Results of trials with new oral anticoagulant drugs and vitamin K antagonists (VKA) might not be directly applicable to Dutch clinical practice due to the high level of control of anticoagulation in the Netherlands. In addition, the Dutch method foassessing anticoagulation control uses cross-sectional international normalised ratio (INR) test results while the method used in the trials is based on person-time. To enable comparisons, the two calculation methods were applied to INR data of a cohortof 5422 atrial fibrillation patients treated with VKA. Overall, 74% of test results and 77% of person-time were in the therapeutic range [2.0-3.5]. For the narrower target INR interval [2.5-3.5], 59% of test results and 61% of person-time were in range.It was only between two and six months after the start of treatment that the percentage of person-time in range was lower than the percentage of test results in range. Control of anticoagulation, expressed as a percentage of person-time spent in range,in this Dutch dataset was similar to recent trials with new oral anticoagulants, although it should be noted that the Dutch INR target is higher than the target in these trials. INR control as estimated by the two calculation methods (cross-sectional andlongitudinal) was similar.