.
2. Davis C, Webster T, eds. DK Illustrated Oxford Dictio¬nary. Toronto: Dorling Kindersley & Oxford University Press, 1998:59.
3. Barlow CE, Kohl III HW, Gibbons LW, et al. Physical fit¬ness, mortality, and obesity. Int J Obes 1995; 19:S41–S44.
4. Appel LJ, Moore TJ, Obarzanek E, et al. A clinical trial of the effects of dietary patterns on blood pressure. N Engl J Med 1997; 336:1117–1124.
5. Hammer RL. Health and fitness benefits for the obese person. Healthy Weight J 1999; 13:52–53, 57.
This article is from the Healthy Weight Journal (Health at Every Size Journal) and can be cited as Cindy Byfield, PhD, “Is Latest Death Figure New Information? Obesity Deaths May Relate More to Diet and Activity,” from Health At Every Size 14:3 (May/June 2000).
