The Cockcroft-Gault and the Modification of Diet in Renal Disease (MDRD) Study equations have not been validated in Asian Americans with varying degrees of glucose tolerance. We compared both equations to 24-h urinary creatinine clearance, the latter as a standard measurement of glomerular filtration rate (GFR), in 398 Japanese Americans (62.1+/-5.8 years, mean+/-S.D.) who had normal glucose tolerance (NGT) (n=138), impaired glucose tolerance (IGT) (n=136) and diabetes (n=124). Although both the Cockcroft-Gault (r=0.65, P<0.001) and the MDRD (r=0.74, P<0.001) equations correlated well with creatinine clearance, the latter was significantly superior (P=0.013 between r values). Receiver operating characteristic (ROC) curve analysis showed that the area under the curve (AUC) for the MDRD equation was significantly greater than for the Cockcroft-Gault equation (AUC 0.86 versus 0.80, P=0.015) in classifying subjects as having mildly reduced GFR (<90ml/min per 1.73m(2)). However, both equations overestimated the number of individuals with decreased GFR. We conclude therefore that while the MDRD equation more accurately identifies Asians who are in the early stages of kidney disease, as for other groups, a correction term appears necessary in order to reduce the number of Asian subjects being falsely diagnosed with CKD.