CV3 - Comparative Effectiveness of Direct Oral Anticoagulants Versus Warfarin in Patients with Atrial Fibrillation

Abstract

Atrial fibrillation (AF) management focuses on the prevention of blood clots to reduce the risk of stroke. Warfarin has successfully reduced the risk of stroke and mortality. Since 2010, a new class of medications, direct oral anticoagulants (DOACs), entered the market. Clinical trials provide good evidence that DOACs compare favorably to warfarin on a variety of outcomes, including mortality. Initial observational studies have concluded these medications may also be cost-effective. Our research objective was to compare the effectiveness of DOACs to warfarin in clinical practice. This is a retrospective observational study of data from patient-level administrative and claims data from the Veterans Health Administration (VHA) and Medicare. We identified patients with a new diagnosis of AF who initiated warfarin or a DOAC between 2011 and 2015 (N=35,478). We examined the relationship between initiating warfarin or a DOAC on mortality using a two-stage residual inclusion approach, using facility-level prescribing patterns as an instrumental variable. Models controlled for facility quality measures, patient demographic variables, comorbidities, and stroke risk. Eighty percent of the sample initiated warfarin with the remainder initiating the DOACs. The average follow-up time was 1.97 years. In the first stage model, facility prescribing patterns significantly predicted the likelihood of receiving a DOAC (OR=6.04, 95% CI=2.5-13.8). In the second-stage, Cox proportional hazard models that control for selection bias, Veterans who started on a DOAC had significantly lower risk of mortality (hazard ratio=0.34, C.I.=0.22-0.53). Naïve models and propensity score models that don’t explicitly account for unobserved selection bias have hazard ratios of 0.66 (CI=0.61-0.71) and 0.70 (CI=0.65-0.76), respectively. Veterans who initiate a DOAC compared to warfarin have significantly lower risk of death. The ease of administration and monitoring advantages for DOACs suggest these drugs may be the preferred option for treatment in clinical populations with significant comorbidities.