A multicenter, observational, retrospective, cross-sectional study of patients, receiving oral anticoagulation therapy (OAT) for stroke prophylaxis in chronic non-valvular atrial fibrillation (NVAF) was conducted in the US, Canada, France, Italy and Spain according to their predominant model of care [routine medical care (RMC) or Anticoagulation Clinic care (ACC)]. The study objectives were to assess anticoagulation control (time in target range), and to describe the features of the local model of care. Consecutive patients were recruited on the basis of a minimum of 60 days of oral anticoagulant treatment over a 12 month period, and clinic and physician details were captured by means of a structured face-to-face or telephone interview. Time in therapeutic range (TTR) was calculated by using linear interpolation between INR values. A total of 1511 patients were recruited, of whom 1445 were included in the analysis of TTR. TTR was higher in ACC (69.5% and 64.9% for Italy and Spain, respectively) with respect to RMC (58.1%, 62.8% and 59.3% for the US, Canada and France, respectively). Mean intervals between INR determinations were between 3 and 4 weeks. Dose changes in case of INR outside therapeutic range were more frequent in Spain and less frequent in France. Striking differences were observed in type of VKA used, specialists involved in patient management, and dosage instructions. Studying of anticoagulation management based on local models of care highlights important discrepancies among countries and suggests further standardization of the management of this important therapy is necessary.