To evaluate the clinical course of patients with AF at high risk for both bleeding and stroke, according to OAC use. Data were analysed from three registries across the Middle East, Europe, and Asia-Pacific regions. The study only included ‘high risk’ patients with AF and CHA₂DS₂-VASc scores ≥ 2 and HAS-BLED scores ≥ 3, who were divided into two groups based on OAC use: OAC users and OAC non-users. Of the 2,535 patients (41.7% female; mean age 75.4 ± 7.8 years), 80.3% (n = 2,037) received OAC therapy. OAC non‑users showed significantly higher crude 1‑year event rates of all‑cause death (116 [23.3%]), MACE (96 [19.3%]) and major bleeding (31 [6.2%]); after multivariable adjustment, they had higher odds of all‑cause death (adjusted odds ratio (aOR) 2.23, 95% CI 1.65-3.01), MACE (aOR 1.92, 95% CI 1.38-2.64) and major bleeding (aOR 2.38, 95% CI 1.42-3.92) compared to OAC users. Enrolment in a non-European setting was associated with a lower risk of all-cause death (aOR 0.61, 95%CI 0.44-0.85) and MACE (aOR 0.42, 95%CI 0.28-0.62). In patients with AF at high risk of both bleeding and stroke, OAC non-use was associated with higher risk of adverse events and bleeding. Decisions on discontinuation of OACs in this subset of patients with AF should be cautiously made and such patients require careful re-evaluation and follow-up.