BackgroundEstablishing the appropriate rivaroxaban dose in older patients with non-valvular atrial fibrillation (NVAF) is important because of the high risk of adverse events. In this EXPAND study subanalysis, we examined the safety and efficacy of standard-dose (15 mg/day) and non-recommended reduced-dose (10 mg/day) rivaroxaban in patients aged ≥65 years with NVAF and preserved renal function. MethodsThe entire analysis population (ALL cohort [n = 3982]; ≥65 years) was divided into early elderly (ELD) (65–74 years [n = 1444]) and late ELD (≥75 years [n = 2386]) sub-cohorts. Each sub-cohort was divided into reduced-dose and standard-dose groups. Kaplan–Meier survival curves with adjusted hazard ratios (aHRs) and 95% confidence intervals (CIs) were used to assess efficacy (thromboembolic events) and safety (hemorrhagic events) outcomes. ResultsThe aHR for major bleeding did not differ between the dosages in any of the cohorts (aHRs: 0.86–0.93). There were no significant differences in the occurrence of stroke + systemic embolism (SE) or stroke + SE + myocardial infarction (MI) + cardiovascular (CV) death among the cohorts. The aHR for MI/unstable angina + interventional/CV surgery + CV death was higher with 10-mg/day rivaroxaban than 15-mg/day rivaroxaban in the ALL cohort (aHR: 1.56 [95% CI 1.02–2.37], p = 0.039) and the late ELD sub-cohort (aHR: 1.86 [95% CI 1.01–3.42], p = 0.045). ConclusionsReduced-dose rivaroxaban may increase the risk of coronary artery events. The use of rivaroxaban 15 mg/day in patients with NVAF aged ≥75 years with preserved renal function was supported.