OBJECTIVE To determine the use of oral anticoagulants in polypathological patients with atrial fibrillation and its influence on mortality and loss of functionality. METHODS Patients with polypathological patient criteria and atrial fibrillation were included in an observational, prospective and multicenter study. Data on demographic, clinical, functional and sociofamilial characteristics, CHADS2 score, levels of hemoglobin, albumin and creatinine, use of oral anticoagulants and survival and functional status at one year were collected. RESULTS Five hundred and thirty-two (32.6%) of 1,632 polypathological patients had atrial fibrillation. The stroke risk was high in 505 (94.9%), moderate in 24 (4.5%) and low in 3 (0.6%) patients. Oral anticoagulants were used in 61% of patients with CHADS2 score≥2 and in 37.5% with CHADS2 score=1. Oral anticoagulants were less used in older patients, with more functional and cognitive impairment. Heart failure was associated with more use of oral anticoagulants. There was no difference by the presence of hypertension, diabetes, anemia, renal insufficiency or stroke. In multivariate analysis the use of oral anticoagulants was independently associated with lower age, lower cognitive impairment, absence of hepatic disease and with higher stroke risk. The prescription of oral anticoagulants was independently associated with more survival at one year with no influence on functional status. CONCLUSIONS Oral anticoagulants are underused in polypathological patients with atrial fibrillation despite being associated with more survival.