The Use of HbA1c for Diagnosis of Type 2 Diabetes in Korea

Lee, Young Sil; Moon, Seong Su

Abstract

BACKGROUND/AIMS
An HbA1c test result > or = 6.5% has recently been recommended as the defining criterion for diabetes by the American Diabetes Association. We compared the detection of diabetes using oral glucose tolerance test (OGTT)-based and HbA1c-based diagnostic criteria. Additionally, we identified the optimal HbA1c threshold for diabetes in Korea.
METHODS
This study was a cross-sectional retrospective design using data from 405 subjects without known diabetes who underwent OGTT and HbA1c tests for the diagnosis of diabetes between January, 2005, and June, 2010, at Dongguk University Gyeongju Hospital, Korea. The OGTT is considered to be the gold standard for diagnosing diabetes. Based on a receiver operating characteristics (ROC) curve, optimal sensitivity and specificity were derived for identifying the HbA1c threshold.
RESULTS
For diabetes, HbA1c at 6.5% gave a sensitivity of 52.9%, specificity of 95.1%, positive predictive value of 92.9%, and a negative predictive value of 62.7%. A total of 47.1% of subjects with newly diagnosed diabetes had HbA1c levels <6.5%. Results were similar in sex- and age-stratified analyses. To identify diabetes based on OGTT, the HbA1c threshold of 6.1% had an area under the curve (AUC) of 0.849 with 77.8% sensitivity and 71.7% specificity.
CONCLUSIONS
In a Korean population, diabetes prevalence was lower using the HbA1c-based diagnostic criteria. An HbA1c threshold of 6.1% optimally identified diabetes by OGTT-based diagnostic criteria in Korea. Further large-scale, population-based studies are needed to evaluate the HbA1c threshold for diagnosing diabetes in Korea.