Purpose There are few comparative studies for Apixaban and Rivaroxaban in Asian Atrial Fibrillation (AF) patients. Our primary objectives were to investigate the relative safety and effectiveness of Apixaban versus Rivaroxaban for 1) all bleeding (major and clinically relevant non-major bleeding), and 2) composite of ischemic strokes and all-cause mortality outcomes. Methods In an active comparator, new-user cohort study with a follow-up period of 2-years, we followed 2,532 patients (Apixaban: n=1,273, Rivaroxaban: n=1,259) who were adults aged 21 years and above from Singapore with an inpatient index diagnosis of AF from 1 January 2019 to 31 December 2020. We used a database containing electronic health records from all public healthcare institutions in Singapore. We ascertained baseline covariates and time to their first events by comprehensive chart review of discharge summaries, emergency department notes and laboratory tests. Results In modified intention-to-treat analyses adjusted on confounders identified in directed acyclic graphs, initiating Rivaroxaban at the point of index AF was associated with a small increased risk of all bleeding as compared to Apixaban over 2-years (HR=1.26, 95% CI: 1.04-1.53), although we observed attenuated effects in on-treatment analyses (HR=1.17, 95% CI: 0.92-1.49). We observe no evidence for differences between initiating either agent for the composite of ischemic strokes and all-cause mortality (HR=0.85, 95% CI: 0.68-1.06). Conclusions In patients newly diagnosed with AF in Singapore, there is a small increased risk of major and clinically relevant non-major bleeding in patients initiated on Rivaroxaban as compared to Apixaban. This is consistent with many other observational studies, although we observe a much smaller magnitude of a bleed risk difference in our population. We are unable to dismiss that small differences may be present for ischemic stroke or all-cause mortality among Apixaban and Rivaroxaban users and recommend larger further studies to confirm these findings.