Semiparametric accelerated intensity models for correlated recurrent and terminal events

Sangbum Choi, Xuelin Huang, Hyunsu Ju, Jing Ning

Research output: Contribution to journalArticle

Abstract

In clinical and epidemiological studies, recurrent events can arise when a subject repeatedly experiences the event of interest. Often, a terminal event such as death may preclude further occurrence of recurrent events in an informative manner such that the terminal event is strongly correlated with the recurrent event process. In this article, we propose a semiparametric joint analysis of correlated recurrent and terminal events. Specifically, we consider an accelerated intensity model for the recurrent events and an accelerated failure time model for the terminal event. We assess the dependency between the two event processes through a commonly used log-normal or gamma shared frailty. To estimate regression parameters and unspecified baseline intensity functions, we develop an EM algorithm with kernel smoothing adapted for both intensity functions, and perform variance estimation via numerical differentiation of the profile likelihoods. We evaluated the finite sample performance of the proposed method via simulation studies for both gamma and log-normal frailty models, and applied our method to the analysis of tumor recurrences and patient survival times in a soft tissue sarcoma study.

Original languageEnglish (US)
Pages (from-to)625-643
Number of pages19
JournalStatistica Sinica
Volume27
Issue number2
DOIs
StatePublished - Apr 1 2017

Fingerprint

Recurrent Events
Intensity Function
Model
Accelerated Failure Time Model
Kernel Smoothing
Numerical Differentiation
Profile Likelihood
Regression Estimate
Frailty Model
Frailty
Variance Estimation
Soft Tissue
Survival Time
EM Algorithm
Recurrence
Baseline
Tumor
Simulation Study

Keywords

  • Accelerated intensity regression
  • Frailty model
  • Informative censoring
  • Kernel smoothing
  • Nonparametric likelihood

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Semiparametric accelerated intensity models for correlated recurrent and terminal events. / Choi, Sangbum; Huang, Xuelin; Ju, Hyunsu; Ning, Jing.

In: Statistica Sinica, Vol. 27, No. 2, 01.04.2017, p. 625-643.

Research output: Contribution to journalArticle

Choi, Sangbum ; Huang, Xuelin ; Ju, Hyunsu ; Ning, Jing. / Semiparametric accelerated intensity models for correlated recurrent and terminal events. In: Statistica Sinica. 2017 ; Vol. 27, No. 2. pp. 625-643.
@article{6c5c661acbbb481dac8d9da15478cc88,
title = "Semiparametric accelerated intensity models for correlated recurrent and terminal events",
abstract = "In clinical and epidemiological studies, recurrent events can arise when a subject repeatedly experiences the event of interest. Often, a terminal event such as death may preclude further occurrence of recurrent events in an informative manner such that the terminal event is strongly correlated with the recurrent event process. In this article, we propose a semiparametric joint analysis of correlated recurrent and terminal events. Specifically, we consider an accelerated intensity model for the recurrent events and an accelerated failure time model for the terminal event. We assess the dependency between the two event processes through a commonly used log-normal or gamma shared frailty. To estimate regression parameters and unspecified baseline intensity functions, we develop an EM algorithm with kernel smoothing adapted for both intensity functions, and perform variance estimation via numerical differentiation of the profile likelihoods. We evaluated the finite sample performance of the proposed method via simulation studies for both gamma and log-normal frailty models, and applied our method to the analysis of tumor recurrences and patient survival times in a soft tissue sarcoma study.",
keywords = "Accelerated intensity regression, Frailty model, Informative censoring, Kernel smoothing, Nonparametric likelihood",
author = "Sangbum Choi and Xuelin Huang and Hyunsu Ju and Jing Ning",
year = "2017",
month = "4",
day = "1",
doi = "10.5705/ss.202015.0152",
language = "English (US)",
volume = "27",
pages = "625--643",
journal = "Statistica Sinica",
issn = "1017-0405",
publisher = "Institute of Statistical Science",
number = "2",

}

TY - JOUR

T1 - Semiparametric accelerated intensity models for correlated recurrent and terminal events

AU - Choi, Sangbum

AU - Huang, Xuelin

AU - Ju, Hyunsu

AU - Ning, Jing

PY - 2017/4/1

Y1 - 2017/4/1

N2 - In clinical and epidemiological studies, recurrent events can arise when a subject repeatedly experiences the event of interest. Often, a terminal event such as death may preclude further occurrence of recurrent events in an informative manner such that the terminal event is strongly correlated with the recurrent event process. In this article, we propose a semiparametric joint analysis of correlated recurrent and terminal events. Specifically, we consider an accelerated intensity model for the recurrent events and an accelerated failure time model for the terminal event. We assess the dependency between the two event processes through a commonly used log-normal or gamma shared frailty. To estimate regression parameters and unspecified baseline intensity functions, we develop an EM algorithm with kernel smoothing adapted for both intensity functions, and perform variance estimation via numerical differentiation of the profile likelihoods. We evaluated the finite sample performance of the proposed method via simulation studies for both gamma and log-normal frailty models, and applied our method to the analysis of tumor recurrences and patient survival times in a soft tissue sarcoma study.

AB - In clinical and epidemiological studies, recurrent events can arise when a subject repeatedly experiences the event of interest. Often, a terminal event such as death may preclude further occurrence of recurrent events in an informative manner such that the terminal event is strongly correlated with the recurrent event process. In this article, we propose a semiparametric joint analysis of correlated recurrent and terminal events. Specifically, we consider an accelerated intensity model for the recurrent events and an accelerated failure time model for the terminal event. We assess the dependency between the two event processes through a commonly used log-normal or gamma shared frailty. To estimate regression parameters and unspecified baseline intensity functions, we develop an EM algorithm with kernel smoothing adapted for both intensity functions, and perform variance estimation via numerical differentiation of the profile likelihoods. We evaluated the finite sample performance of the proposed method via simulation studies for both gamma and log-normal frailty models, and applied our method to the analysis of tumor recurrences and patient survival times in a soft tissue sarcoma study.

KW - Accelerated intensity regression

KW - Frailty model

KW - Informative censoring

KW - Kernel smoothing

KW - Nonparametric likelihood

UR - http://www.scopus.com/inward/record.url?scp=85016323083&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85016323083&partnerID=8YFLogxK

U2 - 10.5705/ss.202015.0152

DO - 10.5705/ss.202015.0152

M3 - Article

VL - 27

SP - 625

EP - 643

JO - Statistica Sinica

JF - Statistica Sinica

SN - 1017-0405

IS - 2

ER -