OBJECTIVES: Atrial Fibrillation (AF) is the most common type of heart rhythm disorder, with a prevalence of 0.5% in the adult population. It causes a significant increase of cardiovascular complications and reduction of long-term survival. The objective of this analysis was to assess the epidemiologic and economic burden of AF from a large population based-study. METHODS: The study population was identified through the DENALI data warehouse of the Italian Lombardy Region: with a probabilistic linkage DENALI matches demographic, clinical and economic data of different Healthcare Administrative databases. The study population was made by all subjects who, during the period January 2000–December 2010, were hospitalized for AF and flutter (ICD-9-CM: 427.3) or received ablation of heart tissue (ICD-9-CM: 37.33 and 37.34) and/or conversion of cardiac rhythm (ICD-9-CM: 99.61,99.62, 99.69). The first hospital admission date was used as index date. We estimated incidence, mortality and healthcare costs (hospitalizations, drugs and outpatient examinations/visits) per patient-year from the National Health Service’s perspective. RESULTS: During the study period, around 510,000 subjects (50.2% male) were detected, corresponding to 5 AF cases on 1,000 Lombardy inhabitants per year. Subjects had median(min-max) age of 74(0-110) years. According to the CHADS score, 62.8% and 15.6% of the study population had a moderate and high risk of stroke, respectively. The overall mortality rate (x100 person-years) was 9.9, increasing significantly with the increasing of CHADS score: 4.6, 8.9 and 18,5 for low, moderate and high stroke risk, respectively. The average cost during the index year was around 9,600€/patient-years: the main driver was represented by hospitalizations (83%), followed by drugs (9%). CONCLUSIONS: Administrative database analysis is an efficient tool to track epidemiologic and medical picture in patients with AF, which poses a significant burden in term of incidence, mortality and costs.