During studies on warfarin, heparin and various anticoagulants with novel mechanisms of action, the activated partial thromboplastin time (aPTT) and the (apparent) international normalized ratio (INR) from a bedside monitor (Coagucheck Plus(R)) were compared with laboratory assay results. Data were compared using the Bland and Altman method of comparison where systematic differences result in significant slopes of the regression line. During heparin treatment, the bedside monitor largely underestimated the aPTT (slope = -0.80). During treatment with the direct thrombin inhibitor napsagatran (slope = 0.99), the pentasaccharides Org31540/SR90107A (slope = 0.77) and SanOrg34006 (slope = 0.35), and warfarin (slope = 0.60), the bedside monitor underestimated the aPTT at lower aPTT levels, while at higher aPTT levels it overestimated the laboratory values. The bedside monitor slightly overestimated the INR during treatment with warfarin (slope = 0.33). Apparent INR was largely overestimated during treatment with Org31540/SR90107A (slope = 1.38), SanOrg34006 (slope = 0.97), Napsagatran (slope = 1.23), and recombinant tissue factor pathway inhibitor (slope = 1.48, P < 0.001 for all regression lines). These results indicate that a substantial disagreement in aPTT or (apparent) INR exists between the bedside monitor and laboratory assay during treatment with the studied 'classic' and novel anticoagulants. The amount of disagreement depended on the anticoagulant given.