Hyperinsulinemic euglycemic clamp technique can be used to evaluate insulin sensitivity in vivo. However, it is not fit for assessing the insulin sensitivity in large scale epidemiological studies in which some simple index is needed. In this paper we present some additional data to describe characteristics of the insulin action index(IAI)=1/(FPG×FIns). The Pima Indians recruited in this study,from which we suggested the IAI, had a very wide range of fasting plasma levels of glucose (75~306 mg/dl or 4.2~17.1 mmol/L)and insulin (9.7~120 mU/L)concentrations, The insulin-mediated glucose disposal rate (M value) assessed by the clamp technique were closely correlated to a various complex parameters involved to fasting glucose and insulin levels.The correlation coefficient of M value and IAI was significant (r>0.7, P=0.0001) and it was greater than that between M and fasting plasma insulin or the ratio of glucose to insulin and not inferior to the correlation between M and the products of areas of glucose and insulin during OGTT. The quintile distribution of the products of plasma glucose and insulin demonstrated that 90.4% IAI value were within the predicted M vatue quintile or its adjacent quintiles. The reverse of the product of fasting plasma glucose and insulin, 1/(FPG×FIns),closely correlates to human insulin sensitivity, thus it can be taken as an index of insulin sensitivity in epidemiological studies.