Applicability of a combination of hemoglobin A(1c) and fasting plasma glucose in population-based prediabetes screening

Abstract

Background: The purpose of this study is to determine: (i) the concordance between a combination of hemoglobin A(1c) (Hb(A1c)) and fasting plasma glucose (FPG) (Hb(A1c) + FPG) and a combination of FPG and 2-h plasma glucose (2hPG) (FPG + 2hPG); and (ii) whether substituting FPG + 2hPG with Hb(A1c) + FPG can enhance the detection of prediabetes in diabetes-free non-Hispanic Whites, non-Hispanic Blacks, and Mexican-Americans adults.

Methods: Data (n = 1376) from the 2007 to 2008 U.S. National Health and Nutrition Examination Surveys were used for this investigation. Prediabetes cut points were determined using 5.7-6.4%, 100-125, and 140-199 mg/dL for Hb(A1c), FPG, and 2hPG, respectively. Concordances between Hb(A1c) and FPG, Hb(A1c) and 2hPG, Hb(A1c) + FPG and FPG + 2hPG in screening for undiagnosed prediabetes were determined using sensitivity, specificity, and positive and negative likelihood ratios.

Results: The overall concordance between Hb(A1c) + FPG and FPG + 2hPG in screening for prediabetes was high, as indicated by a sensitivity of 92.4% (95% CI = 90.5-94.5) and specificity of 84.1% (81.2-87.0). The application of Hb(A1c) + FPG was associated with a higher prevalence of prediabetes compared to FPG + 2hPG. Compared with FPG + 2hPG, screening with Hb(A1c) + FPG was associated with 3.2%, 24.3%, and 4.2% relative increases in the identification of prediabetes in nondiabetic non-Hispanic Whites, non-Hispanic Blacks and Mexican-Americans, respectively.

Conclusions: The enhanced prevalence of prediabetes using Hb(A1c) + FPG compared with FPG + 2hPG calls for the need to redefine at a more basic and practical level how to apply Hb(A1c) in screening for prediabetes. A redefined Hb(A1c) that incorporates FPG, age, race/ethnicity, and body mass index may be a better way to use Hb(A1c) in population-based and clinical settings.