The use of oral contraceptives is associated with altered plasma concentrations of many components of the coagulation and fibrinolysis system, increased plasma levels of markers indicating in vivo coagulation and fibrinolysis, and a modified response to challenge tests both in vivo and in vitro. None of these effects seems to be specific for users of oral contraceptives and none was found uniformly in all users. The predictive value of each of these effects, or even of certain combinations of tests, for the prediction of venous thrombosis is low. There is no established way to assess the "thrombogenicity" of particular pills. The individual susceptibility, however, to develop venous thrombosis varies considerably. Recently, several congenital abnormalities of the hemostatic system have been found that are associated with an increased risk of venous thrombosis in general. The risk associated with the use of oral contraceptives appears to act synergistically with some of these thrombophilic conditions. Although the prevalence of these congenital predispositions varies among different populations, screening for these conditions is not feasible: negative results would not exclude the occurrence of about two-thirds of oral contraceptives-associated thromboses and positive results are likely to be disregarded because of their poor predictive value. Future research has to evaluate the role of a more targeted screening strategy aiming at women with risk factors such as a positive personal or family history of venous thromboembolism.