BackgroundOral anticoagulants (OACs) are effective in reducing the risk of cardioembolic stroke due to atrial fibrillation. While most nursing home residents with atrial fibrillation qualify for anticoagulation based on clinical guidelines, the net clinical benefits of OACs may diminish as residents approach the end of life.MethodsWe conducted a cross‐sectional study of 30,503 US nursing home residents with atrial fibrillation (based on Minimum Data Set 3.0 and Medicare Part A records) who used OACs in the year before enrolling in hospice care during 2012–2016. Whether residents discontinued OACs before hospice enrollment was determined using Part D claims and date of hospice enrollment. Modified Poisson models estimated adjusted prevalence ratios (aPR).ResultsAlmost half (45.7%) of residents who had recent OAC use discontinued prior to hospice enrollment. Residents who were underweight (aPR: 1.02; 95% confidence interval [CI]: 1.01–1.03), those with high bleeding risk (aPR: 1.04, 95% CI: 1.03–1.05), and those with moderate or severe cognitive impairment (aPR: 1.02, 95% CI: 1.02–1.03) had a higher prevalence of OAC discontinuation before entering hospice. Residents with venous thromboembolism (aPR: 0.94, 95% CI: 0.93–0.96), statin users (aPR: 0.88, 95% CI: 0.87–0.89), and those on polypharmacy (≥10 medications, aPR: 0.72; 95% CI: 0.71–0.73) were less likely to discontinue OACs before enrollment in hospice.ConclusionAnticoagulants are often discontinued among older nursing home residents with atrial fibrillation before hospice enrollment; it is not clear that these decisions are driven solely by net clinical benefit considerations. Further research is needed on comparative outcomes to inform resident‐centered decisions regarding OAC use in older adults entering hospice.