Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

rhDNase as an example of recurrent event analysis

T M Therneau¹, S A Hamilton

¹Mayo Clinic, Rochester, MI, USA.

Statistics in Medicine

|October 6, 1997

Summary

This study compares counting process methods for analyzing time-to-event data with multiple outcomes. The Anderson and Gill (AG) and marginal models are recommended for their robustness in complex scenarios.

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

Reproductive performance of early- and late-calving dairy cows artificially inseminated after ovulation synchronization and estrous resynchronization or artificially inseminated after observed estrus.

JDS communications·2022

Same author

Erratum to "Reproductive performance of early- and late-calving dairy cows artificially inseminated after ovulation synchronization and estrous resynchronization or artificially inseminated after observed estrus" (JDS Commun. 2:80-85).

JDS communications·2022

Same author

Securing access to personnel and materials in transboundary animal disease responses.

Revue scientifique et technique (International Office of Epizootics)·2020

Same author

Initial results from the New Horizons exploration of 2014 MU<sub>69</sub>, a small Kuiper Belt object.

Science (New York, N.Y.)·2019

Same author

Prevalence of myeloma precursor state monoclonal gammopathy of undetermined significance in 12372 individuals 10-49 years old: a population-based study from the National Health and Nutrition Examination Survey.

Blood cancer journal·2017

Same author

Factors predictive of survival and estimated years of life lost in the decade following nontraumatic and traumatic spinal cord injury.

Spinal cord·2017

Area of Science:

Biostatistics
Survival Analysis
Statistical Modeling

Background:

Analyzing time-to-event data with multiple or recurrent outcomes presents unique statistical challenges.
Existing methods, including those by Anderson and Gill (AG), Wei, Lin, and Weissfeld (WFW), and Prentice, Williams, and Peterson (PWP), offer different approaches.

Purpose of the Study:

To compare the performance and applicability of established counting process methods for analyzing complex time-to-event data.
To provide practical guidance on implementing these methods using common statistical software.
To highlight the strengths and limitations of each approach through real-world data examples.

Main Methods:

Application of counting process models, specifically Anderson and Gill (AG), Wei, Lin, and Weissfeld (WFW), and Prentice, Williams, and Peterson (PWP).
Comparative analysis using three distinct datasets: a simulated dataset with a hidden covariate, a gamma interferon trial dataset, and a dataset with multiple events and discontinuous at-risk intervals.
Implementation details demonstrated using popular statistical software.

Main Results:

The three methods yielded dissimilar results for the dataset with multiple events and discontinuous at-risk intervals, indicating sensitivity to data structure.
The simulated and gamma interferon datasets showed comparable behaviors, illustrating method performance under specific conditions.
Strengths and pitfalls of each method were identified through practical data analysis.

Conclusions:

The Anderson and Gill (AG) and marginal models are recommended for analyzing time-to-event data with multiple or recurrent outcomes, especially in complex situations.
Careful consideration of the chosen statistical method is crucial, as different approaches can lead to divergent conclusions.
Practical implementation guidance is provided for researchers analyzing similar data types.

Related Experiment Videos