Reproducibility of a glucose tolerance test

Abstract

A statistical method is described for testing the significance of changes in the glucose tolerance of experimental subjects in different conditions. Statistically significant changes in glucose tolerance from day to day were observed in seven normal men who underwent glucose tolerance tests on 3 successive days. No statistically significant changes in glucose tolerance were found on the 2nd day after a glucose tolerance test in further experiments using five normal men. Possible physiological mechanisms by which the performance of a glucose test could cause alterations in glucose tolerance are discussed. Submitted on December 19, 1960