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

Survival analysis with time-varying regression effects using a tree-based approach.

Ronghui Xu1, Sudeshna Adak

  • 1Department of Biostatistics, Harvard School of Public Health and Dana-Farber Cancer Institute, Boston, Massachusetts 02115, USA. rxu@jimmy.harvard.edu

Biometrics
|June 20, 2002
PubMed
Summary

This study introduces a novel tree-based method for survival analysis, effectively estimating time-dependent effects in non-Hodgkin

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

Prior exposure to advanced therapy and timing of discontinuation and risk of serious infections in patients with inflammatory bowel disease initiating a new advanced therapy.

Journal of Crohn's & colitis·2026
Same author

Adjunctive GLP1 Receptor Agonists in Patients With Inflammatory Bowel Diseases and Obesity and/or Diabetes: A Target Trial Emulation.

Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association·2026
Same author

Treatment and Outcomes of Crohn's Disease and Ulcerative Colitis in Newly Diagnosed Adults in the United States, 2007 to 2023.

Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association·2026
Same author

Comparative Safety of Advanced Therapies for Crohn Disease.

JAMA network open·2026
Same author

A Response to the Letter to the Editor: Refining Risk and Redefining Stage: Hidden Implications of BAP1-Mutant Mesothelioma Shiuan-Chih Chen, MD, PhD; Yuan-Ti Lee, MD, PhD.

Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer·2025
Same author

High-precision fiber optic temperature monitoring system based on FBG-GaAs dual-mode collaborative measurement.

Optics express·2025

Area of Science:

  • Biostatistics
  • Survival Analysis
  • Machine Learning

Background:

  • Nonproportional hazards are common in survival analysis, notably in Non-Hodgkin's Lymphoma data.
  • Existing methods may not fully capture time-varying regression effects.

Purpose of the Study:

  • To develop a tree-based method for estimating time-dependent regression effects in survival analysis.
  • To identify change points in regression parameters for Cox-type models.
  • To improve clinical interpretability of prognostic factors.

Main Methods:

  • A tree-based approach approximating time-varying effects as piecewise constants.
  • A fast algorithm using maximized score statistics for recursive time-axis segmentation.
  • A pruning algorithm similar to CART for sparse segmentation.

Related Experiment Videos

  • Bootstrap resampling to correct for overoptimism.

    • Main Results:

    • The method effectively estimates piecewise constant, time-dependent regression effects.
    • Identified change points in regression parameters for Non-Hodgkin's Lymphoma data.
    • Developed a time-varying risk index from the International Risk Index.

    Conclusions:

    • The tree-based method provides a clinically interpretable alternative to spline models for time-varying effects.
    • The developed algorithm enhances survival data analysis, particularly for complex datasets like lymphoma.
    • The time-varying risk index offers a dynamic prognostic tool for Non-Hodgkin's Lymphoma.