Body Weight and Longevity

To the Editor.— The article by Manson et al1entitled "Body Weight and Longevity" represents a worthwhile attempt to eliminate confounding variables in the examination of the relationship between body weight and longevity. Unfortunately, three additional variables that probably should have been looked at have not been. They are (1) physical activity, (2) social status, and (3) diet. Physical inactivity is positively correlated both with greater body weight and with mortality. Therefore, investigations of the relationship between body weight and longevity should include assessments of physical activity. Unfortunately, this is obviously extremely difficult to measure, but again, the failure to look at this variable weakens the establishment of a possible causal relationship between body weight and increased mortality. The second variable, that of social status, should also be examined since persons in the United States with a lower social status also have higher mortality. Lower social status is also