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

A mixed effects Markov model for repeated binary outcomes with non-ignorable dropout.

Robert J Gallop1, Thomas R Ten Have, Paul Crits-Christoph

  • 1Department of Mathematics, Applied Statistics Program, West Chester University, West Chester, PA 19383, USA. rgallop@wcupa.edu

Statistics in Medicine
|October 28, 2005
PubMed
Summary

This study introduces an enhanced Markov model for analyzing two-state processes with irregular observations and non-ignorable dropout. The model effectively handles subject-specific variations and demonstrates robustness in simulations for various research applications.

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

Understanding time as a predictor of mental health, academic performance, and help-seeking behavior among undergraduates: a mixed methods analysis.

Journal of American college health : J of ACH·2026
Same author

Factors Associated With Nocturnal Hypoglycemia After Exercise in the Type 1 Diabetes and Exercise Initiative (T1DEXI) Study.

Diabetes care·2026
Same author

Clinical Presentation and Outcomes of Insulinomas in Pediatric Patients.

Hormone research in paediatrics·2026
Same author

Cognitive Therapy vs. Short-Term Dynamic Psychotherapy in a Community Mental Health Setting: A Preliminary Analysis of Effects in Several Severity Domains.

Clinical psychology & psychotherapy·2026
Same author

Longitudinal changes in bone mineral density after initiation of elexacaftor-tezacaftor-ivacaftor in youth and adults with cystic fibrosis: PROMISE-ENDO.

Journal of cystic fibrosis : official journal of the European Cystic Fibrosis Society·2026
Same author

Enduring effects of brief interpersonal therapy on depression from pregnancy through 1-year postpartum: A randomized clinical trial.

Journal of mood and anxiety disorders·2026

Area of Science:

  • Statistics
  • Biostatistics
  • Stochastic Processes

Background:

  • Repeated binary measures often model two-state stochastic processes with transitions between states.
  • Continuous-time transitions are frequently observed at discrete, individual-specific time points.
  • Standard analyses often rely on the Markov assumption, with prior work introducing conditional Markov models for random effects.

Purpose of the Study:

  • To extend existing conditional Markov models by incorporating a non-ignorable dropout component.
  • To develop a methodology that accommodates subject-to-subject variation and complex state transitions.
  • To ensure a closed-form expression for the marginal likelihood through specific random effects distributions.

Main Methods:

  • Extension of Cook's conditional Markov model (1999) to include non-ignorable dropout.

Related Experiment Videos

  • Specification of random effects distributions to achieve a closed-form marginal likelihood.
  • Application and illustration using diverse datasets: parasitic infection, cocaine treatment, and aging studies.

    • Main Results:

    • The proposed shared parameter model effectively handles non-ignorable dropout in two-state stochastic processes.
    • The methodology is demonstrated to be applicable across various research fields.
    • Simulations indicate the model's robustness against at least one alternative non-ignorable model.

    Conclusions:

    • The enhanced Markov model provides a flexible framework for analyzing longitudinal binary data with complex dependencies.
    • The inclusion of non-ignorable dropout improves the analysis of real-world data where dropouts are informative.
    • The model's robustness and demonstrated applications highlight its utility in biostatistical research.