Importance of Shrinkage in Empirical Bayes Estimates for Diagnostics: Problems and Solutions

Journal Title: The AAPS Journal - Year 2009, Vol 11, Issue 3

Abstract

Empirical Bayes (“post hoc”) estimates (EBEs) of ηs provide modelers with diagnostics: the EBEs themselves, individual prediction (IPRED), and residual errors (individual weighted residual (IWRES)). When data are uninformative at the individual level, the EBE distribution will shrink towards zero (η-shrinkage, quantified as 1-SD(ηEBE)/ω), IPREDs towards the corresponding observations, and IWRES towards zero (ε-shrinkage, quantified as 1-SD(IWRES)). These diagnostics are widely used in pharmacokinetic (PK) pharmacodynamic (PD) modeling; we investigate here their usefulness in the presence of shrinkage. Datasets were simulated from a range of PK PD models, EBEs estimated in non-linear mixed effects modeling based on the true or a misspecified model, and desired diagnostics evaluated both qualitatively and quantitatively. Identified consequences of η-shrinkage on EBE-based model diagnostics include non-normal and/or asymmetric distribution of EBEs with their mean values (“ETABAR”) significantly different from zero, even for a correctly specified model; EBE–EBE correlations and covariate relationships may be masked, falsely induced, or the shape of the true relationship distorted. Consequences of ε-shrinkage included low power of IPRED and IWRES to diagnose structural and residual error model misspecification, respectively. EBE-based diagnostics should be interpreted with caution whenever substantial η- or ε-shrinkage exists (usually greater than 20% to 30%). Reporting the magnitude of η- and ε-shrinkage will facilitate the informed use and interpretation of EBE-based diagnostics.

Authors and Affiliations

Radojka M. Savic, Mats O. Karlsson

Keywords

Related Articles

Influence of Copolymer Composition on In Vitro and In Vivo Performance of Celecoxib-PVP/VA Amorphous Solid Dispersions

Previous studies suggested that an amorphous solid dispersion with a copolymer consisting of both hydrophobic and hydrophilic monomers could improve the dissolution profile of a poorly water-soluble drug compared to the...

US pharmacopeia council of experts 2005–2010: Work plans, new revision approaches, and other enhancements

The United States Pharmacopeial Convention (the USP Convention), which meets at 5-year intervals, last convened in 2005. At that meeting, the convention membership elected a new Council of Experts for the 2005–20...

Polymeric Nanocarriers for Transport Modulation across the Pulmonary Epithelium: Dendrimers, Polymeric Nanoparticles, and their Nanoblends

The online version of this article (doi:10.1208/s12248-014-9588-5) contains supplementary material, which is available to authorized users.

Multiple Imputation of Missing Covariates in NONMEM and Evaluation of the Method’s Sensitivity to η-Shrinkage

Multiple imputation (MI) is an approach widely used in statistical analysis of incomplete data. However, its application to missing data problems in nonlinear mixed-effects modelling is limited. The objective was to impl...

Mechanistic Modeling of Monocarboxylate Transporter-Mediated Toxicokinetic/Toxicodynamic Interactions Between γ-Hydroxybutyrate and l -Lactate

Overdose of γ-hydroxybutyrate (GHB) can result in severe respiratory depression. Monocarboxylate transporter (MCT) inhibitors, including l -lactate, increase GHB clearance and represent a potential treatment for...

Download PDF file
  • EP ID EP681490
  • DOI  10.1208/s12248-009-9133-0
  • Views 131
  • Downloads 0

How To Cite

Radojka M. Savic, Mats O. Karlsson (2009). Importance of Shrinkage in Empirical Bayes Estimates for Diagnostics: Problems and Solutions. The AAPS Journal, 11(3), -. https://europub.co.uk/articles/-A-681490