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

Accommodating negative intracluster correlation with a mixed effects logistic model for bivariate binary data

T R Ten Have1, A Kunselman, E Zharichenko

  • 1Center for Biostatistics and Epidemiology, Hershey Medical Center, Pennsylvania State University, Hershey 17033, USA.

Journal of Biopharmaceutical Statistics
|April 21, 1998
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

Evaluation of four pediatric cardiopulmonary bypass circuits in terms of perfusion quality and capturing gaseous microemboli.

Perfusion·2012
Same author

Evaluation of Quadrox-i and Capiox FX neonatal oxygenators with integrated arterial filters in eliminating gaseous microemboli and retaining hemodynamic properties during simulated cardiopulmonary bypass.

Perfusion·2012
Same author

Evaluation of three hollow-fiber membrane oxygenators without integrated arterial filters for neonatal cardiopulmonary bypass.

Perfusion·2011
Same author

Atomoxetine for depression and other neuropsychiatric symptoms in Parkinson disease.

Neurology·2010
Same author

Adjustments for center in multicenter studies: an overview.

Annals of internal medicine·2001
Same author

Mixed effects models with bivariate and univariate association parameters for longitudinal bivariate binary response data.

Biometrics·2001

This study introduces a new mixed effects model for bivariate binary data, effectively handling negative intracluster correlations. This advanced statistical method improves analysis in complex biological and toxicological studies.

Area of Science:

  • Biostatistics
  • Statistical Modeling
  • Toxicology

Background:

  • Standard random intercept logistic models struggle with negative intracluster correlations in bivariate binary data.
  • Existing mixed effects approaches often fail to accurately model such data structures.
  • Accurate statistical modeling is crucial for analyzing complex biological and toxicological outcomes.

Purpose of the Study:

  • To extend the random intercept logistic model to accommodate negative intracluster correlations for bivariate binary response data.
  • To propose a novel mixed effects approach that utilizes separate affine transformations of a single random effect for paired responses.
  • To evaluate the performance of the proposed model against existing methods using real-world data and simulations.

Main Methods:

Related Experiment Videos

  • Development of a modified random intercept logistic model with distinct affine transformations for each response within a cluster.
  • Application of the proposed model to two datasets: a crossover trial and a developmental toxicity study.
  • Comparison with conditional likelihood and generalized estimating equations (GEE) approaches for population-averaged models.

Main Results:

  • The proposed mixed effects approach successfully handles negative intracluster correlations where other methods failed.
  • Simulations indicated that the conditional likelihood approach performs poorly with moderate to strong negative correlations.
  • The proposed model showed slightly more conservative confidence interval coverage compared to population-averaged models.

Conclusions:

  • The extended random intercept logistic model provides a robust solution for bivariate binary data with negative intracluster correlations.
  • This method offers valuable insights, particularly in contexts like crossover trials, complementing population-averaged models.
  • The approach can be extended to higher-dimensional responses, though with constraints on the correlation structure.