dataset to compute DeltaOFV. The power of correctly estimating CL-CLCR significance was computed as the percentage of simulated datasets using the following 3 decision criteria DeltaOFV larger than 3.84 (P<0.05), 6.64 (P<0.01), and 10.8 (P<0.001). When the significance level was 0.001
power becomes 81.6%, 60.2% and 34.7% for 3 decision criteria, respectively, yielding the expected model rejection ratio of higher than about 20% when the covariate that might be present in data was marginally significant. Although this work was carried out based on the data obtained from one particular clinical trial, we hope that this work can provide an insight into covariate selectivity associated with healthy volunteer data.