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

Bayesian inference for randomized clinical trials with treatment failures.

Michele L Shaffer1, Vernon M Chinchilli

  • 1Department of Health Evaluation Sciences, A210, Penn State College of Medicine, 600 Centerview Drive, Suite 2200, Hershey, PA 17033-0855, U.S.A. mshaffer@hes.hmc.psu.edu

Statistics in Medicine
|April 15, 2004
PubMed
Summary

This study introduces a Bayesian approach to handle rescue medication bias in clinical trials. It offers a supplemental analysis method to improve data interpretation when patients receive additional treatments.

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

Evolving roles of Data Coordinating Centers in multisite research: Challenges and adaptations from a rapid scoping review.

Journal of clinical and translational science·2026
Same author

Correction: Working to Increase Stability through Exercise (WISE): screening, recruitment, and baseline characteristics.

Trials·2026
Same author

Postanesthesia Apnea in Former Preterm and Term Infants: A Qualitative Systematic Review and Meta-analysis.

Anesthesiology·2026
Same author

Multivariate equivalence and component-wise superiority tests for paired samples.

Journal of biopharmaceutical statistics·2026
Same author

Semaglutide, weight loss, and cardiovascular outcomes in the SELECT trial.

Lancet (London, England)·2026
Same author

Hyperglycemia During Acute Pancreatitis and Progression to Early-Onset Diabetes After Recovery: Preliminary Findings From the Diabetes Related to Acute Pancreatitis and Its Mechanisms (DREAM) Study.

Diabetes care·2026

Area of Science:

  • Biostatistics
  • Clinical Trial Methodology
  • Pharmacoeconomics

Background:

  • Clinical trials require ethical administration of rescue medications for treatment failures.
  • Rescue medications can introduce bias into response measurements, particularly in control groups.
  • Standard intent-to-treat (ITT) analysis may not fully account for rescue medication effects.

Purpose of the Study:

  • To propose a Bayesian, counterfactual approach for supplemental analysis of clinical trial data.
  • To address the bias introduced by rescue medications in clinical trial outcomes.
  • To compare the proposed Bayesian method with a likelihood-based approach.

Main Methods:

  • Utilized a Bayesian, counterfactual framework.
  • Employed the data augmentation (DA) algorithm for supplemental analysis.

Related Experiment Videos

  • Conducted a simulation study to compare performance against an EM algorithm-based approach.
  • Illustrated the method with a case study from the Asthma Clinical Research Network (ACRN).

    • Main Results:

    • The Bayesian data augmentation approach provides a viable method for supplemental analysis.
    • Simulation results indicate the operating characteristics of the Bayesian procedure.
    • The ACRN example demonstrates practical application in handling rescue medication bias.

    Conclusions:

    • The proposed Bayesian, counterfactual approach offers a valuable tool for supplemental analysis in clinical trials.
    • This method can help mitigate bias from rescue medications, enhancing data integrity.
    • The study provides statistical evidence supporting the use of Bayesian methods in complex clinical trial scenarios.