Dissertation/Thesis Abstract

Methods for population pharmacokinetics and pharmacodynamics
by Colby, Emily Anne, Dr.P.H., The University of North Carolina at Chapel Hill, 2012, 109; 3549548
Abstract (Summary)

Current applications of cross validation have been unsuccessful at identifying covariate effects in the population Pharmacokinetic/Pharmacodynamic (PK/PD) setting when other methods find a covariate effect may exist. Software that does population PK/PD modeling has a nice feature of being able to do a post hoc step without any major iterations to obtain Bayesian parameter estimates and hence predictions for subjects that were not in the dataset that was used to fit the model. This work proposes cross validation methods for longitudinal mixed effects models that are effective at identifying covariate effects when they exist.

Indexing (document details)
Advisor: Bair, Eric
Commitee: Bair, Eric, Ivanova, Anastasia, Koch, Gary, Kosorok, Michael, Weiner, Daniel
School: The University of North Carolina at Chapel Hill
Department: Biostatistics
School Location: United States -- North Carolina
Source: DAI-B 74/05(E), Dissertation Abstracts International
Subjects: Biostatistics, Applied Mathematics, Pharmacology
Keywords: Cross validation, Nonlinear mixed effects, Pharmacodynamics, Pharmacokinetics, Population
Publication Number: 3549548
ISBN: 9781267860033