Atrial fibrillation (AF) is the most common abnormal heart rhythm in elderly patients. Rivaroxaban has been widely used for stroke prevention. The anticoagulant response to rivaroxaban increases with age, which may make elderly patients susceptible to adverse outcomes resulting from small differences in bioavailability between generic and brand products.We designed a cohort study of ≥65-year-old inpatients with AF. Sociodemographic and laboratory measures of qualified patients who received brand or generic rivaroxaban for at least 72 hours at the study hospital from January 2021 to June 2023 were collected retrospectively. The primary outcome was the incidence of bleeding.A total of 1008 qualifying patients were included for analysis, with 626 (62.1%) receiving brand rivaroxaban and 382 (37.9%) receiving generic rivaroxaban. After propensity score matching and weighting to account for confounders, the odds ratios comparing brand vs generic rivaroxaban (95% confidence intervals) for the bleeding was 1.15 (0.72-1.82). Results from subgroup analyses of patients with age ≥85, HAS-BLED score ≥ 3, containment of antiplatelet drugs, and female patients were consistent with the primary analysis.It provides evidence regarding the clinical safety outcome of generic rivaroxaban in the elderly AF population that may be particularly susceptible to adverse outcomes resulting from small allowable differences in pharmacokinetics.