The relationship between in vivo insulin-stimulated glucose utilization (euglycemic clamp technique) and various estimates of the plasma insulin response to oral glucose was defined in 62 subjects with normal glucose tolerance. Both the incremental insulin increase above fasting (r = 0.61) and the total integrated insulin response (r = 0.65) were highly correlated (P less than 0.001) with in vivo insulin action, and the relationship between total insulin response and insulin action remained significant (r = 0.61, P less than 0.001) when corrected for variations in total glucose response, age, and obesity. In a subset of these subjects (N = 27) we were also able to assess state of habitual physical activity by estimating maximal oxygen consumption during bicycle ergometry. A significant correlation also existed between insulin action and response in these subjects (r = 0.67, P less than 0.001), which remained significant (r = 0.65) when differences in total glucose response, obesity, age, and maximal oxygen consumption were taken into account. These data demonstrate that there is a significant correlation between insulin response and insulin action in normal individuals that can account for approximately one-third of the total variance in insulin action seen in subjects with normal glucose tolerance. Thus, determination of plasma insulin levels after an oral glucose challenge can only provide a qualitative estimate of insulin-stimulated glucose utilization.