Bivariate frailty model for the analysis of multivariate survival time

Xiaonan (Nan) Xue, Ron Brookmeyer

Research output: Contribution to journalArticle

65 Citations (Scopus)

Abstract

Because of limitations of the univariate frailty model in analysis of multivariate survival data, a bivariate frailty model is introduced for the analysis of bivariate survival data. This provides tremendous flexibility especially in allowing negative associations between subjects within the same cluster. The approach involves incorporating into the model two possibly correlated frailties for each cluster. The bivariate lognormal distribution is used as the frailty distribution. The model is then generalized to multivariate survival data with two distinguished groups and also to alternating process data. A modified EM algorithm is developed with no requirement of specification of the baseline hazards. The estimators are generalized maximum likelihood estimators with subject-specific interpretation. The model is applied to a mental health study on evaluation of health policy effects for inpatient psychiatric care.

Original languageEnglish (US)
Pages (from-to)277-289
Number of pages13
JournalLifetime Data Analysis
Volume2
Issue number3
StatePublished - 1996
Externally publishedYes

Fingerprint

Frailty Model
Survival Time
Survival Analysis
Multivariate Survival Data
Frailty
Multivariate Analysis
Health
Health Policy
Negative Association
Psychiatry
Inpatients
Mental Health
Bivariate Distribution
Log Normal Distribution
Survival Data
EM Algorithm
Hazard
Maximum Likelihood Estimator
Univariate
Baseline

Keywords

  • Frailty
  • Multivariate Survival Data
  • Subject-specific Parameter

ASJC Scopus subject areas

  • Applied Mathematics
  • Medicine(all)

Cite this

Bivariate frailty model for the analysis of multivariate survival time. / Xue, Xiaonan (Nan); Brookmeyer, Ron.

In: Lifetime Data Analysis, Vol. 2, No. 3, 1996, p. 277-289.

Research output: Contribution to journalArticle

@article{e37f5f978e12432485fe4cb23c3c80ec,
title = "Bivariate frailty model for the analysis of multivariate survival time",
abstract = "Because of limitations of the univariate frailty model in analysis of multivariate survival data, a bivariate frailty model is introduced for the analysis of bivariate survival data. This provides tremendous flexibility especially in allowing negative associations between subjects within the same cluster. The approach involves incorporating into the model two possibly correlated frailties for each cluster. The bivariate lognormal distribution is used as the frailty distribution. The model is then generalized to multivariate survival data with two distinguished groups and also to alternating process data. A modified EM algorithm is developed with no requirement of specification of the baseline hazards. The estimators are generalized maximum likelihood estimators with subject-specific interpretation. The model is applied to a mental health study on evaluation of health policy effects for inpatient psychiatric care.",
keywords = "Frailty, Multivariate Survival Data, Subject-specific Parameter",
author = "Xue, {Xiaonan (Nan)} and Ron Brookmeyer",
year = "1996",
language = "English (US)",
volume = "2",
pages = "277--289",
journal = "Lifetime Data Analysis",
issn = "1380-7870",
publisher = "Springer Netherlands",
number = "3",

}

TY - JOUR

T1 - Bivariate frailty model for the analysis of multivariate survival time

AU - Xue, Xiaonan (Nan)

AU - Brookmeyer, Ron

PY - 1996

Y1 - 1996

N2 - Because of limitations of the univariate frailty model in analysis of multivariate survival data, a bivariate frailty model is introduced for the analysis of bivariate survival data. This provides tremendous flexibility especially in allowing negative associations between subjects within the same cluster. The approach involves incorporating into the model two possibly correlated frailties for each cluster. The bivariate lognormal distribution is used as the frailty distribution. The model is then generalized to multivariate survival data with two distinguished groups and also to alternating process data. A modified EM algorithm is developed with no requirement of specification of the baseline hazards. The estimators are generalized maximum likelihood estimators with subject-specific interpretation. The model is applied to a mental health study on evaluation of health policy effects for inpatient psychiatric care.

AB - Because of limitations of the univariate frailty model in analysis of multivariate survival data, a bivariate frailty model is introduced for the analysis of bivariate survival data. This provides tremendous flexibility especially in allowing negative associations between subjects within the same cluster. The approach involves incorporating into the model two possibly correlated frailties for each cluster. The bivariate lognormal distribution is used as the frailty distribution. The model is then generalized to multivariate survival data with two distinguished groups and also to alternating process data. A modified EM algorithm is developed with no requirement of specification of the baseline hazards. The estimators are generalized maximum likelihood estimators with subject-specific interpretation. The model is applied to a mental health study on evaluation of health policy effects for inpatient psychiatric care.

KW - Frailty

KW - Multivariate Survival Data

KW - Subject-specific Parameter

UR - http://www.scopus.com/inward/record.url?scp=0030323029&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030323029&partnerID=8YFLogxK

M3 - Article

C2 - 9384637

AN - SCOPUS:0030323029

VL - 2

SP - 277

EP - 289

JO - Lifetime Data Analysis

JF - Lifetime Data Analysis

SN - 1380-7870

IS - 3

ER -