Aims: In patients with a low AF burden and long periods of sinus rhythm, 'pill-in-the-pocket' oral anticoagulation (OAC) may, taken as needed in response to AF episodes, offer the same thromboembolic protection as continuous, life-long OAC, while reducing bleeding complications at the same time. The purpose of this study is to systematically summarise available evidence pertaining to the feasibility, safety and efficacy of pill-in-the-pocket OAC. Methods: Medline and Embase were searched from inception to July 2022 for studies adopting a pill-in-the-pocket OAC strategy in AF patients guided by daily rhythm monitoring (PROSPERO/CRD42020209564). Outcomes of interest were extracted and event rates per patient-years of follow-up were calculated. A random effects model was used for pooled estimates. Results: Eight studies were included (711 patients). Daily rhythm monitoring was continuous in six studies and intermittent in two (pulse checks or smartphone single-lead electrocardiograms were used). Anticoagulation criteria varied across studies, reflecting the uncertainty regarding the AF burden that warrants anticoagulation. The mean time from AF meeting OAC criteria to its initiation was not reported. Adopting pill-in-the-pocket OAC led to 390 (54.7%) patients stopping OAC, 85 (12.0%) patients taking pill-in-the-pocket OAC and 237 (33.3%) patients remaining on or returning to continuous OAC. Overall, annualised ischaemic stroke and major bleeding rates per patient-year of follow-up were low at 0.005 (95% CI [0.002-0.012]) and 0.024 (95% CI [0.013-0.043]), respectively. Conclusion: Current evidence, although encouraging, is insufficient to inform practice. Additional studies are required to improve our understanding of the relationships between AF burden and thromboembolic risk to help define anticoagulation criteria and appropriate monitoring strategies.