1996 Volume 27 Issue 4 Pages 741-757
The present paper reports a new approach to building a population pharmacokinetic and pharmacodynamic (PK/PD) model and interpreting clinical data. We discuss about a historical background of PK/PD modeling, then, two special parametric models. One is the model to assess the time course of the disease progression and the effects of placebo and actual drug treatment. This model is able to distinguish active drug treatment quantitatively from placebo effect. The other model is a subject-specific PK/PD model for analgesia using population approach. It is capable of combining characteristic features of clinical trial, that is, a subject-specific random effect model for a multivariate response that consists of noncontinuous component responses, analyze of nonrandomly censored longitudinal data and statistics to estimate marginal distribution such as Monte Carlo integration. This approach is hopeful about the future in the field of drug development and drug evaluation.