Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Mixed effects versus fixed effects modelling of binary data with inter-subject variability.

Valda Murphy1, Adrian Dunne

  • 1Department of Statistics and Actuarial Science, University College Dublin, Belfield, Ireland.

Journal of Pharmacokinetics and Pharmacodynamics
|November 12, 2005
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Dynamic population pharmacokinetic-pharmacodynamic modelling and simulation supports similar efficacy in glycosylated haemoglobin response with once or twice-daily dosing of canagliflozin.

British journal of clinical pharmacology·2017
Same author

Population Pharmacokinetic Modeling of Canagliflozin in Healthy Volunteers and Patients with Type 2 Diabetes Mellitus.

Clinical pharmacokinetics·2015
Same author

The method of averaging applied to pharmacokinetic/pharmacodynamic indirect response models.

Journal of pharmacokinetics and pharmacodynamics·2015
Same author

Mixed-effects beta regression for modeling continuous bounded outcome scores using NONMEM when data are not on the boundaries.

Journal of pharmacokinetics and pharmacodynamics·2013
Same author

Shrinkage in nonlinear mixed-effects population models: quantification, influencing factors, and impact.

The AAPS journal·2012
Same author

Canakinumab for acute gouty arthritis in patients with limited treatment options: results from two randomised, multicentre, active-controlled, double-blind trials and their initial extensions.

Annals of the rheumatic diseases·2012
Same journal

Cohort population analysis of sparse data: Dexamethasone pharmacokinetics in mother and fetus based on blood sampling at birth.

Journal of pharmacokinetics and pharmacodynamics·2026
Same journal

Bridging the operational gap in population pharmacokinetic-pharmacodynamic analysis: an international perspective on the 2025 Chinese group standard.

Journal of pharmacokinetics and pharmacodynamics·2026
Same journal

Advancing quantitative clinical pharmacology competencies in Francophone Africa through an on-line learning framework.

Journal of pharmacokinetics and pharmacodynamics·2026
Same journal

Optimizing Subcutaneous Antibody Dosing Regimens Through Operating Space Maps: rHuPH20 Case Study.

Journal of pharmacokinetics and pharmacodynamics·2026
Same journal

Mechanistic modeling of FcRn-dependent IgG drug interactions: Clinical applications and dosing implications.

Journal of pharmacokinetics and pharmacodynamics·2026
Same journal

Comparing heavy-tailed residual error models for outlier handling in population PK modeling.

Journal of pharmacokinetics and pharmacodynamics·2026
See all related articles

For binary data with inter-subject variability, adaptive Gaussian quadrature is superior to Laplace approximation in mixed effects models. Increasing data points per subject improves Laplace approximation accuracy, while more subjects favor mixed effects models.

Area of Science:

  • Pharmacokinetics and Pharmacodynamics
  • Statistical Modeling
  • Biostatistics

Background:

  • Mixed effects models are commonly used for binary data with inter-subject variability and within-subject correlation.
  • Previous studies suggested fixed effects models may yield more accurate estimates under certain conditions.
  • The choice of approximation method for the likelihood function in mixed effects models can impact parameter estimation.

Purpose of the Study:

  • To compare the performance of Laplace and adaptive Gaussian quadrature approximations in mixed effects models for binary data.
  • To investigate the impact of the number of observations per subject and the total number of subjects on parameter estimation accuracy.
  • To re-evaluate the necessity of mixed effects models versus fixed effects models for binary data analysis.

Related Experiment Videos

Main Methods:

  • Simulation experiments were conducted, replicating a previous study.
  • Two binary observations per subject were used initially, with subsequent increases in observations per subject and total subjects.
  • Mixed effects models were fitted using both Laplace and adaptive Gaussian quadrature approximations.

Main Results:

  • The Laplace approximation produced extreme outliers in parameter estimates, unlike the adaptive Gaussian quadrature.
  • Outliers from the Laplace approximation occurred when it overestimated the likelihood in extreme parameter spaces.
  • Increasing observations per subject reduced outliers for the Laplace approximation; increasing subjects favored mixed effects model estimates over fixed effects.

Conclusions:

  • Adaptive Gaussian quadrature is a more robust approximation for mixed effects models with binary data compared to the Laplace approximation.
  • The number of observations per subject and the total number of subjects influence the choice and performance of statistical models.
  • Mixed effects models offer advantages in parameter estimation accuracy with larger sample sizes and allow for estimation of fixed effects parameters, unlike fixed effects models.