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

Randomization-based methods for correcting for treatment changes: examples from the Concorde trial.

I R White1, A G Babiker, S Walker

  • 1Medical Statistics Unit, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, U.K.

Statistics in Medicine
|September 25, 1999
PubMed
Summary

New methods analyze clinical trial time-to-event data, adjusting for treatment changes. These causal models compare randomized groups, offering insights into treatment strategies like zidovudine use in HIV trials.

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

Kintsugi: A novel cause of urushiol-induced allergic contact dermatitis.

Contact dermatitis·2024
Same author

Analysis of exhaled breath to identify critically ill patients with ventilator-associated pneumonia.

Anaesthesia·2023
Same author

Position statement: The need for EU legislation to require disclosure and labelling of the composition of medical devices.

Journal of the European Academy of Dermatology and Venereology : JEADV·2021
Same author

Occupational dermatology in the time of the COVID-19 pandemic: a report of experience from London and Manchester, UK.

The British journal of dermatology·2020
Same author

Improved two-stage estimation to adjust for treatment switching in randomised trials: g-estimation to address time-dependent confounding.

Statistical methods in medical research·2020
Same author

The epidemic of methylisothiazolinone contact allergy in Europe: follow-up on changing exposures.

Journal of the European Academy of Dermatology and Venereology : JEADV·2019

Area of Science:

  • Biostatistics
  • Clinical Trials Methodology
  • Epidemiology

Background:

  • Time-to-event outcomes are common in clinical trials.
  • Treatment adherence and changes during follow-up can complicate analysis.
  • Existing methods may not adequately account for dynamic treatment regimens.

Purpose of the Study:

  • To develop and validate novel analysis methods for time-to-event clinical trial data.
  • To adjust for treatment modifications occurring after randomization.
  • To enable causal inference by comparing randomized groups, not selected subgroups.

Main Methods:

  • Utilized a causal modeling framework to relate observed to potential event times under alternative treatment scenarios.
  • Employed a semi-parametric approach based on Robins and Tsiatis, minimizing assumptions about treatment-prognosis relationships.

Related Experiment Videos

  • Applied methods to the Concorde trial data concerning zidovudine treatment for HIV.

    • Main Results:

    • Demonstrated a method to estimate treatment effects accounting for time-varying treatments.
    • Quantified the potential impact of adherence to deferred zidovudine strategy in the Concorde trial.
    • Explored the influence of model choice, non-constant treatment effects, and censoring on results.

    Conclusions:

    • The developed methods provide robust analysis for clinical trials with treatment changes.
    • Causal inference is maintained by focusing on randomized comparisons.
    • The approach is applicable to various time-to-event outcomes and dynamic treatment regimes.