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

Bootstrap analysis of multivariate failure time data.

Jane Monaco1, Jianwen Cai, James Grizzle

  • 1Department of Biostatistics, The University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-7420, USA. jmonaco@bios.unc.edu

Statistics in Medicine
|October 21, 2005
PubMed
Summary

The bootstrap method effectively estimates standard errors for survival probabilities and treatment effects in correlated multivariate failure time data. This approach is valuable for analyzing complex clinical trial outcomes, like those in the ACAS study.

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

A more interpretable regression model for count data with excess of zeros.

Statistical methods in medical research·2026
Same author

Association of lipoprotein(a) and peripheral artery disease: Findings from The hispanic community health study/study of latinos (HCHS/SOL).

American journal of preventive cardiology·2026
Same author

Non-Vigorous Physical Activity Associated with Reduced Hospitalization Risk with or without Diabetes or Peripheral Artery Disease: Study of Latinos.

Medicine and science in sports and exercise·2026
Same author

Visual Impairment and Cardiovascular Risk Factors in Hispanic and Latino Adults.

JAMA network open·2026
Same author

Non-genetic component of height as a surrogate marker for childhood socioeconomic position and its association with cardiovascular and brain health: results from HCHS/SOL.

medRxiv : the preprint server for health sciences·2026
Same author

Intergenerational Socioeconomic Mobility and Adulthood All-Cause Mortality in the Hispanic Community Health Study/Study of Latinos.

American journal of epidemiology·2026

Area of Science:

  • Biostatistics
  • Survival Analysis
  • Clinical Trials

Background:

  • Multivariate failure time data are common in research, but standard Cox proportional hazards models are invalid for correlated observations.
  • Existing methods for correlated failure times often average treatment effects, which may obscure time-specific findings.

Purpose of the Study:

  • To evaluate the bootstrap method for estimating standard errors in multivariate failure time data.
  • To specifically assess its utility for analyzing survival probabilities or treatment effects at a single time point, relevant to surgical trials.

Main Methods:

  • Utilized the bootstrap resampling technique to derive standard errors for correlated failure time data.
  • Conducted extensive simulation studies under various conditions (treatment effects, cluster sizes, correlations, proportional/non-proportional hazards).

Related Experiment Videos

  • Applied the method to analyze correlated carotid artery outcomes from the Asymptomatic Carotid and Atherosclerosis Study (ACAS).

    • Main Results:

    • The bootstrap method accurately estimated standard errors for survival probabilities at specific time points.
    • It also provided adequate standard error estimates for survival differences and relative risks at a given time.
    • Demonstrated successful application for standard error calculation and statistical testing at a specific time point using ACAS data.

    Conclusions:

    • The bootstrap method is a reliable approach for estimating standard errors in multivariate failure time analyses with correlated data.
    • It is particularly useful when the research focus is on survival outcomes or treatment effects at a specific time point.
    • This technique offers a valid alternative to traditional methods when dealing with correlated failure time data in clinical research.