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 relative risks: a tree-based approach.

R Xu1, S Adak

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

Methods of Information in Medicine
|June 27, 2001
PubMed
Summary

This study introduces a novel tree-based method to analyze how patient characteristics impact survival over time. The approach models time-varying risks as piecewise constants, enhancing clinical interpretability.

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

[In situ recurrence and sacral metastasis after surgery: a case of adrenal pheochromocytoma].

Zhonghua nei ke za zhi·2022
Same author

Impaired ascending aortic elasticity in fetuses with tetralogy of Fallot.

Ultrasound in obstetrics & gynecology : the official journal of the International Society of Ultrasound in Obstetrics and Gynecology·2022
Same author

Finite element analysis of optimal design of distal geometry of cementless femoral prosthesis.

Nigerian journal of clinical practice·2022
Same author

Observation of WWW Production in pp Collisions at sqrt[s]=13  TeV with the ATLAS Detector.

Physical review letters·2022
Same author

[An overview of the establishment of cell lines from sinonasal malignancies].

Zhonghua er bi yan hou tou jing wai ke za zhi = Chinese journal of otorhinolaryngology head and neck surgery·2022
Same author

Hypertriton Production in p-Pb Collisions at sqrt[s_{NN}]=5.02  TeV.

Physical review letters·2022

Area of Science:

  • Biostatistics
  • Survival Analysis
  • Machine Learning in Healthcare

Background:

  • Estimating time-varying effects of patient characteristics on survival is crucial for clinical decision-making.
  • Existing continuous time-varying models (e.g., splines, loess) can be complex for clinical interpretation.
  • There is a need for interpretable methods to model dynamic survival predictions.

Purpose of the Study:

  • To introduce a novel tree-based method for estimating time-varying effects on survival.
  • To model time-varying relative risks as piecewise constants for improved interpretability.
  • To provide a computationally efficient and clinically relevant approach to survival analysis.

Main Methods:

  • Utilized a Cox-type model for censored survival data.

Related Experiment Videos

  • Developed a three-step tree-based algorithm: tree growing, pruning, and final tree selection.
  • Employed a fast algorithm with maximized score statistics for change point detection.
  • Incorporated bootstrap resampling or explained variation for model selection.

    • Main Results:

    • The proposed tree-based method effectively estimates time-varying effects of baseline characteristics on survival.
    • Piecewise constant modeling of relative risks enhances the clinical interpretability of regression parameters.
    • The method offers a more parsimonious and understandable alternative to continuous time-varying models.

    Conclusions:

    • The introduced tree-based method provides a valuable tool for analyzing time-varying survival data.
    • Its piecewise constant approach significantly improves the clinical interpretability of survival models.
    • This method offers a practical and efficient alternative for researchers and clinicians in survival analysis.