American Diabetes Association diabetes diagnostic criteria, advancing age, and cardiovascular disease risk profiles - Results from the Third National Health and Nutrition Examination Survey

OBJECTIVE - To evaluate age-specific effects on diabetes prevalence estimates resulting from the American Diabetes Association (ADA) recommendation against use of the oral glucose tolerance rest (OGTT), we contrasted the prevalence of two mutually exclusive groups: undiagnosed diabetes according to ADA criteria (no report of diabetes and fasting glucose [FG] greater than or equal to 126 mg/dl) and isolated postchallenge hyperglycemia (IPH) (FG <126 mg/dl and OGTT greater than or equal to 200 mg/dl), a group designated to have diabetes by World Health Organization (WHO) criteria bur not ADA criteria. RESEARCH DESIGN AND METHODS - The weighted age-specific I arius of undiagnosed diabetes:IPH were calculated for 2,844 subjects aged 40-74 years without reported diabetes who had both FG and OGTT. A ratio >1.0 indicated that the proportion of undiagnosed diabetes was greater than that of IPH. Mean levels of HbA(1c) and cardiovascular disease (CVD) risk factors were contrasted among people with undiagnosed diabetes and IPH and those without either abnormality ("nondiabetic"). RESULTS - Both undiagnosed diabetes and IPH increased with age, but age-specific undiagnosed diabetes:IPH ratios decreased from 5.49 in the 40-44 age-group to 0.77 in the 70-74 age-group. Regression analysis showed a significant (P = 0.006) negative association between age and these ratios. Mean HbA(1c) was 7.1% in the undiagnosed diabetes group and differed significantly from that of the IPH and nondiabetic groups (5.6 and 5.3%, respectively). Individuals with undiagnosed diabetes had less favorable triglycerides, BMI, and HDL cholesterol compared with people with IPH. CONCLUSIONS - Compared with WHO criteria, the ADA criteria underestimate glucose abnormalities more with increasing age. However, compared to those with undiagnosed diabetes, individuals with IPH had a mean HbA(1c) level that is considered in the nondiabetic range, and this group had significantly more favorable levels of several key CVD risk factors. These findings suggest that the ADA criteria, although underestimating the abnormalities of postchallenge hyperglycemia that occur frequently with increasing age, appear to be effective at identifying a group of individuals with both unfavorable CVD risk factor profiles and evidence of long-term exposure to hyperglycemia.