Diabetes is a common disease of the older population of the United States. Estimates of the prevalence of this disease from the National Health Interview Survey indicate that about 2.4% of the U.S. population, approximately 5.5 million Americans, consider themselves to have diabetes. It is estimated that for every known case of diabetes, there is an unrecognized case. In addition to definite diabetes, impaired glucose tolerance, a state of borderline glucose metabolism, exists. Diabetes is a known risk-factor for microvascular (renal, retinal, and neurologic) disease and macrovascular disease (accelerated atherogenesis). The effect of these morbid conditions on American health care expenses is enormous. In an effort to identify risk factors for the development of diabetes, and identify potential areas for intervention, 30 years of longitudinal data from the Baltimore Longitudinal Study of Aging are being examined. We previously reported that fasting glucose values have a graded effect on the rate of development of overt diabetes. The borderline zone of 115-139 mg/dl is artificial. Increased incidence of diabetes can be demonstrated in subjects aged 28-59 yr with fasting glucose levels as low as 103-107 mg/dl. We now have examined the predictive power of the two hour value on a standardized oral glucose tolerance test. Values of 140-199 mg/dl have been labeled as """"""""impaired"""""""" glucose tolerance, the implication being that risks of future development of diabetes and its complications are increased. These values were found to be appropriate for younger adults (28-59 yr) but are set too low for adults aged 60 and over. No cases of diabetes developed in men whose two-hour glucose value was below 157 and no statistically significant increase in development of overt diabetes occurred until levels of 200 or more mg/dl were reached. Thus standards need to age-specific; they are set too low for older adults. Female sex hormone status was evaluated as a potential modifier of glucose tolerance. No effect of the phase of the menstrual cycle was found, but women on oral contraceptive therapy had significantly poorer tolerance than those not on treatment. In contrast, post-menopausal women on estrogen replacement had better tolerance than those older women not on replacement.
