The aim of this proposal is to assess the efficacy of prolonged treatment with low dose warfarin in the secondary prevention of venous thromboembolism (VTE) among patients with and without factor V Leiden mutation, a common inherited defect of hemostasis associated with increased risk of recurrent thrombus formation. The investigators propose to conduct a randomized, double-blind, placebo-controlled, multi-center trial among 800 men and women age 40 or greater with a documented idiopathic venous thrombosis who have completed a standard course of anticoagulation therapy within the last two years. Enrollment will be targeted to ensure at least 300 patients with factor V Leiden are randomized. Patients will be randomized to usual care plus placebo or to usual care plus a three to four year regimen of low-dose warfarin (intended INR 1.5-2.0). After an initial titration period, double-blind INR assessment and dose adjustment will be performed for all patients every two months, with a weekly follow-up whenever a blinded dose adjustment greater than 1 mg is made, and on an as needed basis consistent with best clinical judgment. Trial endpoints will include recurrent VTE, major bleeding episodes, and all cause mortality in the total patient population and separately in those patients who have factor V Leiden. Both qualifying events and reported endpoints will be confirmed by detailed medical record review using standardized diagnostic criteria based upon venographic and doppler ultrasound data, ventilation- perfusion scans, and pulmonary angiography. The trial has sufficient power to discern whether prolonged anticoagulation with low dose warfarin will reduce rates of recurrent disease with low bleeding risks both for all patients enrolled and for the subgroup of those who carry factor V Leiden mutation. The potential clinical impact of this trial is broad since a positive finding would strongly support chronic anticoagulation among the 300,000 US patients hospitalized each year with venous thrombosis who are at risk for recurrent disease following cessation of standard outpatient anticoagulation and for whom there is no regimen currently proven to have an acceptable risk to benefit ratio to support long term prophylaxis.