This study examined warfarin therapy discontinuation and its risk factors among patients with unprovoked venous thromboembolism (VTE) in the US clinical practice setting. Adult patients with unprovoked VTE were identified from the MarketScan claims database from January 1, 2006 to December 31, 2012. The index date was defined as the date of first VTE diagnosis. Patients were required to have no VTE diagnosis in the 6 months before index date and continuous health plan enrollment for 6 months before and 12 months after the index date. Warfarin discontinuation rates and adjusted hazard ratios (HRs) were reported. Of 21,163 eligible patients, 15,463 were diagnosed with deep vein thrombosis (DVT) only (73.1%), 5027 with pulmonary embolism (PE) only (23.7%), and 673 with DVT and PE (3.2%). The average duration of warfarin therapy was 5.2 months (SD = 3.0). During 1-year follow-up, 21.4% patients discontinued therapy within 3 months, 42.8% within 6 months, and 70.1% within 12 months. PE versus DVT [HR = 0.77, 95% confidence interval (CI) = 0.74-0.80], comorbid atrial fibrillation (HR = 0.73, 95% CI = 0.66-0.81), thrombophilia (HR = 0.62, 95% CI = 0.54-0.71), and age >40 years (41-65 years: HR = 0.86, 95% CI = 0.81-0.91; >65 years: HR = 0.82, 95% CI = 0.77-0.87) were significantly associated with reduced risk of warfarin discontinuation. Alcohol abuse/dependence (HR = 1.36, 95% CI = 1.20-1.55), cancer history (HR = 1.13, 95% CI = 1.07-1.19), bleeding (HR = 1.07, 95% CI = 1.01-1.15), and catheter ablation (HR = 1.10, 95% CI = 1.00-1.20) in the 6 months before index date were significantly associated with increased risk for warfarin discontinuation. In conclusion, nearly 1 of 4 patients with unprovoked VTE discontinued warfarin within 3 months. Three of 4 patients discontinued therapy within 1 year. Younger age and multiple clinical factors are associated with warfarin therapy discontinuation.