OBJECTIVES: To estimate how the number of diabetics known as type 2 is modified by applying the American Diabetes Association (ADA) new diagnostic criteria. To calculate the sensitivity, specificity and predictive values of the ADA diagnostic criteria. DESIGN: Transversal descriptive study. SCOPE OF THE STUDY: 15,451 people belonging to two urban health care centers. 1292 individuals were studied by routine random sampling. SUBJECTS OF THE STUDY: General population between 40 and 75 years of age. METHOD: Review of clinical histories, selecting the basal glycemias performed over the last three years, considering the last one in the event that more than one existed, and performing the necessary analysis if no data existed. A new glycemia measurement was carried out as well as a glucose tolerance test for those values > or = 110-139. Age, sex, and prior diagnosis of diabetes were also recorded. MEASUREMENTS AND RESULTS: The mean age was 56 years, 56.1% were females. Normal glycemias (< 110)--830 individuals (86.2%). The prevalence of diabetes was 10.5% when the WHO criteria were applied and 8.7% when ADA criteria were applied. Sensitivity was 39.29%, specificity was 100%, the positive predictive value was 100% and the negative predictive value was 98.5%. 49 individuals presented an altered basal glycemia (5.5%), 17 of whom (34.7%) were diabetic according to the WHO. CONCLUSIONS: The prevalence of diabetes when the WHO criteria are applied is significantly higher than when ADA criteria are applied (p = 0.000). The basal glycemia value of > or = 126 is less sensitive than the glucose tolerance test. No normoglycemic patient according to the ADA would be diabetic according to the WHO; however the ADA and the WHO classify the non-normoglycemics in different groups. The WHO criteria (scrupulously applied) are the better diagnostic method for diabetes in principle and the glucose tolerance test is a test not to be done away with.
OBJECTIVES: To estimate how the number of diabetics known as type 2 is modified by applying the American Diabetes Association (ADA) new diagnostic criteria. To calculate the sensitivity, specificity and predictive values of the ADA diagnostic criteria. DESIGN: Transversal descriptive study. SCOPE OF THE STUDY: 15,451 people belonging to two urban health care centers. 1292 individuals were studied by routine random sampling. SUBJECTS OF THE STUDY: General population between 40 and 75 years of age. METHOD: Review of clinical histories, selecting the basal glycemias performed over the last three years, considering the last one in the event that more than one existed, and performing the necessary analysis if no data existed. A new glycemia measurement was carried out as well as a glucose tolerance test for those values > or = 110-139. Age, sex, and prior diagnosis of diabetes were also recorded. MEASUREMENTS AND RESULTS: The mean age was 56 years, 56.1% were females. Normal glycemias (< 110)--830 individuals (86.2%). The prevalence of diabetes was 10.5% when the WHO criteria were applied and 8.7% when ADA criteria were applied. Sensitivity was 39.29%, specificity was 100%, the positive predictive value was 100% and the negative predictive value was 98.5%. 49 individuals presented an altered basal glycemia (5.5%), 17 of whom (34.7%) were diabetic according to the WHO. CONCLUSIONS: The prevalence of diabetes when the WHO criteria are applied is significantly higher than when ADA criteria are applied (p = 0.000). The basal glycemia value of > or = 126 is less sensitive than the glucose tolerance test. No normoglycemic patient according to the ADA would be diabetic according to the WHO; however the ADA and the WHO classify the non-normoglycemics in different groups. The WHO criteria (scrupulously applied) are the better diagnostic method for diabetes in principle and the glucose tolerance test is a test not to be done away with.