Abstract 210: Rates of Drug Switching and Discontinuation among Nonvalvular Atrial Fibrillation Elderly Patients Treated with Direct Oral Anticoagulants in the US

Abstract

Background: While many studies have demonstrated the efficacy of direct-acting oral anticoagulants (DOACs) for stroke risk reduction among nonvalvular atrial fibrillation (NVAF) patients, there is little information regarding drug switching and discontinuation patterns in real-world settings, particularly in older adults. Thus, we evaluated rates of switching and discontinuation among elderly NVAF patients who initiated any of the three most commonly prescribed DOACs in the US. Methods: Patients ≥65 years who initiated apixaban, rivaroxaban or dabigatran (index event/date) were identified from the Humana database (1/1/2013-9/30/2017) and grouped into cohorts by index DOAC. Patients were required to have an NVAF diagnosis and continuous health plan enrollment for ≥12 months prior to the index date and ≥3 months during the follow-up period after the index date. Percentages of patients in each study cohort who switched to any other DOAC or warfarin or discontinued their index DOAC were evaluated. Multivariable Cox regression analyses were carried out to evaluate the impact of index DOAC treatments on the likelihood of switching or discontinuation while controlling for differences in patient characteristics. Results: Of the final study population (n=38,250), 56% initiated apixaban (n=21,376; mean age: 78.6 years), 37% rivaroxaban (n=14,277; mean age: 77.4 years), and 7% dabigatran (n=2,597; mean age: 77.0 years). During follow-up, the switching rate was lowest for patients treated with apixaban at 5.2%, followed by patients treated with rivaroxaban (10.6%), and patients treated with dabigatran (16.9%, p<0.001 across the 3 cohorts). Patients treated with apixaban also had the lowest discontinuation rate (63.2% vs. 68.7% vs. 71.7%, p<0.001). After controlling for patient differences, those treated with rivaroxaban and dabigatran had significantly greater likelihood of switching treatment than patients treated with apixaban (Figure A). Also, rivaroxaban and dabigatran treated patients were significantly more likely to discontinue DOAC treatment (Figure B). Conclusion: Of the DOACs evaluated in this real-world study, apixaban was associated with the lowest rates of switching and discontinuation among elderly NVAF patients.