Comparison of Drug Switching and Discontinuation Rates in Patients with Nonvalvular Atrial Fibrillation Treated with Direct Oral Anticoagulants in the United States

INTRODUCTION: Continuous usage of direct oral anticoagulants (DOACs) among nonvalvular atrial fibrillation (NVAF) patients is essential to maintain stroke prevention. We examined switching and discontinuation rates for the three most frequently initiated DOACs in NVAF patients in the USA. METHODS: Patients who initiated apixaban, rivaroxaban, or dabigatran (index event/date) were identified from the Pharmetrics Plus claims database (Jan 1, 2013-Sep 30, 2016, includes patients with commercial and Medicare coverage) and grouped into cohorts by index DOAC. Patients were required to have a diagnosis of NVAF and continuous health plan enrollment for 12 months prior to the index date (baseline period) and at least 3 months during the follow-up period. Drug switching rates to any other DOAC or warfarin and index DOAC discontinuation rate were evaluated separately with descriptive statistics, Kaplan-Meier analysis, and multivariable Cox regression analysis. RESULTS: Of the NVAF study population (n = 41,864), 37% initiated apixaban (n = 15,352; mean age 62 years), 51% initiated rivaroxaban (n = 21,250; mean age 61 years), and 13% initiated dabigatran (n = 5262; mean age 61 years). During the follow-up period, the unadjusted drug switching rates of patients treated with apixaban, rivaroxaban, and dabigatran were 3.6%, 6.3%, and 11.1%, respectively (p < 0.001 across the three cohorts); while the index DOAC discontinuation rates were 52.8%, 60.3%, and 62.9%, respectively (p < 0.001). After we controlled for differences in patient characteristics, patients treated with rivaroxaban (HR 1.8; 95% CI 1.6-2.0; p < 0.001) and dabigatran (HR 3.4; 95% CI 3.0-3.8, p < 0.001) had a significantly greater likelihood for drug switching than patients treated with apixaban. Also, both rivaroxaban (HR 1.1; 95% CI 1.1-1.2, p < 0.001) and dabigatran (HR 1.3; 95% CI 1.2-1.3, p < 0.001) treated patients were more likely to discontinue treatment. CONCLUSION: In the real-world setting, patients with NVAF newly treated with apixaban were less likely to switch or discontinue treatment compared to patients treated with rivaroxaban or dabigatran.