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

Explaining heterogeneity in meta-analysis: a comparison of methods.

S G Thompson1, S J Sharp

  • 1Department of Medical Statistics and Evaluation, Imperial College School of Medicine, Hammersmith Hospital, Du Cane Road, London W12 0NN, U.K. simon.thompson@ic.ac.uk

Statistics in Medicine
|October 16, 1999
PubMed
Summary

Investigating study heterogeneity in meta-analysis is crucial. This study compares methods to identify covariates explaining heterogeneity, like cholesterol reduction

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

Intervention to optimise body mass index in adolescents and address the triple burden of malnutrition-the Ntshembo (Hope) trial in rural and urban South Africa study: a study protocol for a randomised controlled trial.

Trials·2026
Same author

Complement factor I: Regulatory nexus, driver of immunopathology, and therapeutic.

Immunobiology·2023
Same author

Experiences that matter in bipolar disorder: a qualitative study using the capability, comfort and calm framework.

International journal of bipolar disorders·2023
Same author

An investigation of factors affecting changes in health behaviours during the COVID-19 pandemic in a UK population-based cohort study.

Public health·2022
Same author

Validity and severity thresholds for the depression subscale of the affective self rating scale: An equipercentile equating study using classical test theory.

Journal of affective disorders·2021
Same author

Correction: External validation of risk prediction models for incident colorectal cancer using UK Biobank.

British journal of cancer·2020

Area of Science:

  • Biostatistics
  • Epidemiology
  • Medical Research Methodology

Background:

  • Heterogeneity between studies is a key challenge in meta-analysis.
  • Identifying sources of heterogeneity is essential for accurate interpretation of results.
  • Covariate analysis helps explain variability in treatment effects across studies.

Purpose of the Study:

  • To compare methods for investigating whether study-level covariates explain heterogeneity in meta-analyses.
  • To assess the impact of cholesterol reduction on coronary events using meta-analysis.
  • To examine the relationship between treatment effect estimates and precision to detect publication bias.

Main Methods:

  • Comparison of weighted normal errors regression and random effects logistic regression.

Related Experiment Videos

  • Application of covariate analysis to meta-analysis data, including serum cholesterol reduction trials.
  • Examination of methods for assessing publication bias using treatment effect estimates and their precision.

    • Main Results:

    • Methods quantifying explained heterogeneity and the effect of cholesterol reduction on coronary events were compared.
    • The study assessed the relationship between treatment effects and their precision to evaluate publication bias.
    • Weighted regression with restricted maximum likelihood estimators is recommended.

    Conclusions:

    • Methods allowing for residual heterogeneity should be employed in meta-analysis.
    • While computationally feasible, explicitly using original data forms (e.g., binomial models) may yield similar results to normality assumptions.
    • Covariate analysis is vital for understanding heterogeneity and improving meta-analysis validity.