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

Joint models for efficient estimation in proportional hazards regression models.

Peter Slasor1, Nan Laird

  • 1Genzyme Corporation, One Kendall Square, Cambridge, MA 02139-1562, USA. peter.slasor@genzyme.com

Statistics in Medicine
|June 24, 2003
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

Real-world clinical outcomes of patients with CLN2 disease treated with cerliponase alfa.

Frontiers in neurology·2025
Same author

Exploring concurrent validity of the CLN2 Clinical Rating Scale: Comparison to PedsQL using cerliponase alfa clinical trial data.

PloS one·2024
Same author

Safety and efficacy of cerliponase alfa in children with neuronal ceroid lipofuscinosis type 2 (CLN2 disease): an open-label extension study.

The Lancet. Neurology·2023
Same author

Region-based analysis of rare genomic variants in whole-genome sequencing datasets reveal two novel Alzheimer's disease-associated genes: DTNB and DLG2.

Molecular psychiatry·2022
Same author

Clinical Pharmacokinetics and Pharmacodynamics of Cerliponase Alfa, Enzyme Replacement Therapy for CLN2 Disease by Intracerebroventricular Administration.

Clinical and translational science·2020
Same author

Identification of Novel Alzheimer's Disease Loci Using Sex-Specific Family-Based Association Analysis of Whole-Genome Sequence Data.

Scientific reports·2020

This study introduces joint models for survival and repeated measures data to improve efficiency and reduce bias in survival estimates. The proposed piecewise exponential model demonstrated modest efficiency gains in simulations and clinical trials.

Area of Science:

  • Biostatistics
  • Survival Analysis
  • Longitudinal Data Analysis

Background:

  • Censoring in survival studies leads to information loss.
  • Repeated measures data can recapture lost information and reduce bias.
  • Joint models integrate survival and longitudinal data for comprehensive analysis.

Purpose of the Study:

  • To develop efficient joint models for survival and repeated measurements.
  • To assess covariate effects on survival time using non-parametric models for longitudinal data.
  • To avoid bias from misspecification of the distribution for repeated measures.

Main Methods:

  • Utilized mixture models for joint analysis of survival (T) and repeated measurements (Y) given covariates (Z).
  • Modeled survival (T|Z) using a piecewise exponential distribution with proportional hazards.

Related Experiment Videos

  • Employed a multinomial model for the longitudinal component (Y|T,Z).
  • Maximized the joint likelihood using the Expectation-Maximization (EM) algorithm.

    • Main Results:

    • The joint piecewise exponential model demonstrated modest efficiency gains over standard methods.
    • Simulations showed an estimated efficiency gain of 6.4% compared to the standard proportional hazards model.
    • Clinical trial data analysis revealed an estimated efficiency gain of 10.2% over the standard proportional hazards model.

    Conclusions:

    • Joint models incorporating repeated measures data offer improved efficiency in survival analysis.
    • The proposed non-parametric approach for longitudinal data mitigates bias due to distributional misspecification.
    • This methodology provides a valuable tool for analyzing complex survival data with longitudinal components.