To study the efficacy and safety of an oral anticoagulation protocol for the treatment of nonvalvular atrial fibrillation, based on scientific associations' recommendations, in unselected patients seen in daily clinical practice. The study included all consecutive patients with permanent nonvalvular atrial fibrillation who attended two outpatient cardiology clinics between February 1, 2000 and February 1, 2002. They were treated according to an anticoagulation protocol based on Spanish Society of Cardiology and American College of Cardiology/American Heart Association/European Society of Cardiology guidelines. Patients were followed up prospectively for major events, such as death, stroke, transient ischemic attack, peripheral embolism and severe hemorrhage, which were recorded by treatment group. A total of 624 patients were included in the study. Those receiving anticoagulation therapy (n=425; 68%) more frequently had hypertension, diabetes and previous embolism as well as a greater number of cardioembolic risk factors (P< .001). Overall, 93% of non-anticoagulated patients received platelet aggregation inhibitors (92% received aspirin). After a median follow-up of 21 months, the probability of an embolic event was lower in anticoagulated patients (0.81% vs 14.04%; P< .001), as was all-cause mortality (3.27% vs 6.42%; P=.003). However, there was no significant difference in the probability of severe bleeding (2.75% vs 2.93%; P=.96). Results were unchanged after adjustment for age, sex, and previous embolic events. Oral anticoagulation therapy for nonvalvular atrial fibrillation implemented according to scientific associations' recommendations is effective and safe in daily clinical practice.