Tacrolimus is widely used to prevent acute rejection after transplantation, but achieving therapeutic blood concentrations of tacrolimus is often difficult because of large pharmacokinetic variability. In this study, the applicability of the Bayesian method to individualize tacrolimus dose was prospectively examined. Twenty adult recipients (Bayesian group) and another 20 adult patients (control group), all of whom underwent living-donor liver transplantation, were enrolled in this study. In the Bayesian group, the dose of tacrolimus during the first 3 and 4 weeks after surgery was adjusted with the Bayesian method using a population pharmacokinetic model, targeting a trough level of 5 to 12 ng/mL. The interindividual variability in tacrolimus concentrations was significantly reduced in the Bayesian group compared with the control group (average percentage coefficient of variation for all occasions, 32% vs 44% and 31% vs 39% in the first 3 and 4 weeks, respectively). In addition, more patients achieved the target concentrations in the Bayesian group than in the control group (average for all occasions, 85% vs 59% and 83% vs 70% in the first 3 and 4 weeks, respectively). These findings suggest that the Bayesian method can be used to calculate maintenance doses of tacrolimus in adult patients early after living-donor liver transplantation.