Nephropathy clusters in Pima Indian families with non-insulin-dependent diabetes mellitus (NIDDM), suggesting that susceptibility to nephropathy is distinct from NIDDM per se. The authors compared the family history of end-stage renal disease (ESRD) from 52 African-American patients with NIDDM-induced ESRD (cases) with 45 age-, sex-, and and race-matched non-insulin-dependent diabetics without nephropathy (controls) to assess whether the risk of renal disease was independent from NIDDM in African-Americans as well. Thirty-seven percent (19 of 52) of NIDDM-induced ESRD patients had either a first-, second-, or third-degree relative with ESRD, in contrast to only 7% (3 of 45) of diabetic controls. African-American individuals with NIDDM were at eightfold increased risk for developing subsequent ESRD in the presence of a close relative with ESRD (odds ratio = 8.06; 95% confidence interval, 2.2 to 29.6; P < or = 0.0005). No significant differences were observed in yearly income, years of formal education, total serum cholesterol level, prevalence of smoking, or hypertension between the groups. Diabetic control (assessed by glycosylated hemoglobin and random glucose levels) was suboptimal in nonrenal disease controls, suggesting that hyperglycemia alone fails to cause nephropathy in patients with NIDDM. Family size was unlikely to have influenced the results because diabetic cases had significantly fewer first-degree relatives than did diabetic controls. Familial clustering of ESRD is present in certain African-American families with NIDDM. Differences in family size and degree of diabetic control are unlikely to account for the differences observed between families.(ABSTRACT TRUNCATED AT 250 WORDS)