The assessment of in vivo pancreatic beta cell function as well as insulin action in humans is challenging because of a complex interplay between insulin secretion, insulin sensitivity and hepatic insulin extraction. Simplified, the relationship between insulin secretion and insulin sensitivity can be described by an approximate hyperbola with the product of the two variables being constant for individuals with the same degree of glucose tolerance (the disposition index). Strengths and limitations of the disposition index have been widely debated, and there is an increasing awareness of different measurements of both in vivo insulin secretion and of in vivo insulin action influencing the results and the interpretation of calculated insulin secretion disposition indices. Examples of differential results of insulin secretion disposition indices are the use of oral versus intravenous insulin secretion tests as well as the use of methods assessing in vivo insulin action preferentially measuring the degree of hepatic or peripheral insulin action. Examples of how uncritical use of common tests to measure insulin secretion and insulin action in humans may lead to erroneous results will be illustrated by recent studies of prediabetic subjects with impaired glucose tolerance (IGT) or impaired fasting glucose (IFG), from studies of subjects with genetic or non-genetic predisposition to type 2 diabetes, as well as results from human overfeeding and bed test studies. The underlying assumption in the use of the disposition index that the pancreatic insulin responds to insulin action is to some extent challenged by studies showing that increased insulin secretion may precede peripheral insulin resistance during short term overfeeding in humans. Finally the lecture will address another and until recently unrecognized dimension of the disposition index, namely the issue of adjusting insulin secretion for hepatic versus peripheral insulin sensitivity. An overall take home lesson from the lecture may be that none of the most commonly used in vivo tests of insulin secretion and of insulin action may be used uncritical as “gold standards”, but that the results from different tests should be interpreted within the physiological context in which the test was performed.