Comparison of the performance of HbA1c and fasting plasma glucose in identifying dysglycaemic status in Chinese high-risk subjects

Abstract

The aim of the present study was to compare the performance of HbA1c and fasting plasma glucose (FPG) in identifying dysglycaemic status among Chinese participants. Fasting plasma glucose and HbA1c were measured in 2318 subjects with at least one risk factor for diabetes but without being previously diagnosed with diabetes. Using HbA1c to diagnose diabetes resulted in the same classification as FPG for 90.5% of the study participants, with 21.0% (n = 487) classified as having diabetes by both FPG and HbA1c and 69.5% (n = 1610) classified as not having diabetes by both FPG and HbA1c. The kappa (κ) coefficient of the FPG criterion with the HbA1c criterion for diabetes was 0.75 (95% confidence interval (CI) 0.72-0.78). The overlap index regarding diabetes diagnosed by FPG or HbA1c was 68.8%. Of 1610 subjects with FPG < 126 mg/dL and HbA1c < 6.5%, 220 (13.7%) had FPG ≥ 100 mg/dL and HbA1c < 5.7%, whereas 277 (17.2%) had FPG < 100 mg/dL and HbA1c ≥ 5.7%. The κ coefficient of the FPG criterion with the HbA1c criterion for prediabetes was 0.30 (95% CI 0.25-0.35). The overlap index between subjects diagnosed as having prediabetes by FPG of 100-125 mg/dL (impaired fasting glucose (IFG)) or HbA1c of 5.7-6.4% (increased HbA1c (IGH)) was 35.9%. The HbA1c criterion demonstrates reasonable concordance with the FPG criterion for diabetes. Hence, HbA1c and FPG can be used for the diagnosis of diabetes. However, the IGH shows limited overlap with IFG for prediabetes. Introduction of the IGH criterion in addition to IFG for the screening of prediabetes could lead to the identification of more people with this condition.