Background: The prospective, single-arm, observational, phase 4 ETNA-AF Europe study collected real-world data about safety, effectiveness and therapeutic adherence in European patients with non-valvular atrial fibrillation newly prescribed with edoxaban and followed up for 4 years., Methods: Overall, 13 164 patients were included in the full-analysis set, which means that they had at least one documentation after baseline at 4 years. The current paper reports about the 3329 Italian patients out of the whole European population., Results: In the Italian cohort, median age was 76.0 (69.0-82.0) years, with 57.4% of the patients being ≥75 years old. The CHA2DS2-VASc score was >4 in 586 (18.1%) patients. At baseline, 670 (20.8%) patients were classified as frail by the investigators. Edoxaban 30 mg/day was prescribed to 1013 (31.8%) patients: these were older, with more comorbidities and a lower estimated creatinine clearance compared with those receiving 60 mg/day. All-cause mortality was 4.1%/year and there were very low yearly rates of bleeding and thromboembolic events: major bleeding, 0.9%; intracranial hemorrhage, 0.2%; ischemic stroke, 0.3%; systemic embolism, <0.1%. These events were more frequent in patients ≥75 years or in patients with renal impairment or treated with edoxaban 30 mg/day. Advancing age was not associated with an increased incidence of intracranial bleeding., Conclusions: These findings confirm the favorable long-term safety and effectiveness profile of edoxaban in non-valvular atrial fibrillation patients treated in routine clinical care in Italy.