BACKGROUND: The optimal means of achieving therapeutic oral anticoagulation in the outpatient setting has not been determined. OBJECTIVE: To compare a 10-mg dosing nomogram with a 5-mg nomogram that has been suggested to be sufficient for warfarin initiation. DESIGN: Randomized, controlled clinical trial. SETTING: Outpatient venous thromboembolism services of four tertiary care hospitals. PATIENTS: 201 of 210 consecutive patients with objectively confirmed diagnoses of acute venous thromboembolism. INTERVENTION: All patients were treated with subcutaneous low-molecular-weight heparin for a minimum of 5 days until a therapeutic international normalized ratio (INR) was achieved. Patients were randomly assigned to initially receive a 10-mg or 5-mg dose of warfarin. MEASUREMENTS: The primary end point was time in days to therapeutic INR. Secondary end points were the proportion of patients who had achieved a therapeutic INR by day 5, the total number of INR assessments, the number of INR measurements greater than 5.0, incidence of recurrent venous thromboembolism and major bleeding, and survival. RESULTS: 210 consecutive patients met the inclusion criteria. Of these, 9 were excluded and 201 were randomly assigned to study groups (104 to the 10-mg group and 97 to the 5-mg group). Demographic characteristics of both groups were similar. Patients in the 10-mg group achieved therapeutic INR 1.4 days earlier than patients in the 5-mg group (P < 0.001). Eighty-three percent of patients in the 10-mg group achieved a therapeutic INR by day 5 versus 46% in the 5-mg group (P < 0.001). Fewer INR assessments were performed in the 10-mg group than in the 5-mg group (8.1 vs. 9.1; P = 0.04). There were no significant differences between the two groups in recurrent events, major bleeding, survival, and number of INR measurements greater than 5.0. CONCLUSION: The 10-mg warfarin initiation nomogram is superior to the 5-mg nomogram because it allows more rapid achievement of a therapeutic INR.