Dabigatran (D) and rivaroxaban (R) showed better benefit-risk than VKA for stroke prevention in non-valvular atrial fibrillation (NVAF), but no randomized trial compared D versus R. Purpose To compare the 2-year risk of major events for D vs. R new users for NVAF in real-life: standard dosage (D150mg vs. R20mg) and reduced dosage (D110mg vs. R15mg).

Cohort of D or R new users for NVAF in 2013 identified and followed for 2 years in the SNDS French nationwide claims database. NVAF was defined from long-term disease registration, hospitalisation diagnosis or procedure for atrial fibrillation without valvular disease (3-year database history). D and R patients were 1:1 matched according to standard or reduced dosage, on gender, age, drug start date and high-dimensional propensity scores (hdPS) including individual stroke and bleeding risk factors from CHA2DS2-VASc and HAS-BLED. Hazard ratios (HR) [95% confidence interval] were estimated on treatment using Cox proportional hazard risk or Fine and Gray models.

From the 10,847 D150, 15,532 D110, 18,829 R20 and 11,195 R15 new users for NVAF in 2013, 8290 D150/R20 per arm, and 7639 D110/R15 per arm were matched, i.e. 76% and 68% of patients of the lowest group (D150 and R15 groups), respectively; with very good overlap of hdPS distributions, all standardized differences<0.1 and most of them<0.02. The D vs. R HR for clinically relevant bleeding was 0.55 [0.43–0.70] for standard dosage and 0.77 [0.64–0.92] for reduced dosage, and respectively 0.92 [0.67–1.26] and 0.73 [0.56–0.94] for stroke and systemic embolism, 0.93 [0.66–1.29] and 0.95 [0.71–1.26] for acute coronary syndrome, 0.84 [0.65–1.11] and 0.95 [0.83–1.09] for death, and 0.74 [0.64–0.86] and 0.88 [0.79–0.97] for the composite of the 4 major events.

This nationwide study of new anticoagulant users for NVAF shows a better benefit-risk of dabigatran standard and reduced dosages compared to rivaroxaban in real-world setting.