Hemoglobin A1C Criterion for Diabetes Diagnosis Among Hispanic and Non-Hispanic Populations

Abstract

To report on the performance of the recently recommended hemoglobin A(₁c) (A1C) criterion for diabetes diagnosis in comparison with the standard fasting plasma glucose and 2-hour post-glucose challenge (PG) test criteria across racial and ethnic groups. We evaluated local and national survey data from 689 Dominican, 4,862 Hispanic, 4,694 African American, and 6,883 white study subjects. We compared rates of diabetes classification by diagnostic criteria, agreement and disagreement between A1C and PG criteria for diagnosing diabetes, and differences in cardiometabolic risk among the 3 diagnostic groups across racial and ethnic stratifications. The A1C-based diabetes diagnoses were higher among Dominican and African American study subjects (81.6% and 67.0%, respectively), and lower among Hispanic and white subjects (46.0% and 37.9%, respectively). Among those not meeting any PG criterion for diabetes, the A1C criterion identified diabetes in 8.3% of Dominican, 3.5% of African American, 0.9% of Hispanic, and 0.5% of white study subjects. The A1C criterion, however, did not identify diabetes in 64.5% of white, 46.1% of Dominican, 44.0% of African American, and 41.9% of Hispanic subjects who were diagnosed with diabetes by a PG criterion. For single tests, the agreement was greatest between A1C and fasting plasma glucose test criteria among Dominican, Hispanic, and African American study populations-76.9%, 65.6%, and 60.7%, respectively. There was no clear difference in selected cardiometabolic risks between A1C and PG-only diabetes diagnoses across racial and ethnic groups. The A1C criterion yields racial- and ethnic-specific differences in diagnosing diabetes and in test agreements with PG-based criteria. Furthermore, diagnostic differences were observed between the Dominican subgroup and the Hispanic study population, of whom 91.5% were Mexican American.