OBJECTIVES: Vitamin K antagonists (VKAs) have a narrow therapeutic window and numerous drug and food interactions and therefore require frequent INR monitoring. This study aimed to estimate VKA treatment related costs in patients with chronic nonvalvular atrial fibrillation (NVAF). METHODS: Data were collected as part of the French arm of ISAM - International Study of Anticoagulation Management. Cost analysis was from a societal perspective and limited to direct medical costs. A randomly selected sample of general practitioners (GPs) and cardiologists (43 GPs, 20 cardiologists) established a register of all patients who had received at least 60 consecutive days of VKA treatment for chronic NVAF in the preceding year. Study data came from medical records and patient interviews. VKA treatment related costs included INR tests, blood sample draws, physician consultations, nurse cost and VKA related hospitalisations. Drug costs were not considered. RESULTS: Of 278 patients interviewed, 264 patients had at least two consecutive INR results within the study period. In total, 3,026 INR tests were collected over a cumulative follow-up time of 188.4 years. The mean cost for INR tests was €164.8/patient/year (ppy); 41.2% of blood samples were taken at the patient's home by a nurse, with a mean driving distance of 4.6 km, representing an additional cost of €6.3 for each INR taken at home (average €41.5/ppy). VKA treatment generated 630 GP and 171 cardiologist face-to-face consultations, at a cost of €87.8/ppy. Of 71 hospitalisations reported, 13 were judged attributable to VKAs by an event adjudication committee. These hospitalisations added a further €193.1/ppy. CONCLUSIONS: The annual cost of ambulatory VKA follow-up was €294.1 per patient. VKA related hospitalisations in this patient sample added €193.1, but the exclusion of patients unable to be interviewed may have resulted in underestimation of costs associated with severe and fatal adverse events (stroke).