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 Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Improved Speech Recognition with Automated ForwardFocus in Cochlear Kanso 3 Sound Processor Users.

Otology & neurotology open·2026
Same author

Time to achieve response and depth of maximal tumor response as potential surrogates for overall survival in advanced non-small cell lung cancer.

Future oncology (London, England)·2026
Same author

Target trial emulation of early docetaxel and enzalutamide for metastatic hormone-sensitive prostate cancer.

BJU international·2025
Same author

Randomised controlled trial to assess the efficiency and effectiveness of remote care compared with in-clinic care for adult cochlear implant recipients in the first 12 months after activation: a protocol of the INSPIRE study.

BMJ open·2025
Same author

INTEGRATE pooled phase 2/3 results are robust to postprogression switching and the winner's curse.

JNCI cancer spectrum·2025
Same author

Phenotypic characterization of ENPP1 deficiency: generalized arterial calcification of infancy and autosomal recessive hypophosphatemic rickets type 2.

JBMR plus·2025

Related Experiment Video

Updated: Aug 31, 2025

Integrating Computerized Linguistic and Social Network Analyses to Capture Addiction Recovery Capital in an Online Community
08:53

Integrating Computerized Linguistic and Social Network Analyses to Capture Addiction Recovery Capital in an Online Community

Published on: May 31, 2019

5.2K

Analysis of adaptive platform trials using a network approach.

Ian C Marschner1, I Manjula Schou1

  • 1NHMRC Clinical Trials Centre, University of Sydney, Sydney, NSW, Australia.

Clinical Trials (London, England)
|August 22, 2022
PubMed
Summary

Adaptive platform trials can be biased by time trends. Network meta-analysis offers a framework to separate direct and indirect evidence, improving bias assessment for these complex clinical trials.

Keywords:
Adaptive designindirect treatment comparisonmixed treatment comparisonmulti-arm multi-stage studynetwork meta-analysisplatform trial

More Related Videos

The Innovation Arena: A Method for Comparing Innovative Problem-Solving Across Groups
14:14

The Innovation Arena: A Method for Comparing Innovative Problem-Solving Across Groups

Published on: May 13, 2022

6.0K
Assessment of Social Cognition in Non-human Primates Using a Network of Computerized Automated Learning Device ALDM Test Systems
08:42

Assessment of Social Cognition in Non-human Primates Using a Network of Computerized Automated Learning Device ALDM Test Systems

Published on: May 5, 2015

12.2K

Related Experiment Videos

Last Updated: Aug 31, 2025

Integrating Computerized Linguistic and Social Network Analyses to Capture Addiction Recovery Capital in an Online Community
08:53

Integrating Computerized Linguistic and Social Network Analyses to Capture Addiction Recovery Capital in an Online Community

Published on: May 31, 2019

5.2K
The Innovation Arena: A Method for Comparing Innovative Problem-Solving Across Groups
14:14

The Innovation Arena: A Method for Comparing Innovative Problem-Solving Across Groups

Published on: May 13, 2022

6.0K
Assessment of Social Cognition in Non-human Primates Using a Network of Computerized Automated Learning Device ALDM Test Systems
08:42

Assessment of Social Cognition in Non-human Primates Using a Network of Computerized Automated Learning Device ALDM Test Systems

Published on: May 5, 2015

12.2K

Area of Science:

  • Clinical trial methodology
  • Biostatistics
  • Epidemiology

Background:

  • Adaptive platform trials offer flexible, multi-treatment comparisons within a common infrastructure.
  • Criticism exists regarding potential bias from time trends in population risk, confounding study design with risk levels.
  • Non-concurrent controls, where treatments aren't randomized simultaneously with controls, can introduce bias in adaptive platform trials.

Purpose of the Study:

  • To evaluate bias in adaptive platform trials due to time trends and non-concurrent controls.
  • To compare bias-mitigation strategies: stratification and adjustment.
  • To propose a unified framework using network meta-analysis for analyzing platform trial data.

Main Methods:

  • Network meta-analysis principles were adapted to model platform trials as networks of direct randomized and indirect non-randomized comparisons.
  • Two analysis methods, stratification (using only concurrent controls) and adjustment (using time-period modeling), were compared within this framework.
  • Simulations and a case study from the STAMPEDE platform trial were analyzed using the netmeta R package.

Main Results:

  • Unadjusted methods showed bias in simulations with time trends; both adjustment and stratification were unbiased when direct and indirect evidence were consistent.
  • Network tests identified inconsistency between direct and indirect evidence in the STAMPEDE trial, supporting the use of only direct comparisons.
  • The network meta-analysis framework transparently separated direct randomized evidence from indirect non-randomized evidence.

Conclusions:

  • Network meta-analysis provides a robust methodology for analyzing complex adaptive platform trials, enabling clear assessment of non-concurrent control impact.
  • Time-stratified analysis of concurrently controlled comparisons is recommended for primary analyses.
  • Time-adjusted analyses incorporating non-concurrent controls are suitable for secondary analyses, with network analysis as a valuable supplement.