Abstract 339: Hospitalization Rates and Healthcare Costs Among Non-valvular Atrial Fibrillation Patients Who Were Naïve Users of Novel Oral Anticoagulants

Abstract

Background: Novel oral anticoagulants (NOACs) are recommended for stroke risk reduction in non-valvular atrial fibrillation (NVAF) patients withCHADS 2 ≥2. Efficacy and safety of the different NOACs are well documented in the clinical trial setting, however, real-world evidence of the effectiveness and safety of these agents is limited. This study provides an early assessment of all-cause hospitalizations and economic outcomes among NVAF patients initiated on different NOAC treatments in the real-world. Methods: A retrospective cohort study of patients diagnosed with AF (ICD-9-CM 427.31 and 427.32) who were naïve to anticoagulant therapy and initiated on a NOAC (dabigatran, rivaroxaban and apixaban) was conducted using PharMetrics Plus data from 01/ 2012 through 03/2014. The first NOAC prescription date served as the index date. Patients were required to be ≥18 years old and have no evidence of valvular disease, CHADS 2 score ≥ 1, and ≥ 1 month of follow-up. Follow-up ended upon the earliest of the following: treatment discontinuation, switch to a different NOAC, end of continuous enrollment, 1 year after the index date, or end of study. A Cox proportional hazards model were used to model the likelihood and timing of all-cause hospitalizations while a generalized linear model (GLM) was used for costs. Results: There were 9,150 patients included in the study. The majority of patients were prescribed rivaroxaban (61.5%), followed by dabigatran 1,374 (23.5%), and apixaban (15.0%). Males were 71.4% of the sample and mean age was 63.4 years, with the apixaban patients being slightly older (P=0.0170). Mean CHADS 2 score was 1.8. Mean HAS-BLED score was 2.0, and was slightly higher for patients on apixaban and rivaroxaban (P=0.0025). After adjusting for baseline characteristics, treatment with dabigatran (HR 1.37, 95% CI 1.10-1.69) and rivaroxaban (HR 1.57, 95% CI 1.30-1.90) was associated with increased rate of all-cause hospitalization relative to apixaban. Mean monthly all-cause costs were lower for patients on apixaban compared to those on dabigatran ($3,581 vs $4,236; P<0.0001) and rivaroxaban ($3,581 vs $4,144; P<0.0001). Conclusions: This early assessment shows that among anticoagulant naïve NVAF patients, treatment with apixaban was associated with a lower rate of all-cause hospitalization, as well as lower overall costs compared to other NOACs. Further evaluation is needed to provide additional detail on potential drivers of utilization differences.