Obesity and Mortality
http://www.100md.com
《新英格兰医药杂志》
To the Editor: It has been suggested that the magnitude of the association between obesity and total mortality is decreasing over time because of improvements in medical care for conditions related to obesity.1 This hypothesis was generated to explain higher observed relative risks of death from any cause associated with obesity in the first National Health and Nutrition Examination Survey (NHANES I) as compared with two later surveys, NHANES II and NHANES III.1 Whether similar trends are detectable in other cohorts that span recent decades is not known.
We examined secular trends in the association between body-mass index (BMI, defined as the weight in kilograms divided by the square of the height in meters) and mortality in the prospective cohort of the Cancer Prevention Study II, begun in 1982 by the American Cancer Society.2 We followed 317,875 men and women who were healthy and were nonsmokers at baseline, in 1982, when their average age was 56 years. During 20 years of follow-up, 69,229 subjects died. We stratified deaths and person-years of follow-up into three periods according to calendar year (1982 to 1991, 1992 to 1997, and 1998 to 2002) and estimated the association between BMI in 1982 and subsequent mortality separately for each calendar period.
Relative-risk estimates increased across the range of overweight (BMI, 25.0 to 29.9) and were substantially higher among persons who were obese (i.e., BMI 30.0) (Table 1). There was no indication that the relative risks associated with overweight and obesity had lessened in more recent calendar periods. In fact, in the earliest period, 1982 to 1991, the relative risks for overweight and obesity were slightly lower, and the relative risks for leanness were slightly higher, than in later calendar periods, possibly because weight loss related to an undiagnosed illness would be most likely to influence the measure of BMI for deaths occurring within 10 years of baseline.
Table 1. Relative Risks of Death from All Causes According to Body-Mass Index, 1982 to 2002.
These secular trends were seen in both men and women and within the following strata of attained age: 30 to 64, 65 to 74, and 75 or more years. Both smoking and preexisting illness cause a lower BMI over a period of decades of the adult lifespan, and both predict increased mortality. To the extent that the problem of reverse causality cannot be entirely eliminated in the database under study, the adverse prospective effect of adiposity on mortality will be underestimated. When the potential for confounding by smoking and preexisting disease is eliminated, overweight and obesity are significant predictors of death from any cause during 20 years of follow-up and as recently as calendar year 2002. In this large cohort, there is no evidence that the magnitude of the association between obesity and mortality is decreasing over time.
Eugenia E. Calle, Ph.D.
Lauren R. Teras, M.P.H.
Michael J. Thun, M.D., M.P.H.
American Cancer Society
Atlanta, GA 30329
jcalle@cancer.org
References
Flegal KM, Graubard BI, Williamson DF, Gail MH. Excess deaths associated with underweight, overweight, and obesity. JAMA 2005;293:1861-1867.
Calle EE, Thun MJ, Petrelli JM, Rodriguez C, Heath CW Jr. Body-mass index and mortality in a prospective cohort of U.S. adults. N Engl J Med 1999;341:1097-1105.
We examined secular trends in the association between body-mass index (BMI, defined as the weight in kilograms divided by the square of the height in meters) and mortality in the prospective cohort of the Cancer Prevention Study II, begun in 1982 by the American Cancer Society.2 We followed 317,875 men and women who were healthy and were nonsmokers at baseline, in 1982, when their average age was 56 years. During 20 years of follow-up, 69,229 subjects died. We stratified deaths and person-years of follow-up into three periods according to calendar year (1982 to 1991, 1992 to 1997, and 1998 to 2002) and estimated the association between BMI in 1982 and subsequent mortality separately for each calendar period.
Relative-risk estimates increased across the range of overweight (BMI, 25.0 to 29.9) and were substantially higher among persons who were obese (i.e., BMI 30.0) (Table 1). There was no indication that the relative risks associated with overweight and obesity had lessened in more recent calendar periods. In fact, in the earliest period, 1982 to 1991, the relative risks for overweight and obesity were slightly lower, and the relative risks for leanness were slightly higher, than in later calendar periods, possibly because weight loss related to an undiagnosed illness would be most likely to influence the measure of BMI for deaths occurring within 10 years of baseline.
Table 1. Relative Risks of Death from All Causes According to Body-Mass Index, 1982 to 2002.
These secular trends were seen in both men and women and within the following strata of attained age: 30 to 64, 65 to 74, and 75 or more years. Both smoking and preexisting illness cause a lower BMI over a period of decades of the adult lifespan, and both predict increased mortality. To the extent that the problem of reverse causality cannot be entirely eliminated in the database under study, the adverse prospective effect of adiposity on mortality will be underestimated. When the potential for confounding by smoking and preexisting disease is eliminated, overweight and obesity are significant predictors of death from any cause during 20 years of follow-up and as recently as calendar year 2002. In this large cohort, there is no evidence that the magnitude of the association between obesity and mortality is decreasing over time.
Eugenia E. Calle, Ph.D.
Lauren R. Teras, M.P.H.
Michael J. Thun, M.D., M.P.H.
American Cancer Society
Atlanta, GA 30329
jcalle@cancer.org
References
Flegal KM, Graubard BI, Williamson DF, Gail MH. Excess deaths associated with underweight, overweight, and obesity. JAMA 2005;293:1861-1867.
Calle EE, Thun MJ, Petrelli JM, Rodriguez C, Heath CW Jr. Body-mass index and mortality in a prospective cohort of U.S. adults. N Engl J Med 1999;341:1097-1105.