The FRAIL-AF trial was a pragmatic, multicenter, open-label, randomized controlled superiority trial on the safety of switching from a vitamin K antagonist (VKA) to a non- vitamin K oral anticoagulant (NOAC in frail older patient with atrial fibrillation (AF). The primary outcome was major and clinically non-major bleeding. Follow-up was 1 year. The dataset in the repository contains the full pseudonymised data of 661 patient in the control arm continuing VKA, 661 patient in the intervention arm switching from VKA to NOAC, and 66 patients in the observational arm excluded for randomisation due to an estimated glomerular filtration rate (eGFR) <30 mL/min–1/1.73 m–2.