Reduced prevalence of diabetes according to 1997 American Diabetes Association criteria

Abstract Aims/hypothesis. We examined the prevalence of diabetes and investigated the characteristics of subjects diagnosed by the American Diabetes Association and the World Health Organization criteria. Methods. A total of 1235 Japanese-Americans living in Hawaii and Los Angeles was studied. Of the subjects 114 were classified as previously diagnosed as having diabetes. A 75-g glucose tolerance test was given to the rest of the subjects. Results. When abnormal glucose tolerance was diagnosed by the American Diabetes Association criteria, it was possible to identify only 40 % of diabetic subjects who had not been previously diagnosed compared with the current World Health Organization criteria based on a glucose tolerance test. In addition, the subjects identified by the American Diabetes Association criteria had higher glucose concentrations and had less insulin secretory capacity and they were in need of intensive treatment for diabetes. On the other hand, the subjects not diagnosed by the American Diabetes Association criteria alone were those whose glucose tolerance would be more likely to improve with lifestyle modification. Conclusion/interpretation. It might be better to use the fasting plasma glucose criterion advocated by the American Diabetes Association in combination with a glucose tolerance test after taking a detailed medical history. To reduce the number of subjects requiring the glucose tolerance test, priority should be given to subjects with impaired fasting glucose (6.1 ≤ fasting plasma glucose < 7.0 mmol/l). [Diabetologia (1999) 42: 1168–1170] Received: 22 March 1999 and in revised form: 26 May 1999