Interpretation of an abnormal oral glucose tolerance test encountered during multiphasic laboratory screening

162 Healthy male flyers were studied who had repeat glucose tolerance tests at intervals of 6 mth to 2 yr. On the premise that an abnormal oral glucose tolerance test (OGTT) was not reproducible because standard criteria were not strict enough, 11 different criteria were evaluated for diagnosing an abnormal OGTT. The number of abnormal tests varied according to the criteria but, no matter how the criteria were varied, the best reproducibility of any abnormal test on repeat testing was 50%. Factors affecting the reproducibility of an abnormal test are discussed. It was concluded that a single, standard 2 hr oral glucose tolerance test should be considered only as a screening test for diabetes mellitus, and that a final diagnosis of chemical diabetes mellitus should be made only if the carbohydrate intolerance consistently worsens or if insulin curves show hypofunction of the Islets of Langerhans. The recommendations for followup studies of a patient with an initial abnormal OGTT are outlined. This study illustrates the importance of basic definitions of diagnostic criteria for diseases which have a long term effect on the health of the flying population.