DIABETES MELLITUS

"While there is little doubt that obesity is in some way associated with both the prevalence and subsequent incidence of non-insulin-dependent diabetes, its exact role and its interaction with other factors have remained somewhat elusive " (1, p. 982). Epidemiologic studies of obesity and di-abetes, one of the oldest known chronic condition associations in medicine, ante-dated studies of obesity and cancer or heart disease. Joslin (2) was among the first to draw attention to the association between obesity and diabetes in a manner that would be acceptable to modern epidemiol-ogists. In 1921, he described a study of weight and height in 1,000 diabetics com-pared by age with insured lives. He under-stood the importance of using consecutive cases and large samples, of actually mea-suring both height and weight, of giving some variation around the "normal stan-dard " of the life insurance tables, of consid-ering the degree of leanness as well as over-weight, and the importance of the basic concept of risk ratios. Describing his data, Joslin wrote, "The table as a whole shows that of the 1,000 diabetics considered, the maximum weights of only 10 per cent were below the standard weight zone, while 15 per cent came in that zone and 75 per cent above it " (2, p. 81)