Statistical Assessment of Linear Pharmacokinetics in Clinical Pharmacological Studies

Abstract

In clinical pharmacological studies, the test drug is considered to show linear pharmacokinetics when AUC and C_max increase proportionally to dose level. In this paper, we reviewed three statistical analysis methods such as linear regression analysis, one-way analysis of variance, and a power model, which have been used to assess the dose proportionality. Power model is a simple regression analysis of log of the pharmacokinetic parameter and log of the dose. We also assessed the validity of these methods by means of computer simulation, and confirmed the usefulness of the power model. We therefore recommend the use of the power model, and reporting both the point estimate and its confidence interval of the slope.