The 1997 American Diabetes Association and 1999 World Health Organization criteria for hyperglycemia in the diagnosis and prediction of diabetes

OBJECTIVE - The 1997 American Diabetes Association (ADA) and the 1985 and 1 999 World Health Organization (WHO) criteria for diabetes and hyperglycemia differ. The appropriateness of these diagnostic criteria in terms of indiv iduals identified as abnormal and their prognosis has been debated. The pur pose of this study is to compare the classifications of people by these cri teria and to compare fasting and postload plasma glucose concentrations in the prediction of diabetes. RESEARCH DESIGN AND METHODS - The frequencies of diabetes by the 3 sets of criteria were compared in 5,023 adult Pima Indians not taking hypoglycemic drugs. Among nondiabetic subjects, Easting plasma glucose (FPG) and 2-h pos tload plasma glucose (2-h PG) concentrations and categories of impaired glu cose regulation or diabetes were evaluated as predictors of diabetes define d by 1999 WHO criteria. RESULTS - The frequency of diabetes was 12.5% by 1997 ADA criteria, 14.6% b y 1985 WHO criteria, and 15.3% by 1999 WHO criteria. The incidence of diabe tes was strongly related to higher FPG and 2-h PG, each of which had very s imilar predictive powers. Impaired glucose tolerance (IGT) was more common than impaired fasting glucose (IFG) (15 vs. 5%), but the 5-year incidence o f diabetes was higher in IFG than IGT (37 vs. 24%). CONCLUSIONS - The prevalence and incidence of diabetes are somewhat lower w ith the ADA criteria than with the 1985 or 1999 WHO criteria. The intermedi ate categories of glycemia differ substantially. IFG defines a smaller numb er of people who are at higher risk of developing diabetes than those with IGT. More people at high risk of diabetes could be identified by using eith er IFG or IGT, as recommended by the 1999 WHO criteria, or by using the FPG concentration alone, but with a lower cutoff value.