OBJECTIVE: Because we previously found that weight loss was associated with increased risk for death in all but very overweight men in a cohort of U.S. adults, we undertook a new analysis to determine whether inadequate control for preexisting illness or cigarette smoking contributed to this association. DESIGN: Cohort study. SETTING: The first National Health and Nutrition Examination Survey (NHANES I, 1971 to 1975) collected information on maximum lifetime weight and measured current weight on a probability sample of U.S. adults. The NHANES I Epidemiologic Follow-up Study determined the vital status of participants through 1987. PARTICIPANTS: Men (n = 2453) and women (n = 2739) who were 45 to 74 years old at the time of the NHANES I examination. RESULTS: The effect of excluding persons who died within the first 5 and first 8 years after baseline was examined to limit the influence of weight loss due to preexisting illness. For women, extension of the exclusionary period weakened the association between weight loss and increased risk for death from noncardiovascular disease. However, excluding death for as much as 8 years after baseline did not affect the strong association between weight loss and increased risk for death from cardiovascular disease among men and women with maximum body mass indexes between 26 and 29 (relative risks of up to 2.1 and 3.6 for men and women, respectively, after excluding deaths in the first 8 years). Results were not substantially altered by limiting the analysis to persons who never smoked. CONCLUSIONS: Preexisting illness may influence the association between weight loss and death principally through deaths from noncardiovascular disease. For some persons, weight loss is associated with an increased risk for death, even after excluding deaths occurring in the first 8 years.
OBJECTIVE: Because we previously found that weight loss was associated with increased risk for death in all but very overweight men in a cohort of U.S. adults, we undertook a new analysis to determine whether inadequate control for preexisting illness or cigarette smoking contributed to this association. DESIGN: Cohort study. SETTING: The first National Health and Nutrition Examination Survey (NHANES I, 1971 to 1975) collected information on maximum lifetime weight and measured current weight on a probability sample of U.S. adults. The NHANES I Epidemiologic Follow-up Study determined the vital status of participants through 1987. PARTICIPANTS: Men (n = 2453) and women (n = 2739) who were 45 to 74 years old at the time of the NHANES I examination. RESULTS: The effect of excluding persons who died within the first 5 and first 8 years after baseline was examined to limit the influence of weight loss due to preexisting illness. For women, extension of the exclusionary period weakened the association between weight loss and increased risk for death from noncardiovascular disease. However, excluding death for as much as 8 years after baseline did not affect the strong association between weight loss and increased risk for death from cardiovascular disease among men and women with maximum body mass indexes between 26 and 29 (relative risks of up to 2.1 and 3.6 for men and women, respectively, after excluding deaths in the first 8 years). Results were not substantially altered by limiting the analysis to persons who never smoked. CONCLUSIONS: Preexisting illness may influence the association between weight loss and death principally through deaths from noncardiovascular disease. For some persons, weight loss is associated with an increased risk for death, even after excluding deaths occurring in the first 8 years.