In a study population consisting of healthy men (n = 8), women not using oral contraceptives (OC) (n = 28) and women using different kinds of OC (n = 187) we used calibrated automated thrombography (CAT) in the absence and presence of added activated protein C (APC) to compare parameters that can be obtained from thrombin generation curves, i.e. lag time, time to peak, peak height and endogenous thrombin potential (ETP). Both with and without APC, plasmas of OC users exhibited the shortest lag time and time to peak, and the highest peak height and ETP. In the absence of APC none of these parameters differed between users of OC containing different progestogens. In contrast, in the presence of APC shorter lag times and time to peak, and higher peak height and ETP were observed in plasma of users of gestodene-, desogestrel-, drospirenone- and cyproterone acetate-containing OC than in plasma of users of levonorgestrel- containing OC. The ETP determined in the absence of APC (ETP(-APC)) had no predictive value for the APCsr (r = 0.11; slope 0.9 x 10(-3); 95% CI: -0.1 x 10(-3) to 2.0 x 10(-3)) whereas the ETP measured in the presence of APC (ETP+APC) showed an excellent correlation with the APCsr (r = 0.95; slope 6.6 x 10(-3); 95% CI: 6.3 x 10(-3) to 6.9 x 10(-3)) indicating that the APCsr is entirely determined by the ETP+APC. In conclusion, OC use increases thrombin generation, but differential effects of second and third generation OCs on the protein C system likely determine the differences in the risk of venous thrombosis between these kinds of OC.