Byline: Gregory Y.H. Lip, From the Institute of Cardiovascular Sciences, University of Birmingham, United Kingdom (G.Y.H.L.) Liverpool Centre for Cardiovascular Science, University of Liverpool and Liverpool Heart & Chest Hospital, United Kingdom (G.Y.H.L.) Aalborg Thrombosis Research Unit, Department of Clinical Medicine, Aalborg University, Denmark (G.Y.H.L.) Health Economics and Outcomes Research, STATinMED Research, Ann Arbor, MI (A.K.) Worldwide Health Economics and Outcomes Research, Bristol-Myers Squibb Company, Lawrenceville, NJ (X. Li, M.H.) Patient Health & Impact, Outcomes & Evidence, Pfizer, Inc, New York, NY (C.M., J.M.) US Health Economics and Outcomes Research, Bristol-Myers Squibb Company, Lawrenceville, NJ (K.G., A.N.) Patient Health & Impact, Outcomes & Evidence, Pfizer, Inc, Groton, CT (X. Luo) Worldwide Medical, Bristol-Myers Squibb Company, Lawrenceville, NJ (K.F.) Center for Observational Research and Data Sciences, Bristol-Myers Squibb Company, Lawrenceville, NJ (X.P.) Deparment of Internal Medicine, University of Michigan, Ann Arbor (O.B.) Department of Hospital Medicine, Ochsner Clinic Foundation, New Orleans, LA; and Ochsner Clinical School, University of Queensland School of Medicine, New Orleans, LA (S.D.).; Allison Keshishian; Xiaoyan Li; Melissa Hamilton; Cristina Masseria; Kiran Gupta; Xuemei Luo; Jack Mardekian; Keith Friend; Anagha Nadkarni; Xianying Pan; Onur Baser; Steven Deitelzweig Abstract BACKGROUND AND PURPOSE-: This ARISTOPHANES study (Anticoagulants for Reduction in Stroke: Observational Pooled Analysis on Health Outcomes and Experience of Patients) used multiple data sources to compare stroke/systemic embolism (SE) and major bleeding (MB) among a large number of nonvalvular atrial fibrillation patients on non-vitamin K antagonist oral anticoagulants (NOACs) or warfarin. METHODS-: A retrospective observational study of nonvalvular atrial fibrillation patients initiating apixaban, dabigatran, rivaroxaban, or warfarin from January 1, 2013, to September 30, 2015, was conducted pooling Centers for Medicare and Medicaid Services Medicare data and 4 US commercial claims databases. After 1:1 NOAC-warfarin and NOAC-NOAC propensity score matching in each database, the resulting patient records were pooled. Cox models were used to evaluate the risk of stroke/SE and MB across matched cohorts. RESULTS-: A total of 285a292 patients were included in the 6 matched cohorts: 57a929 apixaban-warfarin, 26a838 dabigatran-warfarin, 83a007 rivaroxaban-warfarin, 27a096 apixaban-dabigatran, 62a619 apixaban-rivaroxaban, and 27a538 dabigatran-rivaroxaban patient pairs. Apixaban (hazard ratio [HR], 0.61; 95% CI, 0.54-0.69), dabigatran (HR, 0.80; 95% CI, 0.68-0.94), and rivaroxaban (HR, 0.75; 95% CI, 0.69-0.82) were associated with lower rates of stroke/SE compared with warfarin. Apixaban (HR, 0.58; 95% CI, 0.54-0.62) and dabigatran (HR, 0.73; 95% CI, 0.66-0.81) had lower rates of MB, and rivaroxaban (HR, 1.07; 95% CI, 1.02-1.13) had a higher rate of MB compared with warfarin. Differences exist in rates of stroke/SE and MB across NOACs. CONCLUSIONS-: In this largest observational study to date on NOACs and warfarin, the NOACs had lower rates of stroke/SE and variable comparative rates of MB versus warfarin. The findings from this study may help inform the discussion on benefit and risk in the shared decision-making process for stroke prevention between healthcare providers and nonvalvular atrial fibrillation patients. CLINICAL TRIAL REGISTRATION-: URL: https://www.clinicaltrials.gov/. Unique identifier: NCT03087487.