Recent improvements in the methods of clinical trials and the use of accurate objective tests to detect venous thromboembolism have made possible a series of randomized trials to evaluate various treatments for venous thromboembolism. The results of these trials have resolved many of the uncertainties a clinician confronts in selecting an appropriate course of anticoagulant therapy. These trials have shown that the intensity of both initial heparin treatment and long-term anticoagulant therapy must be sufficient to prevent unacceptable rates of recurrence of venous thromboembolism. Patients with proximal deep vein thrombosis who receive inadequate anticoagulant therapy have a risk of clinically evident, objectively documented recurrent venous thromboembolism that approaches 20% to 25%. The need for therapy with heparin and the importance of monitoring blood levels of the effect of heparin have been established. The importance of achieving adequate heparinization was suggested by a nonrandomized trial in 1972 and randomized trials in the 1980s have confirmed this finding. Furthermore, randomized trials have demonstrated the importance of achieving adequate heparinization early in the course of therapy. Unfractionated intravenous heparin has provided an effective therapy for more than half a century, but the need to monitor therapy and establish therapeutic levels is a fundamental problem. It is evident that validated heparin protocols are more successful in establishing adequate heparinization than intuitive ordering by the clinician. However, even with the best of care using a heparin protocol, some patients treated with intravenous heparin will receive subtherapy. In this context, subtherapy reflects a practical limitation of the use of unfractionated heparin, rather than a poor standard of care. Furthermore, it is recognized that the practical difficulties associated with heparin administration are compounded by the substantive practical difficulties of standardizing APTT testing and the therapeutic range. Our findings emphasize the confounding effect that initial heparin treatment has on long-term outcome. In future trials of longterm therapy, it is imperative that the initial therapy is of adequate intensity and duration; failure to administer adequate initial treatment may lead to a poor outcome that is falsely attributed to the long-term therapy under evaluation. Therapy with low-molecular-weight heparin, which does not require monitoring and dose finding, is the likely practical solution to these dilemmas. Based on the experience of difficulties achieving adequate therapy with subcutaneous unfractionated heparin dosing, we administered a low-molecular-weight heparin formulation in a single daily dose, rather than splitting the treatment into 2 equal doses. The initial intensity of therapy was thereby maximized. Therapy with low-molecular-weight heparin proved to be better than therapy with unfractionated heparin.