The study was initiated to evaluate the ability of hyperinsulinemia (as a surrogate measure of insulin resistance) to predict the development in a previously healthy population of three putative outcomes of this abnormality--glucose intolerance, hypertension, and coronary heart disease (CHD). The study involved defining the incidence at which these changes occurred between 1981 and 1993 to 1996 in 647 individuals who were free of any disease when initially studied. The study population consisted of approximately 90% of the subjects evaluated in 1981, divided into quartiles on the basis of the plasma insulin response to a glucose challenge as determined in 1981. The results indicated that the 25% of the population with the highest insulin response in 1981 had significant (P < .001) increases in the incidence of impaired glucose tolerance (IGT) or type 2 diabetes (eightfold), hypertension (twofold), or CHD (threefold). Furthermore, the ability of hyperinsulinemia to predict the three clinical endpoints was independent of differences in age, gender, or body mass index (BMI). Finally, if CHD is considered the clinical endpoint, multiple logistic regression analysis indicates that the values for plasma triglyceride (TG) and mean arterial blood pressure ([MAP] as measured in 1981) also predict the development of CHD. These results indicate that the untoward clinical effects of insulin resistance and/or compensatory hyperinsulinemia, glucose intolerance, hypertension, and CHD clearly can develop in less than 15 years.