Outcomes and direct treatment costs with novel oral anticoagulants compared to clinic-monitored warfarin for stroke prevention in atrial fibrillation



Journal Title

Journal ISSN

Volume Title



Objectives: To describe patient characteristics and evaluate costs and outcomes of novel oral anticoagulants compared to clinic-monitored warfarin for the prevention of stroke and systemic embolism in patients with atrial fibrillation within the Scott & White Healthcare system. Methods: Patients with atrial fibrillation, CHADS₂ score ≥ 1, and a prescription claim for dabigatran, rivaroxaban, or warfarin between 2010 and 2012 were evaluated over 12 months. Patients in the warfarin cohort were enrolled in an Anticoagulation Clinic. Patients were matched 1:1 for age, CHADS₂, and gender for comparisons between groups. Baseline characteristics, medication adherence, occurrence of adverse events, and treatment costs were compared using inferential statistics. Anticoagulation control was assessed for patients in the warfarin cohort. Results: 141 and 471 patients met criteria for the novel cohort group and the warfarin group, respectively. After matching, 136 remained in each cohort. Prior to matching, compared to the warfarin cohort, the novel anticoagulant cohort had a higher proportion of male patients (63% versus 49%), and lower average CHADS₂ score (2.65 versus 3.30), while average age in both cohorts was similar (75 years). Matched cohorts had similar adherence rates (88% for novel versus 87% for warfarin). After matching, annual medication cost in 2014 US dollars for dabigatran or rivaroxaban averaged $2,658 (SD $1,494) compared to $1,066 (SD $633) for warfarin, including monitoring costs. Annual total all-cause healthcare costs averaged $23,711 (SD $22,910) for dabigatran or rivaroxaban, compared to $18,248 (SD $24,184) for warfarin. For the 95 warfarin patients with INR values, time in therapeutic range averaged 70.4%. Conclusion: Compared to clinic-monitored warfarin, more men than women were prescribed new oral anticoagulants and these patients averaged a lower CHADS₂ score. After matching, patient adherence was high and comparable between groups. Anticoagulation control for warfarin patients was similar to clinical trials. Annual medication cost was significantly greater for new oral anticoagulants than clinic-monitored warfarin, including INR monitoring costs. Total annual all-cause healthcare costs were significantly greater for patients taking new oral anticoagulants compared to warfarin, although too few adverse events occurred to draw conclusions regarding event rates and costs of ischemic stroke and major bleeds.