The diagnostic accuracy of HbA1c to identify diabetes and dysglycemia as found in OGTTs is limited. Since opportunistic screening at health visits is cost-effective, and glucose is commonly included in metabolic panels obtained at such visits, we asked if use of glucose followed by HbA1c could improve the accuracy of screening. With the Screening for Impaired Glucose Tolerance (SIGT) study dataset of 1,573 adults without known diabetes, we used Receiver Operating Characteristic (ROC) analysis and OGTTs with ADA diagnostic criteria as a "standard" to evaluate screening; random plasma and capillary glucose (RPG and RCG) were measured on a separate day. Subjects were 58% female and 58% black, with mean age 47.9 year, BMI 30.3 kg/m2 and HbA1c 5.4%. Using OGTTs, 51.5% had normal glucose metabolism, 33.3% prediabetes (PreDM), 4.6% type 2 diabetes (DM), and 37.9% dysglycemia (PreDM or DM). For HbA1c to identify DM by ADA OGTT criteria, the ROC area under the curve (AUC) was 0.818 in all 1573 subjects. However, using an RPG cutoff of 100 mg/dl, the ROC was increased to 0.857 in the 576 with RPG >100 mg/dl but only 0.603 in the 997 with RPG <100 mg/dl (both p<0.001 vs. all subjects). There was similar improvement in identification of DM in the 756 subjects with RCG >100 mg/dl (ROC 0.841, p<0.001), and parallel improvement in identification of dysglycemia (ROC 0.671 for all subjects, vs. 0.740 and 0.741 for those with RPG or RCG >110 mg/dl, respectively, both p<0.001). Results were similar with RCG in a separate dataset (n=1,037). For identification of DM among those with RPG >100 mg/dl in the SIGT dataset, specificity and sensitivity were 79% and 74%, respectively, for HbA1c >5.7%, 84% and 69%, for HbA1c >5.8%, and 90% and 64% for HbA1c >5.9%.
Conclusions: Use of random glucose followed by HbA1c improves the accuracy and efficiency of screening, identifying both individuals who should and should not have an OGTT. Such a strategy might improve recognition of diabetes and prediabetes, permitting initiation of preventive management.
B.T. Legvold: None. L.R. Staimez: None. D. Li: None. L.S. Phillips: Advisory Panel; Self; Janssen Pharmaceuticals, Inc. Research Support; Self; AbbVie Inc., GlaxoSmithKline plc., Kowa Pharmaceutical Europe Co. Ltd., Novartis Pharmaceuticals Corporation, Novo Nordisk Inc., Pfizer Inc. Stock/Shareholder; Self; Diasyst Inc. Other Relationship; Self; Diasyst Inc., Janssen Pharmaceuticals, Inc. M.K. Rhee: None.
U.S. Department of Veterans Affairs; National Institutes of Health