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

Analysis of longitudinal marginal structural models.

Jenny Bryan1, Zhuo Yu, Mark J Van Der Laan

  • 1Statistics Department and Biotechnology Lab., University of British Columbia, 333-6356 Agricultural Road, Vancouver, BC V6T 1Z2, Canada. jenny@stat.ubc.ca

Biostatistics (Oxford, England)
|June 23, 2004
PubMed
Summary
This summary is machine-generated.

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

An approach to nonparametric inference on the causal dose-response function.

Journal of causal inference·2026
Same author

Deep learning-based differentiation of non-tuberculous mycobacterial lung disease and pulmonary tuberculosis using chest CT.

Frontiers in medicine·2026
Same author

SIRT3 deSUMOylation sustains hematopoietic stem cell activities under stressful conditions via the DHX58-IRF7 axis.

Haematologica·2026
Same author

Sequential invitations to FOBT screening and colorectal cancer incidence.

Scientific reports·2026
Same author

BCAA catabolism mediates POU2AF1 propionylation to enhance T-ALL development.

Cellular oncology (Dordrecht, Netherlands)·2026
Same author

Powering RCTs for Marginal Effects With GLMs Using Prognostic Score Adjustment.

Statistics in medicine·2026

This study introduces novel estimators for time-dependent treatment effects on survival in longitudinal data. These methods, including inverse probability of treatment weighting, improve causal effect estimation, even with informative censoring.

Area of Science:

  • Biostatistics
  • Epidemiology
  • Longitudinal Data Analysis

Background:

  • Estimating causal effects of time-dependent treatments on survival is crucial in longitudinal studies.
  • Standard methods can be biased by time-dependent confounding and informative censoring.

Purpose of the Study:

  • To construct and study novel estimators for the causal effect of time-dependent treatments on survival.
  • To extend existing marginal structural models (MSM) to handle informative censoring.
  • To apply these methods to estimate the effect of exercise on mortality in seniors.

Main Methods:

  • Utilized a marginal structural model (MSM) approach.
  • Employed inverse probability of treatment weighted (IPTW) estimators.
  • Developed an improved, one-step estimator for enhanced consistency and asymptotic linearity.

Related Experiment Videos

  • Extended methodology to address informative censoring.

    • Main Results:

    • Simulation studies demonstrated bias in naive estimators with time-dependent confounders.
    • The IPTW estimator showed efficiency gains, even without confounding.
    • The improved one-step estimator demonstrated further efficiency gains.
    • Applied methodology to estimate exercise's causal effect on mortality in a senior cohort.

    Conclusions:

    • The proposed estimators provide robust methods for assessing time-dependent treatment effects on survival.
    • The IPTW and improved one-step estimators offer significant advantages over naive approaches.
    • The methodology is effective for analyzing complex longitudinal data with potential confounding and censoring issues.