OBJECTIVES: Atrial fibrillation has been estimated to affect as many as 2.3 million Americans, making it the second most common cardiovascular condition in the United States. Atrial fibrillation has been found to increase patient’s risk of stroke by 5-fold. We sought to calculate the projected total treatment costs, quality-adjusted survival and cost-effectiveness of dabigatran and adjusted-dose warfarin for stroke prevention in patients with atrial fibrillation.   METHODS: This three-state Markov transition model (healthy with atrial fibrillation, disability, and death) simulated the treatment costs, quality-adjusted survival and cost-effectiveness of dabigatran 150 mg twice daily and adjusted-dose warfarin (international normalized ratio of 2-3) for stroke prevention in atrial fibrillation. Our base-case consisted of a hypothetical cohort of ≥65 year old patients with atrial fibrillation, a moderate risk of stroke (CHADS2≥1) and no contraindications to anticoagulation therapy. The parameters used in the model were adopted from the literature research, Cost-effectiveness was calculated over a patient’s lifetime and using a societal perspective (excluding indirect costs). One-way and threshold sensitivity analyses were performed on all relevant variables. RESULTS: The mean quality-adjusted life expectancy of simulated patients was 12.9 and 12.2 years for those receiving dabigatran and warfarin. Total lifetime treatment costs were $146,649 and $118,904. The incremental cost-effectiveness ratio was $40,580. Upon one-way sensitivity analysis, our conclusions were found to be sensitive to changes in dabigatran cost and the differential efficacy of the two strategies.  Threshold sensitivity analysis further revealed that daily dabigatran costs greater than $13 per day and differential efficacy between the two strategies of less than 0.15% per year resulted in incremental cost-effectiveness ratios greater than $50,000 per quality-adjusted life year gained. CONCLUSIONS: Our analysis suggested that dabigatran is cost-effective for stroke prevention in atrial fibrillation; however, this conclusion was sensitive to changes in dabigatran costs and the antithrombotic efficacy of the two treatment strategies.