The aim of this study was to evaluate the accuracy of the portable coagulometer CoaguChek® (Roche Diagnostics) as a prothrombin time (PT) monitor, and to correlate capillary blood results with those of three different routine methods used for monitoring oral anticoagulant therapy (OAT): capillary, plasma and whole blood samples. Three hospitals participated in the study with a total of 235 patients on OAT. The international normalized ratio (INR) results obtained with CoaguChek were compared with those obtained using each of the routine methods. The study presents a good correlation between the PT monitor and the three methods studied: r = 0.9745 (hospital A), r = 0.9283 (hospital B), r = 0.9136 (hospital C). A simplified concordance test of the methods results in a nine-field comparison table showing concordances of 87.2, 85.7 and 68.4%, respectively. The absolute difference (mean ± SD) between laboratory A and CoaguChek INRs was 0.0571 ± 0.2042, with values of 0.04286 ± 0.3906 for laboratory B and 0.6986 ± 0.6170 for laboratory C. These results confirm that CoaguChek could be used as a new method for oral anticoagulant monitoring, and is in best agreement with the capillary blood PT system.