The gold-standard methods to assess insulin sensitivity (IS) and beta-cell function are time-consuming and difficult to use in large-scale clinical or epidemiological studies where simpler methods are required. This has raised interest in obtaining estimates from glucose and insulin measured in the fasting state or during an OGTT. Several indices of beta-cell function and IS obtained from fasting and OGTT measurements have been described and most of them have been validated with reference methods [1±4]. The aim of our study was to examine if the relation between IS and beta-cell function assessed from fasting and OGTT measurements with these simple indices keeps the physiological relation that for the reference methods has been described to be hyperbolic. In 1993, a study showed that the lower the IS the higher the insulin concentrations and the higher the IS, the lower the insulin concentrations so that the product of beta-cell function and IS is approximately a constant [5]. Other studies have confirmed this relation Diabetologia (2000) Abstract Aims/hypothesis. We aimed to find if the relation between insulin sensitivity and beta-cell function assessed from fasting and OGTT measurements has a physiological shape (hyperbolic with the reference methods).Methods. Healthy women without diabetic first-degree relatives underwent a 75 g OGTT with plasma glucose and insulin (n = 35) concentrations being measured at 0, 30, 60 and 120 min. Beta-cell function and insulin sensitivity were estimated using previously described indices from fasting (1 for beta-cell function, 6 for insulin sensitivity) and OGTT measurements (3 for beta-cell function and 5 for insulin sensitivity). A hyperbolic relation was tested for the 21 beta-cell function-insulin sensitivity pairs using a non-lineal regression method.Results. The assessment of beta-cell function from OGTT was impossible in seven women and one had outlier indices. For the remaining 27 women, only 8 combinations adjusted to a hyperbolic relation. The best adjustment was achieved using the fasting glucose to insulin ratio as the estimation of insulin sensitivity and the homeostasis model assessment (HOMA) index (single fasting sample) as the estimation of beta-cell function (r 2 0.802, k 869.71, p < 0.001). Conclusion/interpretation. In this group of healthy women, the estimation of insulin sensitivity and beta-cell function by most methods using OGTT-derived glucose and insulin measurements did not adjust to a hyperbolic relation but all fasting indices combinations did. Beta-cell function estimated with the HOMA index and insulin sensitivity with fasting glucose to insulin ratio had the best adjustment. [Diabetologia (2000