Abstract 17609: Comparison of All-cause and Bleeding-related Hospitalizations Among Non-valvular Atrial Fibrillation Patients Receiving Oral Anticoagulants
To compare the risk of hospitalization among non-valvular atrial fibrillation (NVAF) patients newly initiated with an oral anticoagulant (OAC): apixaban, dabigatran, rivaroxaban, or warfarin.
Retrospective cohort study using Humana Medicare Advantage data from 7/1/2009 - 9/30/2014. NVAF patients ≥18 years receiving one OAC on the index date with 6 months continuous enrollment prior to index prescription date and 3 months post-index were eligible. Hospitalizations were identified by standard codes for inpatient admission. Bleeding-related hospitalizations required an additional code for major/clinically relevant non-major (CRNM) bleeding. A cox proportional hazards model was used to estimate the hazard ratios (HR) of hospitalizations adjusted for age, sex, region, comorbidities and comedications. Adherence for each OAC was also calculated using a proportion of days covered approach to understand medication taking behaviors.
Among the 53,168 patients initiated on an OAC, 2,028 (3.8%) apixaban, 5,644 (10.6%) dabigatran, 7,667 (14.4%) rivaroxaban and 37,829 (71.1%) warfarin. Patients in apixaban cohort were older (mean 75.5 years, P <0.05) with higher mean CHA2DS2-VASc score (P <0.05). Abixaban patients had a higher mean HAS-BLED score vs. dabigatran (P <0.0001), lower mean score vs. warfarin (P <0.0001) and did not differ significantly vs. rivaroxaban (P =0.46). Patients receiving apixaban had a significantly lower risk for all-cause hospitalization across cohorts, and a sig. lower risk for bleeding-related hospitalization vs. patients receiving rivaroxaban or warfarin (Table). Adherence ranged from 87.8% to 90.4% across cohorts.
In a real-world setting, initiation with apixaban was associated with a significantly lower risk for all-cause hospitalization, and a significantly lower risk of bleeding-related hospitalization compared to rivaroxaban or warfarin.
Table: Adjusted Hazard Ratios of All-cause and Bleeding-related Hospitalizations
Author Disclosures: S. Deitelzweig: Consultant/Advisory Board; Significant; Consultant. A. Bruno: Employment; Significant; Employee of Bristol Myers Squibb. K. Gupta: Employment; Significant; Employee of Bristol Myers Squibb. J. Trocio: Employment; Significant; Employee of Pfizer Inc. N. Tate: Employment; Significant; Employee of Bristol Myers Squibb.
- © 2015 by American Heart Association, Inc.