Subcutaneous fatness and mortality.

1 From the Research Section, Tuberculosis Program, Communicable Disease Center, Public Health Service, Department of Health, Education and Welfare, Bethesda, Maryland. "Present address: Department of Epidemiology, Johns Hopkins University, School of Hygiene and Public Health, Baltimore, Maryland 21205. 3 Present address: Division of Occupational Health, Public Health Service, Department of Health, Education and Welfare, Washington, D.C. This study had considerable support from other agencies. Measurements of the photofluorograms and machine tabulations were supported by the Heart Program, Division of Chronic Diseases, Public Health Service, Department of Health, Education and Welfare. Consolidation of detailed codes for causes of death was accomplished with the aid of computer time made available through the Division of Radiological Health, Public Health Service, Department of Health, Education and Welfare. Final analyses and presentation were supported by Public Health Service Research Career Award No. 1-K6-HE-21, 670-01 from the National Heart Institute; a grant from the Washington County (Md.) Tuberculosis Association; and Graduate Training Grant No. CD-1-01-1-T1 from the Bureau of State Services, Public Health Service, Department of Health, Education and Welfare. by life insurance companies (1, 2, 3). Throughout this vast experience, the attribute of overweight was found to be strongly associated with excess mortality. But obesity is not the only cause of increased weight relative to height or body frame, and the correlation between obesity and excess weight is far from perfect, the correlation coefficient being of the order of 0.7 (4, 5). Thus it is clearly possible that fatness may or may not be the component of overweight responsible for its association with excess mortality. The evidence incriminating obesity is largely indirect, and to a considerable extent is based on observations involving change in weight. In the life insurance studies, it was found that overweight persons who achieved a normal weight did not experience excess mortality. And a study of U.S. Army officers indicated that there was an increased risk of disability and death, mainly from cardiovascular-renal diseases of a degenerative type, associated with weight gain during adult life (6). In both situations, it seems likely that change in weight was caused by changes in body fat. Although a more recent study failed to confirm the pattern of excess deaths among overweight persons (7), this observation does not absolve fatness, for the population under study consisted of California longshoremen, whose work may well be suffici-