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

Estimating risk difference in multicenter studies under baseline-risk heterogeneity.

D Böhning1, J Sarol

  • 1Department of Epidemiology, Free University Berlin, Germany. boehning@zedat.fu-berlin.de

Biometrics
|April 28, 2000
PubMed
Summary

This study addresses bias in estimating risk differences across multiple centers. A new Mantel-Haenszel type estimator is proposed, offering unbiased results and improved small-sample performance for multicenter studies.

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

Assessing injury risk in male and female Royal Navy recruits: does the Functional Movement Screen provide understanding to inform effective injury mitigation?

BMJ military health·2023
Same author

Sensitivity of contact-tracing for COVID-19 in Thailand: a capture-recapture application.

BMC infectious diseases·2022
Same author

Correction on Extending the Lincoln-Petersen estimator for multiple identifications in one source. Statistics in Medicine 2014; 33:4237-4249.

Statistics in medicine·2017
Same author

Erratum to: "Youden's Index and the Likelihood Ratio Positive in Diagnostic Testing".

Methods of information in medicine·2015
Same author

Youden's Index and the Likelihood Ratio Positive in Diagnostic Testing.

Methods of information in medicine·2015
Same author

Extending the Lincoln-Petersen estimator for multiple identifications in one source.

Statistics in medicine·2014

Area of Science:

  • Biostatistics
  • Epidemiology
  • Clinical Trials

Background:

  • Estimating common risk difference in multicenter studies is crucial for synthesizing evidence.
  • Baseline heterogeneity across study centers can introduce bias in pooled estimates.
  • Existing methods may suffer from bias when using estimated weights.

Purpose of the Study:

  • To investigate bias in optimally weighted estimators for common risk difference in multicenter studies.
  • To propose a novel, unbiased estimator for the common risk difference in the presence of heterogeneity.
  • To evaluate the performance of the proposed estimator, particularly for small sample sizes within centers.

Main Methods:

  • Consideration of the optimally weighted estimator for common risk difference.

Related Experiment Videos

  • Analysis of bias arising from using sample estimates of true weights.
  • Development and proposal of a Mantel-Haenszel type estimator.
  • Monte Carlo simulations to compare estimator performance.

    • Main Results:

    • The optimally weighted estimator demonstrates considerable bias when sample estimates replace true weights.
    • The proposed Mantel-Haenszel type estimator is shown to be unbiased.
    • The new estimator exhibits smaller variance for small sample sizes within centers compared to alternatives.

    Conclusions:

    • Standard optimally weighted estimators can be biased in multicenter settings due to estimated weights.
    • The proposed Mantel-Haenszel type estimator provides a robust and efficient alternative for risk difference estimation.
    • This new method is particularly advantageous in studies with limited sample sizes at individual centers.