RESUMEN
OBJECTIVE: To identify the optimal threshold of HbA1c and to evaluate the predictive performance of HbA1c levels in diagnosing diabetes and prediabetes in a middle-aged and elderly Han Chinese population from northwest China. METHODS: In all, 3354 participants aged ≥ 40 years with no history of diabetes from northwest China were enrolled in the present cross-sectional study. All subjects underwent a 75-g oral glucose tolerance test (OGTT), as well as HbA1c testing. HbA1c thresholds for diagnosing diabetes and prediabetes were identified by the highest sum of sensitivity and specificity of each cut-off point, and the receiver operating characteristic (ROC) curve was used to evaluate the diagnostic accuracy of the HbA1c threshold. RESULTS: The mean (± SD) age of the participants was 57 ± 8 years, and 70.75% were women. Based on results of the OGTT, 1347 (40.16%) subjects had impaired fasting glucose and/or impaired glucose tolerance, and 725 (21.62%) had diabetes. The area under the ROC curve for detecting undiagnosed diabetes and prediabetes by HbA1c levels was 0.810 (95% confidence interval [CI] 0.796-0.823) and 0.732 (95% CI 0.717-0.747), respectively. HbA1c threshold of 6.4% and 6.1% produced the highest sum of sensitivity (60.00% and 61.49%) and specificity (87.33% and 73.24%) for diagnosing diabetes and prediabetes, respectively. CONCLUSION: HbA1c is an effective and convenient method for diagnosing diabetes and prediabetes. HbA1c thresholds of 6.4% and 6.1% may be used as diagnostic criteria for diabetes and prediabetes, respectively, in the Han Chinese population living in northwest China.