Epidemiological ramifications of diagnosing diabetes with HbA1c levels

Aims: To compare the prevalence of diabetes by history and using the fasting plasma glucose (FPG) criterion alone, the HbA1c criterion alone or either one in those not known to have diabetes. Methods: Analysis of NHANES population >= 20 years old from 1999 through 2010. Results: In those diagnosed by laboratory tests, 86% met the FPG criterion and 53% met the HbA1c criterion. The prevalence of diabetes (history or laboratory test) was significantly increased when the FPG criterion was used compared with the HbA1c criterion in the entire (11.5% vs 10.5%, P = 0.018) and Caucasian (10.6% vs 9.2%, P = 0.022) populations. In contrast, there were no significant differences in the prevalence when only the FPG criterion was used compared with only the HbA1c criterion in Hispanics (12.9% vs 12.1% P = 0.386) and African Americans (14.5 vs 14.3%, P = 0.960). Using history and either criterion in those not known to have it, diabetes increased by 61% in this 12 year period in adults >= 20 years old. Conclusions: Using the FPG rather than the HbAl c criterion to diagnose diabetes in those without a history significantly increased the total prevalence of diabetes in Caucasians but not in African Americans or Hispanics. (C) 2014 Published by Elsevier Inc.