Abstract 95: Nonvalvular Atrial Fibrillation Emergency Department Encounters where Dabigatran Etexilate was Administered are Associated with Lower Admission Rate Compared to Encounters where Warfarin was Administered

Abstract

Background: Warfarin and dabigatran etexilate (DE) are oral anticoagulants (OAC) that reduce stroke risk among patients with nonvalvular atrial fibrillation (AF). However, DE does not require titration and INR monitoring. This study examined whether emergency department (ED) rate of admissions differed between the two therapies. Methods: Admission rate was evaluated for hospital encounters initiated in the ED, with a primary or secondary discharge diagnosis of AF between 1/1/2011-3/31/2012, with DE or warfarin administered during the encounter, and excluding encounters of valvular AF patients. Encounters were identified from a hospital Charge Detail Masters database containing 387 eligible hospitals. Samples were propensity score matched using nearest neighbor within a caliper of 0.20 standard deviations of the logit, without replacement, and a 2:1 match. Admission rates were estimated for encounters representing previously-treated patients and those representing treatment-naive patients using binominal generalized linear models, fitted by generalized estimating equations (clustered by hospital). Covariates estimating the propensity score and admission rate included age, payer type, use of bridging agents, AF as primary or secondary diagnosis, CHADS 2 and HAS-BLED scores, comorbid conditions, and hospital attributes. As a sensitivity analysis, admission rate was also estimated from the unmatched sample. Results: Matched samples included 2,688 warfarin and 1,344 DE ED encounters of previously-treated patients out of 15,053 and 1,367 ED encounters, respectively; and 2,578 warfarin and 1,289 DE ED encounters of OAC-treatment-naive patients out of 8,361 and 1,406 ED encounters, respectively. There were too few (n<5) matched encounters where the patient had prior OAC use but were new to the drug administered during the encounter, so these were excluded. No covariates used in matching had standardized mean differences > 10% after matching. Among the previously-treated sample, the estimated admission rate was 3.2% lower for DE compared to warfarin (88.3% vs. 91.5%, p=0.010) with sensitivity analysis confirming a lower admission rate for DE (91.1% vs. 93.8%, delta=2.7%, p=0.001). Among the treatment-naive sample, DE had a 1.2% lower admission rate compared to warfarin (95.2% vs. 96.3%, p=0.048). Sensitivity analysis confirmed a lower admission rate for DE (95.5% vs. 97.0%, delta=1.5%, p=0.001). Conclusions: While the vast majority of AF encounters initiated in the ED result in admission, encounters where patients were treated with DE as continuing or new therapy were less likely to be admitted compared to similar encounters where warfarin was administered.