Background: To assess the usefulness of erythrocyte glycated haemoglobin (HbA 1C ) as a screening tool to identify those subjects with impaired fasting glycaemia (IFG) who do not have impaired glucose tolerance (IGT) or diabetes mellitus (DM) on a 75 g oral glucose tolerance test (OGTT). Design and methods: All subjects undergoing an OGTT had HbA 1C measured at baseline. Receiver operator characteristics analysis was used to identify optimal HbA 1C cut-off values for diagnosing and excluding IGT and DM. Results: We studied 140 subjects (69 women) with IFG (fasting capillary plasma glucose between 6.1 -6.9 mmol/L). Using World Health Organisation criteria, 27 had isolated IFG, 56 had IGT and 57 had DM. HbA 1C was higher (P , 0.001) in patients with DM (6.8 + 0.93%) when compared with those with IGT (6.3 + 0.68%) and isolated IFG (6.2 + 0.30%), but HbA 1C was similar in those with IGT and isolated IFG. There was no HbA 1C cut-off value differentiating isolated IFG from IGT or DM. None of the subjects with isolated IFG had HbA 1C concentration of .6.8%, but 76% and 54% subjects with IGT and DM, respectively, had HbA 1C of 6.8%. Conclusions: HbA 1C measurement is of limited value in differentiating isolated IFG, IGT and DM in subjects with IFG. It cannot be used to identify which subjects with IFG do not require an OGTT.