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 hybrid model for nonignorable dropout in longitudinal binary responses.

Kenneth J Wilkins1, Garrett M Fitzmaurice

  • 1Department of Biostatistics, Harvard School of Public Health, 655 Huntington Avenue, Boston, Massachusetts 02115, USA. kwilkins@hsph.harvard.edu

Biometrics
|March 18, 2006
PubMed
Summary

This study introduces a new statistical method to address missing data in longitudinal studies with binary outcomes. The approach effectively handles non-ignorable dropout using a hybrid model and EM algorithm for accurate analysis.

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

Big data integration for enhanced epidemiological research: insights and directions from NHLBI's workshop.

Frontiers in digital health·2026
Same author

Among individuals who die of COVID-19, is the percentage who had diabetes actually higher than in those dying of other viral infections?

Research square·2026
Same author

Phase 1 Study of Oral N-Acetylmannosamine in Primary Podocytopathies.

Kidney international reports·2026
Same author

Altered RBC aggregability in diabetes: a threshold for pathophysiological structure-function RBC changes.

Cardiovascular diabetology. Endocrinology reports·2025
Same author

The Effect of COVID-19 on Incident Diabetes in Pediatric Patients: Findings From the National COVID-19 Cohort Collaborative (N3C).

Pediatric diabetes·2025
Same author

Behavioral Therapy as an Adjunct to Buprenorphine Treatment for Opioid Use Disorder: A Secondary Analysis of 4 Randomized Clinical Trials.

JAMA network open·2025

Area of Science:

  • Biostatistics
  • Longitudinal Data Analysis
  • Clinical Trials

Background:

  • Longitudinal studies with binary responses often suffer from non-ignorable dropout, biasing results.
  • Existing methods like selection and pattern-mixture models have limitations in handling such data.
  • Accurate analysis requires methods that account for the reasons behind missing data.

Purpose of the Study:

  • To develop a likelihood-based method for handling non-ignorable dropout in longitudinal binary response data.
  • To formulate a hybrid model combining strengths of existing approaches.
  • To provide a transparent and identifiable statistical framework for inference.

Main Methods:

  • A hybrid statistical model was formulated to accommodate various forms of non-ignorable dropout.

Related Experiment Videos

  • The Expectation-Maximization (EM) algorithm was employed for likelihood-based estimation.
  • Identifying constraints were imposed to ensure model identifiability.

    • Main Results:

    • The proposed method effectively handles non-ignorable dropout in longitudinal binary data.
    • The hybrid model offers flexibility in specifying dropout mechanisms.
    • The EM algorithm provides a robust estimation procedure.

    Conclusions:

    • The developed likelihood-based method is suitable for analyzing longitudinal studies with non-ignorable dropout.
    • The approach maintains transparency in model constraints and identification.
    • The method was successfully applied to a contraceptive randomized clinical trial dataset.