Context: Although glycated hemoglobin (HbA1c) has recently been incorporated as a diagnostic test by the American Diabetes Association, its validity needs to be established in Asian Indians in a community setting.
Objective: The objective of the study was to assess the validity of HbA1c as a screening and diagnostic test in individuals with newly detected diabetes mellitus.
Design and setting: Community based randomized cross sectional study in urban Chandigarh, a city in north India, from April 2008 to August 2009.
Subjects: Subjects included 1972 subjects aged 20 yr or older.
Intervention: Intervention included an oral glucose tolerance test and glycated hemoglobin in all the subjects.
Main outcome measures: Utility of HbA1c as a diagnostic method in newly detected diabetes mellitus subjects was evaluated.
Results: Using World Health Organization criteria for diagnosis of diabetes mellitus, 134 (6.7%) had newly detected diabetes mellitus, 192 (9.7%) known diabetes mellitus, 329 (16.6%) prediabetes, and 1317 (69.4%) were normal of 1972 people screened. Using only the ADA criteria, 38% people were underdiagnosed. An HbA1c level of 6.1% had an optimal sensitivity and specificity of 81% for diagnosing diabetes. A HbA1c level of 6.5% (+/-2 SD) and 7% (+/-2.7 SD) had sensitivity and specificity of 65 and 88% and 42 and 92%, respectively, with corresponding positive predictive value and negative predictive value of 75.2 and 96.5% and 90.4 and 94.4%, respectively, for diagnosis of newly detected diabetes mellitus.
Conclusion: A HbA1c cut point of 6.1% has an optimal sensitivity and specificity of 81% and can be used as a screening test, and a cut point of 6.5% has optimal specificity of 88% for diagnosis of diabetes.