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 binary data: which within study variance estimate to use?

B H Chang1, C Waternaux, S Lipsitz

  • 1Center for Health Quality, Outcomes, and Economic Research, Bedford VA Medical Center, Boston University SPH, 200 Springs Road (Building 70), Bedford, MA 01730, USA. bhchang@bu.edu

Statistics in Medicine
|June 28, 2001
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

Patients with Down syndrome can be included in early phase clinical trials- a report from the T2016-003 Therapeutic Advances in Childhood Leukemia and Lymphoma (TACL) study.

Leukemia·2023
Same author

Using extended reality (XR) for medical training and real-time clinical support during deep space missions.

Applied ergonomics·2022
Same author

Assessment of the Non-Technical Skills for Surgeons (NOTSS) framework in the USA.

The British journal of surgery·2020
Same author

Variation in global uptake of the Surgical Safety Checklist.

The British journal of surgery·2020
Same author

Variation in expenditure for common, high cost surgical procedures in a working age population: implications for reimbursement reform.

BMC health services research·2019
Same author

Coordinate regulation of residual bone marrow function by paracrine trafficking of AML exosomes.

Leukemia·2015
Same journal

Checking Genetic Homogeneity Between Two Samples Using Summary Statistics With Application to Mendelian Randomization.

Statistics in medicine·2026
Same journal

A Bayesian Learning Model for Joint Risk Prediction of Alcohol and Cannabis Use Disorders.

Statistics in medicine·2026
Same journal

Reluctant Transfer Learning in Penalized Regressions for Individualized Treatment Rules Under Effect Heterogeneity.

Statistics in medicine·2026
Same journal

Predictor-Assisted Nonparametric Graphical Models With Multivariate Error-Prone Data.

Statistics in medicine·2026
Same journal

Optimizing Treatment Decision Estimation for Right-Censored Survival Data Through Parameter Transfer Learning.

Statistics in medicine·2026
Same journal

Latent Class Log-Linear Models for Estimating Diagnostic Test Accuracy Without a Gold Standard: A Simulation Study.

Statistics in medicine·2026
See all related articles

This study on schizophrenia outcome data found that a new method for estimating within-study variance significantly improves meta-analysis results. The improved variance estimation leads to less biased estimates and more accurate confidence intervals in mixed-effects models.

Area of Science:

  • Biostatistics
  • Psychiatry
  • Meta-analysis

Background:

  • Meta-analysis of schizophrenia outcome studies requires accurate estimation of between- and within-study variation.
  • Traditional methods for estimating within-study variance in logit models can be highly correlated with the outcome, potentially biasing results.

Purpose of the Study:

  • To investigate the impact of different within-study variance estimators on meta-analysis results for schizophrenia outcomes.
  • To compare a novel variance estimator with the traditional one using real data and simulations.

Main Methods:

  • Applied mixed-effects models to 180 schizophrenia outcome studies.
  • Utilized rate difference and logit models, comparing two estimators for within-study variance: (p̂(i)(1-p̂(i))n(i))⁻¹ and (p̄(1-p̄)n(i))⁻¹.

Related Experiment Videos

  • Estimated fixed effects regression coefficients (β) and random study effect variance (τ²) using quasi-likelihood estimating equations.
  • Conducted a simulation study to evaluate estimator performance under varying conditions (number of studies, study size).

    • Main Results:

    • Estimates of β and τ² differed substantially between the two variance estimators in the schizophrenia meta-analysis.
    • The novel estimator (p̄(1-p̄)n(i))⁻¹ resulted in less biased estimates of β and τ² and more accurate 95% confidence interval coverage in simulations.
    • Simple regression analysis is inappropriate unless τ² >> σ²(i) or robust variance is used.

    Conclusions:

    • The choice of within-study variance estimator significantly impacts meta-analysis findings in schizophrenia research.
    • The proposed alternative variance estimator (p̄(1-p̄)n(i))⁻¹ offers improved accuracy and reliability for mixed-effects meta-analyses.
    • Careful consideration of variance estimation is crucial for valid meta-analytic conclusions in psychiatric research.