A Note on Parameterizations in Pharmacokinetic Compartment Models

Abstract

In pharmacokinetics, compartment models often play an important role in the description of the concentration of the drug in the blood over time after its administration to a subject. In the inference in the compartment models in pharmacokinetics, unlike that of the usual nonlinear regression models, parameterizations with respect to the parameters related to pharmacokinetic indices are frequently applied, for some reasons of the constraint that the parameter takes only the positive value, facilitation of the compartment models, or interpretation of the relationships between the parameters and the demographic or physiological individual attributes. However, in practice no attention is paid to checking if the utilized parameterization is valid. In this article, the relative curvature measure that enables us to assess the intrinsic and parameter-effects nonlinearity in the model is applied for checking it from the latter nonlinearity. From a viewpoint of the relative curvature measure, we reconsider several well-known examples and demonstrate the parameterization checking.