OBJECTIVE - To measure the 12-year incidence (1982-1994) of nontraumatic lower extremity amputations (LEAs) in Nauruans, a population at high risk for NIDDM, and to determine the risk factors for amputation in Nauruans with diabetes. RESEARCH DESIGN AND METHODS - Amputation data were abstracted from operating theater records in Nauru, hospital databases in Australia, and Nauru government records. Baseline characteristics of a cohort 1,564 Nauruans aged ≤20 years examined during a population-based survey in 1982 were used to determine risk factors for first LEAs. RESULTS - Over this 12 year period, 46 first LEAs were performed on people with NIDDM, of whom 30 were members of the 1982 study cohort. The incidence of first LEAs in Nauruas aged ≤25 years with NIDDM was 8.1 per 1,000 person-years in the study cohort and an estimated 7.6 per 1,000 person-years nationally. Amputations were associated significantly with lower BMI lower blood pressure higher fasting plasma glucose (FPG) level, and longer mean duration of diabetes at baseline, but levels of other risk factors, including cigarette smoking, plasma triglycerides, and plasma cholesterol, were also elevated in amputees. There were no amputations among individuals with baseline FPG levels <7.8 mmol/l, irrespective of diabetes duration. FPG baseline diabetes duration, and male sex were independent risk factors for first amputation using the Cox proportional hazards model. There was a decrease in the incidence of amputations after the commencement of a national foot care health education and prevention campaign in June 1992. CONCLUSIONS - The incidence of LEAs in diabetic Nauruans was higher in other populations after adjusting for age and duration. Given the apparent success of the Nauruan footcare program in reducing amputation rates, other populations with high rates of NIDDM and LEAs should consider population-wide prevention strategies.