Institute of Statistics and Decision Sciences

Duke University

presents:

Jon Wakefield

Imperial College of Science, Technology, and Medicine

London, UK

"The Bayesian Modelling of Covariates for Population Pharmacokinetic Models"

Abstract:

Pharmacokinetic models describe the time course of a drug and its metabolites in a particular individual. Population pharmacokinetic models attempt to identify and quantify sources of between-individual variability in observed concentrations. Crucial to this aim is the identification of those covariates which are responsible for explaining the variability. In this paper we discuss how covariate modelling can be carried out for population pharmacokinetic models. We argue that the importance of a particular covariate can only be discussed with reference to the specific use for which the model is intended. Covariate selection is important in population pharmacokinetic studies as it aids in determining dosage recommendations for specific populations, as defined by particular covariates. We describe a Bayesian predictive procedure which places covariate modelling in the context of dosage determination. The approaches utilize Markov chain Monte Carlo techniques. For covariate selection we extend the approach of George and McCulloch (1993).

The methods are illustrated using population pharmacokinetic data from a study of the antibiotic Vancomycin in babies. These data are sparse, containing just 180 concentrations from 37 babies. Eight covariates are available from which we construct a covariate model. We argue that it is crucial in problems such as these to incorporate relevant prior information.

Keywords :- population pharmacokinetics, covariate selection, Markov chain Monte Carlo, dosage determination.

Friday, September 8, 1995

11:45 - 12:45

116 Old Chemistry Building

Any questions concerning the seminar may be addressed to Cheryl McGhee @ (919) 684-8029, e-mail cheryl@isds.duke.edu, or finger seminar@isds.duke.edu.