Direct Comparison of the Effectiveness and Safety Among Direct Oral Anticoagulants and Warfarin in Japanese Patients: Nationwide Cohort Study in Japan

Abstract

Direct oral anticoagulant drugs (DOACs) are available in addition to warfarin for the treatment of patients with non‐valvular atrial fibrillation (NVAF). Anticoagulants are useful for practical pharmacotherapy in Asian populations, but their responses are different from those in Caucasian populations. Therefore, we aimed to identify the most useful anticoagulant using a nationwide insurance claims database in Japan. We extracted information on patients with NVAF treated with DOACs or warfarin for the first time between September 2015 and February 2018 to conduct a retrospective observational cohort study using the claims database of Japan. We calculated hazard ratios (HR) of effectiveness and safety endpoints among DOACs or warfarin after adjusting for backgrounds by propensity scores (inverse probability weighting). Using negative control outcomes, we considered renal function as an unmeasured confounding factor. After adjusting for their backgrounds, patients treated with DOACs showed higher effectiveness endpoints (stroke or systemic embolism) than those treated with warfarin (HR; 0.72–0.81) and higher safety in terms of safety end points (clinically significant bleeding) (HR; 0.80–0.88). The net clinical benefit, which reflects the actual clinical usefulness, indicates the advantages of DOACs over warfarin (HR; 0.75–0.82). Dabigatran had lower usefulness than edoxaban and rivaroxaban in Japanese patients treated with NVAF, even after considering the unmeasured confounding factor of kidney function. Based on practical clinical data, edoxaban and rivaroxaban were the most useful anticoagulants in Japanese patients with NVAF.