ENDO 2019—Hemoglobin A1c unreliable for assessing DM prevalence


  • Tara Haelle
  • Conference Reports
Access to the full content of this site is available only to registered healthcare professionals. Access to the full content of this site is available only to registered healthcare professionals.

Takeaway

  • Hemoglobin A1c (HbA1c) missed the majority of diabetes mellitus (DM) cases identified by oral glucose tolerance test (OGTT), particularly among racial/ethnic minorities.

Why this matters

  • HbA1c is the preferred way to assess prevalence of glucose tolerance states, but underestimation of prevalence can affect screening guidelines and policy.

Study design

  • Compared prevalence of glucose tolerance states based on HbA1c and OGTT fasting plasma glucose (FPG) and 2-hour post-challenged plasma glucose (2hPG) in 9000 adults (≥20 years) from the 2005-2014 National Health and Nutrition Examination Survey.

Key results

  • 73.07% of DM cases defined by OGTT (FPG≥126 mg/dL or 2hPG≥200 mg/dL) were misclassified as normal with HbA1c.
    • Includes 69.29% of missed cases among Mexican Americans, 80.26% among other Hispanics, 78.89% among non-Hispanic whites, and 58.65% among non-Hispanic blacks. 
  • Compared with OGTT, HbA1c sensitivity was 26.93% and specificity 93.39%.
    • Sensitivity was 30.71% for Mexican Americans, 19.74% for other Hispanics, 21.11% for non-Hispanic whites, and 41.35% for blacks. 
  • HbA1c for normal glucose tolerance had a sensitivity of 84.91% and a specificity of 43.53%.
    • Only 57.52% of those identified with normal glucose tolerance by HbA1c were likewise identified as normal with OGTT.