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

Group sequential comparison of changes: ad-hoc versus more exact method

J W Lee1, D L DeMets

  • 1Department of Statistics, Korea University, Seoul.

Biometrics
|March 1, 1995
PubMed
Summary

This study compares two methods for analyzing sequential clinical trial data. The more exact group sequential method, based on linear mixed effects models, is shown to be robust and performs better than the simpler ad-hoc approach.

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

Statistical issues in interpreting clinical trials.

Journal of internal medicine·2004
Same author

Considerations in the evaluation of surrogate endpoints in clinical trials. summary of a National Institutes of Health workshop.

Controlled clinical trials·2001
Same author

Statistical issues relating to international differences in clinical trials.

American heart journal·2001
Same author

Effect of carvedilol on survival in severe chronic heart failure.

The New England journal of medicine·2001
Same author

The Data and Safety Monitoring Committee: Some final thoughts.

American heart journal·2001
Same author

Synthesis of 2-(allylthio)pyrazines as a novel cancer chemopreventive agent.

Archives of pharmacal research·2001

Area of Science:

  • Biostatistics
  • Clinical Trials
  • Statistical Methods

Background:

  • Sequential clinical trials often compare changes in responses between treatment groups.
  • Existing ad-hoc methods for interim analysis may violate assumptions of independent increments, particularly with repeated measures.
  • Simplicity of ad-hoc methods has led to their use despite theoretical limitations.

Purpose of the Study:

  • To evaluate the robustness of a more exact group sequential method for comparing rates of change.
  • To compare the performance of the exact method against a commonly used ad-hoc approach.
  • To investigate the relationship between Fisher information and surrogate information in sequential analyses.

Main Methods:

  • Utilized linear mixed effects models to derive the asymptotic joint distribution of sequentially computed statistics.

Related Experiment Videos

  • Conducted simulation studies to assess method robustness and compare performance.
  • Examined the behavior of both exact and ad-hoc methods under various scenarios.

    • Main Results:

    • The more exact group sequential method demonstrated robustness to violations of typical assumptions.
    • Simulation results indicated superior performance of the exact method compared to the ad-hoc approach.
    • Insights were gained into the interplay between different information time measures.

    Conclusions:

    • The exact group sequential method provides a more reliable approach for analyzing sequential clinical trial data with repeated measures.
    • The ad-hoc method, while simple, may lead to inaccurate conclusions due to unmet assumptions.
    • Further research into information times can refine sequential analysis techniques.