Abstract Background At least 10 million people in Europe have a diagnosis of atrial fibrillation (AF). People with AF commonly have concomitant frailty, rendering them vulnerable to various adverse outcomes. Whilst appropriate prescription of oral anticoagulation (OAC) is associated with reduced risk of stroke and mortality, there are fears of iatrogenic harm in older people with frailty. Purpose Previous studies give conflicting evidence of the association between frailty and OAC prescription and are based on small samples from select cohorts. Therefore, we provide data of the association between OAC prescription and frailty for a large representative cohort of patients with AF. Methods This cross-sectional study used EHR for 536,955 patients in England aged ≥65 years on 31/12/2015. Clinical Terms Version 3 (CTV-3) codes were used to identify AF and relevant past medical history (PMH, including: cancer, varices, intracranial or gastrointestinal haemorrhage). Frailty was determined according to the validated electronic frailty index (eFI, a cumulative deficit score of 36 possible deficits), and categorised into robust (0–0.12), mild (>0.12–0.24), moderate (>0.24–0.36) or severe (>0.36) frailty. Patients with a CHA2DS2-VASc score of ≥2 were considered eligible for OAC prescription. Prescription of OAC among those eligible (warfarin or direct oral anticoagulant [DOAC]) or not was established using prescribing data within the EHR. Poisson regression modelling was used to determine the odds of OAC prescription for each frailty category compared with non-frail (robust) individuals. Odds ratios (ORs) and 95% confidence intervals (CIs) were reported with and without adjustment for sex, deprivation, PMH, concomitant prescription of medications that increase bleeding risk (including antiplatelets, steroids, non-steroidal anti-inflammatories). Results Of the cohort, 61,177 (11.4%) had AF. Of these, 58,204 (95.1%) were eligible for OAC, which was prescribed in 30,916 (53.1%) people. Individuals prescribed OAC were on average 5 months younger than those not prescribed OAC (80.1 vs 80.5 years, p<0.001), and had a slightly higher CHA2DS2-VASc score (4.0 vs 3.8, p<0.001). Frailty was identified in 54,734 (89.5%) patients with AF. OAC was prescribed in 2,028 of 4,863 (41.7%) patients in the robust category; 10,221 of 19,198 (53.2%) with mild; 11,167 of 20,099 (55.6%) with moderate; and 7,500 of 14,044 (53.4%) with severe frailty. In comparison to the robust category, frailty was associated with higher odds of OAC prescription: mild frailty OR 1.6 (95% CI 1.5 to 1.7); moderate 1.7 (1.6 to 1.9); severe 1.6 (1.5 to 1.7). Adjustment for confounding variables increased the magnitude of the association (Figure 1). Conclusion People with AF and advancing frailty were more likely to be prescribed an anticoagulant than those with AF who are robust. Specific safety and efficacy data for OAC are needed in people with AF and frailty to better inform clinical decision-making. Figure 1 Funding Acknowledgement Type of funding source: Public hospital(s). Main funding source(s): CW was funded by the Hull-York Medical School. He is now an NIHR clinical lecturer.