Serum insulin concentrations have been used as markers of insulin resistance in population studies examining the relationship between insulin resistance and blood pressure, but the relationship is variable among studies. We hypothesized that differences in cross-reactivity of insulin assays with proinsulin and its split/des-amino products might account for the variation. We therefore examined fasting and post-glucose load serum insulin concentrations (determined by both specific and conventional assays), insulin sensitivity (measured by the euglycaemic clamp technique), and blood pressure, in a group of 56 diabetic (NIDDM) and non-diabetic subjects. Insulin concentrations as measured by the two methods were highly correlated (r = 0.97, p < 0.0001), and the relationships among serum insulin concentrations, insulin sensitivity and blood pressure were independent of assay method; for example, in non-diabetic subjects the univariate correlation between log10AUC insulin and insulin sensitivity index was similar with both methods [r = -0.81 vs. r = -0.82, p < 0.0001 (specific vs. conventional assay)]. Discrepancies between studies in the relationship between serum insulin concentrations and blood pressure are unlikely to be due to cross-reactivity of conventional insulin assays with proinsulin-like molecules.