OBJECTIVE - We have previously suggested using the paired values of fasting fasting plasma glucose (FPG) and HbA(1c) to identify potential diabetic subjects. In this article, we followed up on 208 nondiabetic subjects and examined their rates of progression to diabetes. We analyzed their likelihood of becoming diabetic according to their baseline FPG and HbA(1c) concentrations. RESEARCH DESIGN AND METHODS - Between 1988 and 1995, 2,877 Chinese subjects with risk factors for diabetes underwent screening. Of these, 2,250 had FPG <7.8 mmol/l and 2-h plasma glucose (PG) <11.1 mmol/l. Of these 2,250 subjects, 265 were randomly recruited for an annual oral glucose tolerance test (OGTT) until they progressed to develop diabetes. Of those 265 subjects, 57 had baseline FPG greater than or equal to7.0 mmol/l and were excluded from the present analysis. Hence, the progression of glucose tolerance in 208 subjects who were nondiabetic according to the new American Diabetes Association diagnostic criteria (FPG <7.0 mmol/l and 2-h PG <11.1 mmol/l) was examined RESULTS - Of the 208 nondiabetic subjects, 26 (12.5%) were men and 182 (87.5%) were women. After a mean follow-up of 1.60 +/- 1.16 years (range 1-7, median 1), 44 (21.2%) progressed to develop diabetes and 164 (78.8%) remained nondiabetic. Those who were diabetic at the end of the study had a high likelihood ratio (LR) of 9.3 to have baseline FPG greater than or equal to6.1 mmol/l and baseline HbA(1c) greater than or equal to6.1%. This was compared with a low LR of 0.6-1.1 in diabetic subjects who had either FPG <6.1 mmol/l or HbA(1c) <6.1% or both at baseline. The crude rate of progression to diabetes was more than five times higher (44.1 vs. 8.1%) in those whose baseline FPG was greater than or equal to6.1 mmol/l and baseline HbA(1c) was greater than or equal to6.1% compared with those whose baseline FPG was <6.1 mmol/l and baseline HbA(1c) was <6.1%. CONCLUSIONS - For Chinese subjects with risk factors for glucose intolerance, the use of paired FPG and HbA(1c) values helped to identify potential diabetic subjects. Those with an FPG greater than or equal to6.1 mmol/l and HbA(1c) greater than or equal to6.1% had a rate of progression to diabetes more than five times higher than those with an FPG <6.1 mmol/l and an HbA(1c) <6.1% after a mean follow-up of 1.6 years. Those with an FPG greater than or equal to6.1 but <7.0 mmol/l, especially if their HbA(1c) was <greater than or equal to>6.1%, should undergo an OGTT to confirm diabetes. Subjects with an FPG <6.1 mmol/l and/or an HbA(1c) <6.1% should have regular screening using the paired values of FPG and HbA(1c).