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

An analytic method for randomized trials with informative censoring: Part 1

J M Robins1

  • 1Department of Epidemiology, Harvard School of Public Health, Boston, Massachusetts 02115, USA.

Lifetime Data Analysis
|January 1, 1995
PubMed
Summary

This study introduces new statistical methods to test treatment effects in clinical trials with informative censoring, ensuring accurate results even when patient follow-up is compromised. These methods provide reliable analysis for disease progression and repeated measures outcomes.

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

On the asymptotic validity of confidence sets for linear functionals of solutions to integral equations.

Biometrika·2025
Same author

Using negative controls to identify causal effects with invalid instrumental variables.

Biometrika·2025
Same author

Multiple robustness in factorized likelihood models.

Biometrika·2018
Same author

Instrumental variables as bias amplifiers with general outcome and confounding.

Biometrika·2017
Same author

Marginal Mean Models for Dynamic Regimes.

Journal of the American Statistical Association·2009
Same author

Causal models for estimating the effects of weight gain on mortality.

International journal of obesity (2005)·2008

Area of Science:

  • Biostatistics
  • Clinical Trial Methodology
  • Epidemiology

Background:

  • Informative censoring, where censoring time depends on failure time, complicates analysis of time-to-event data in clinical trials.
  • Standard methods like the weighted log-rank test may yield biased results when censoring is informative.
  • Accurate estimation of treatment effects is crucial for evaluating disease progression and intervention efficacy.

Purpose of the Study:

  • To propose two non-identifiable assumptions enabling testing and estimation of treatment effects under informative censoring.
  • To develop asymptotically distribution-free alpha-level tests that remain valid even with informative censoring.
  • To extend these methods for analyzing repeated measures outcomes, such as CD-4 counts in HIV trials.

Main Methods:

Related Experiment Videos

  • Specification of two novel non-identifiable assumptions for handling informative censoring.
  • Development of novel hypothesis tests that are asymptotically distribution-free under these assumptions.
  • Comparison of the proposed tests' validity against standard weighted log-rank tests in scenarios with informative censoring.

Main Results:

  • The proposed tests maintain asymptotic distribution-free alpha-level properties under informative censoring when either of the specified assumptions holds.
  • Weighted log-rank tests are only valid under informative censoring if the censoring distribution is identical across treatment arms and one of the assumptions is met.
  • The methodology is adaptable for analyzing longitudinal data, such as changes in CD-4 counts over time.

Conclusions:

  • The developed statistical framework effectively addresses the challenges posed by informative censoring in randomized trials.
  • These methods offer a robust approach to testing and estimating treatment effects, improving the reliability of clinical trial findings.
  • The extension to repeated measures outcomes broadens the applicability of these statistical advancements in medical research.