The diagnostic criteria of the US National Diabetes Data Group and the World Health Organization have stimulated a major increase throughout the world in epidemiologic studies on the pathogenesis of non-insulin-dependent diabetes mellitus (NIDDM). They have established that much of NIDDM is undiagnosed, that onset of NIDDM occurs at least 7 y before its diagnosis, and that significant morbidity and premature mortality occur in subjects with undiagnosed diabetes. New studies have shown that rural or traditional-living populations are experiencing a major increase in the burden of NIDDM as they move to urban or nontraditional situations, often with 5- to 10-fold increases in NIDDM prevalence. Epidemiologic studies have documented that major risk factors for NIDDM include increasing age, greater obesity, longer duration of obesity, unfavourable body fat distribution, physical inactivity, and hyperinsulinemia. All these factors interact with unknown genetic factors to produce NIDDM. Studies have shown that genes for diabetes, as yet undetermined, are a necessary cause of NIDDM. Hyperinsulinemia exists in childhood in populations at high risk for NIDDM. Stimulated by obesity, upper body obesity, and physical inactivity, insulin resistance develops, accompanied by impaired glucose tolerance. The pressure of the NIDDM risk factors continues this process of insulin resistance/hyperinsulinemia/hyperglycemia, until glucose toxicity to the beta cell results in inability to secrete sufficient insulin, resulting in decompensated fasting hyperglycemia.