Relationship between body weight and mortality in men aged 75 years and older

As our population ages, recommendations about weight control in the elderly will assume increasing importance. But such recommendations are not supported by the literature, in that there is little information about the independent effect of obesity on survival in subjects who are more than 70 years old. We studied body weight, as assessed by Quetelet's index (weight in kilograms divided by the square of the height in meters), in 162 men aged 75 to 98 years (mean age = 81 +/- 5.3). Length of follow-up averaged 28.6 months. Early deaths (those occurring within the first year of the study) were attributed to preexisting morbid conditions. Cox regression was determined for the entire group, and then for the same group with early deaths excluded. Covariates were Quetelet's index, age, race, history of cigarette smoking, glucose intolerance, and hypertension. Two variables, a low Quetelet's index and glucose intolerance, were significantly associated with decreased survival for the entire group. When early deaths were excluded, however, the only variable significantly related to survival was age. These data suggest that the inverse relationship of body weight to mortality in aged men is due to preexisting morbid conditions and that when these morbid conditions are accounted for, body weight is not significantly related to survival. Hence, the relationship between body weight and mortality risk in aged men differs substantially from that in younger adults.