Performance of HbA1c for detecting newly diagnosed diabetes and pre-diabetes in Chinese communities living in Beijing

Aim To determine the performance of glycated haemoglobin (HbA(1c)) as a screening tool for detecting newly diagnosed diabetes (NDM) and pre-diabetes. Methods A diabetes survey was conducted in Beijing among community dwellers who were willing to participate in the survey. Included in the survey were 903 individuals aged 21-79 years without previously diagnosed diabetes and in whom HbA(1c) and other required covariates had been measured. NDM and pre-diabetes (impaired glucose tolerance + impaired fasting glucose) were defined according to the World Health Organization 1999 criteria based on 75-g oral glucose tolerance test. Receiver operating characteristic curve (ROC) was plotted to determine the performance of HbA(1c). Results The prevalence of NDM and pre-diabetes was 11.1% and 22.4%, respectively. At an optimal HbA(1c) cut-off point of >= 6.0%, the test gave a sensitivity of 80.0% and a specificity of 89.8% for diagnosing NDM; at an optimal cut-off point of >= 5.7%, the sensitivity was 59.4% and specificity 73.9% for diagnosing pre-diabetes. Individuals with HbA(1c)>= 6.0% tended to be more obese than those with HbA(1c) < 6.0%, but blood pressure and lipid profiles did not differ between the two groups. Conclusions HbA(1c) as a single screening test is adequate to detect newly diagnosed diabetes but is not able to identify pre-diabetes in this obese Chinese population.