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

Meta-analysis of studies with missing data.

Ying Yuan1, Roderick J A Little

  • 1Department of Biostatistics, University of Texas MD Anderson Cancer Center, Houston, Texas 77030, USA. yyuan@mdanderson.org

Biometrics
|June 21, 2008
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

Multiple Imputation with Massive Data: An Application to the Panel Study of Income Dynamics.

Journal of survey statistics and methodology·2023
Same author

A simulation study of diagnostics for selection bias.

Journal of official statistics·2021
Same author

Bayesian sensitivity analyses for longitudinal data with dropouts that are potentially missing not at random: A high dimensional pattern-mixture model.

Statistics in medicine·2021
Same author

Framework for the treatment and reporting of missing data in observational studies: The Treatment And Reporting of Missing data in Observational Studies framework.

Journal of clinical epidemiology·2021
Same author

A comparison of parametric propensity score-based methods for causal inference with multiple treatments and a binary outcome.

Statistics in medicine·2021
Same author

Measures of the Degree of Departure from Ignorable Sample Selection.

Journal of survey statistics and methodology·2020
Same journal

Fast penalized generalized estimating equations for large longitudinal functional datasets.

Biometrics·2026
Same journal

Causally-interpretable random-effects meta-analysis.

Biometrics·2026
Same journal

Statistical inference for mean function of partially observed functional time series.

Biometrics·2026
Same journal

Subgroup identification via Interaction Tree and Mixed Model for Repeated Measures with application to Alzheimer's disease.

Biometrics·2026
Same journal

Finite mixtures of linear quantile regressions with concomitant variables: a solution to endogeneity in longitudinal data modeling.

Biometrics·2026
Same journal

Discussion on "INTACT: a method for integration of longitudinal physical activity data from multiple sources" by Jingru Zhang, Erjia Cui, Hongzhe Li, and Haochang Shou.

Biometrics·2026
See all related articles

Meta-analyses can be biased by missing patient data, especially when attrition rates vary. This study introduces three novel methods to correct for this bias, improving the accuracy of treatment effect estimates in complex research.

Area of Science:

  • Biostatistics
  • Clinical Trials
  • Psychiatry

Background:

  • Meta-analyses combine results from multiple studies but can be compromised by missing patient data (attrition).
  • Standard meta-analysis models assume missing data are ignorable, which may not hold when attrition rates correlate with treatment effects.
  • This can lead to biased estimates of treatment effect size and variance in meta-analyses.

Purpose of the Study:

  • To address bias in meta-analysis caused by patient-level missing data with varying attrition rates.
  • To propose and evaluate methods for correcting bias in meta-analytic estimates when missing data are not ignorable.

Main Methods:

  • Proposed three bias-correction methods: reweighting DerSimonian-Laird estimates by completion rate, a Bayesian random-effects model incorporating completion rate, and a Bayesian shared-parameter model.

Related Experiment Videos

  • Applied these methods to a meta-analysis of 16 randomized trials on combined pharmacotherapy and psychological treatment for depression.
  • Assessed the impact of varying proportions of missing data and study-level attrition on meta-analytic results.

    • Main Results:

    • Standard random-effects meta-analysis can yield biased estimates if primary study attrition rates depend on treatment effect size.
    • The proposed methods offer potential corrections for bias introduced by non-ignorable missing data in meta-analyses.
    • Illustrative meta-analysis demonstrated the practical application of these bias-correction techniques.

    Conclusions:

    • Missing data, particularly when non-ignorable in meta-analysis, requires specific adjustment methods beyond standard approaches.
    • The developed techniques provide a framework for more accurate synthesis of evidence from studies with differential attrition.
    • Accurate meta-analysis is crucial for reliable treatment effect estimation, especially in fields like depression research.