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

Generalized linear mixed models for meta-analysis.

R W Platt1, B G Leroux, N Breslow

  • 1McGill University, Montreal Children's Hospital Research Institute, Canada. robertp@epid.lan.mcgill.ca

Statistics in Medicine
|April 16, 1999
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

A Prospective Cohort Study of Bisphenol A Exposure from Dental Treatment.

Journal of dental research·2020
Same author

Obstetric haemorrhage and risk of cardiovascular disease after three decades: a population-based cohort study.

BJOG : an international journal of obstetrics and gynaecology·2020
Same author

Incretin-based drugs and risk of lung cancer among individuals with type 2 diabetes.

Diabetic medicine : a journal of the British Diabetic Association·2020
Same author

Pioglitazone and bladder cancer: improving research methods.

Diabetic medicine : a journal of the British Diabetic Association·2020
Same author

Machine learning for the prediction of postpartum complications is promising, but needs rigorous evaluation.

BJOG : an international journal of obstetrics and gynaecology·2019
Same author

A systematic review of observational studies of the association between pioglitazone use and bladder cancer.

Diabetic medicine : a journal of the British Diabetic Association·2018
Same journal

A Mixture of Distributed Lag Non-Linear Models to Account for Spatially Heterogeneous Exposure-Lag-Response Associations.

Statistics in medicine·2026
Same journal

Practical Considerations for Gaussian Process Modeling for Causal Inference in Quasi-Experimental Studies With Panel Data.

Statistics in medicine·2026
Same journal

Covariate Adjustment for Wilcoxon Two Sample Statistic and Test.

Statistics in medicine·2026
Same journal

Beyond Fixed Thresholds: Optimizing Summaries of Wearable Device Data via Piecewise Linearization of Quantile Functions.

Statistics in medicine·2026
Same journal

A Causal Framework for Evaluating the Total Effect of Strategies Aiming to Expand Screening and to Improve Outcomes.

Statistics in medicine·2026
Same journal

Causal Effects on Nonterminal Event Time With Application to Antibiotic Usage and Future Resistance.

Statistics in medicine·2026
See all related articles

This study compares two meta-analysis methods for 2x2 tables. Penalized quasi-likelihood (PQL) and weighted least squares performed adequately, but struggled with sparse data, with PQL showing better inference in some sparse scenarios.

Area of Science:

  • Biostatistics
  • Epidemiology
  • Statistical Modeling

Background:

  • Meta-analysis of 2x2 tables is crucial for synthesizing evidence from multiple studies.
  • Modeling the odds ratio using covariates and random effects accounts for study-level factors and heterogeneity.
  • Accurate estimation of regression coefficients and dispersion parameters is essential for reliable meta-analytic results.

Purpose of the Study:

  • To compare the performance of two statistical strategies for meta-analysis of 2x2 tables.
  • To evaluate the effectiveness of penalized quasi-likelihood (PQL) and weighted least squares (WLS) methods.
  • To identify conditions under which each method provides adequate or superior inference.

Main Methods:

  • Utilized penalized quasi-likelihood (PQL), an approximate inference technique for generalized linear mixed models.

Related Experiment Videos

  • Employed a linear model fitted by weighted least squares (WLS) to observed log-odds ratios.
  • Estimated regression coefficients and dispersion parameters for both methods.

    • Main Results:

    • Both PQL and WLS methods demonstrated adequate approximate inference across various conditions.
    • Neither method performed well when dealing with highly sparse data (small cell frequencies).
    • The PQL method showed improved inference under specific conditions with small cell frequencies.

    Conclusions:

    • Penalized quasi-likelihood and weighted least squares are viable methods for meta-analysis of 2x2 tables.
    • Caution is advised when applying these methods to highly sparse datasets.
    • Further research may be needed to refine methods for sparse data scenarios in meta-analysis.