Abstract
For most of the twentieth century, the measurement of blood glucose after a glucose load was the only generally accepted test to diagnose diabetes. Diagnostic criteria were based on glucose concentrations that best predicted the development of microvascular disease, specifically diabetic retinopathy. The OGTT has declined in utility because of its low reproducibility and because it is so time-consuming. The coefficients of variation can be up to 17% for the 2-h glucose level. This inconsistency can result in over 12% being misclassified by the OGTT.