Bivariate random change point models for longitudinal outcomes

Lili Yang, Sujuan Gao

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Epidemiologic and clinical studies routinely collect longitudinal measures of multiple outcomes, including biomarker measures, cognitive functions, and clinical symptoms. These longitudinal outcomes can be used to establish the temporal order of relevant biological processes and their association with the onset of clinical symptoms. Univariate change point models have been used to model various clinical endpoints, suchasCD4 count in studying the progression of HIV infection and cognitive function in the elderly. We propose to use bivariate change point models for two longitudinal outcomes with a focus on the correlation betweenthetwo change points. We consider three types of change point models in the bivariate model setting: the broken-stick model, the Bacon-Watts model, and the smooth polynomial model. We adopt a Bayesianapproachusing a Markov chain Monte Carlo sampling method for parameter estimation and inference. We assess the proposed methods in simulation studies and demonstrate the methodology using data from a longitudinal study of dementia.

Original languageEnglish
Pages (from-to)1038-1053
Number of pages16
JournalStatistics in Medicine
Volume32
Issue number6
DOIs
StatePublished - Mar 15 2013

Fingerprint

Change-point Model
Cognition
Biological Phenomena
Monte Carlo Method
Markov Chains
Statistical Models
HIV Infections
Longitudinal Studies
Dementia
Epidemiologic Studies
Biomarkers
Outcome Assessment (Health Care)
Multiple Outcomes
HIV Infection
Polynomial Model
Monte Carlo Sampling
Longitudinal Study
Change Point
Sampling Methods
Markov Chain Monte Carlo

Keywords

  • Bayesian method
  • Longitudinal bivariate outcomes
  • Random change point model

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Bivariate random change point models for longitudinal outcomes. / Yang, Lili; Gao, Sujuan.

In: Statistics in Medicine, Vol. 32, No. 6, 15.03.2013, p. 1038-1053.

Research output: Contribution to journalArticle

@article{4d6c935b26f548b28641844d1fff1b11,
title = "Bivariate random change point models for longitudinal outcomes",
abstract = "Epidemiologic and clinical studies routinely collect longitudinal measures of multiple outcomes, including biomarker measures, cognitive functions, and clinical symptoms. These longitudinal outcomes can be used to establish the temporal order of relevant biological processes and their association with the onset of clinical symptoms. Univariate change point models have been used to model various clinical endpoints, suchasCD4 count in studying the progression of HIV infection and cognitive function in the elderly. We propose to use bivariate change point models for two longitudinal outcomes with a focus on the correlation betweenthetwo change points. We consider three types of change point models in the bivariate model setting: the broken-stick model, the Bacon-Watts model, and the smooth polynomial model. We adopt a Bayesianapproachusing a Markov chain Monte Carlo sampling method for parameter estimation and inference. We assess the proposed methods in simulation studies and demonstrate the methodology using data from a longitudinal study of dementia.",
keywords = "Bayesian method, Longitudinal bivariate outcomes, Random change point model",
author = "Lili Yang and Sujuan Gao",
year = "2013",
month = "3",
day = "15",
doi = "10.1002/sim.5557",
language = "English",
volume = "32",
pages = "1038--1053",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "6",

}

TY - JOUR

T1 - Bivariate random change point models for longitudinal outcomes

AU - Yang, Lili

AU - Gao, Sujuan

PY - 2013/3/15

Y1 - 2013/3/15

N2 - Epidemiologic and clinical studies routinely collect longitudinal measures of multiple outcomes, including biomarker measures, cognitive functions, and clinical symptoms. These longitudinal outcomes can be used to establish the temporal order of relevant biological processes and their association with the onset of clinical symptoms. Univariate change point models have been used to model various clinical endpoints, suchasCD4 count in studying the progression of HIV infection and cognitive function in the elderly. We propose to use bivariate change point models for two longitudinal outcomes with a focus on the correlation betweenthetwo change points. We consider three types of change point models in the bivariate model setting: the broken-stick model, the Bacon-Watts model, and the smooth polynomial model. We adopt a Bayesianapproachusing a Markov chain Monte Carlo sampling method for parameter estimation and inference. We assess the proposed methods in simulation studies and demonstrate the methodology using data from a longitudinal study of dementia.

AB - Epidemiologic and clinical studies routinely collect longitudinal measures of multiple outcomes, including biomarker measures, cognitive functions, and clinical symptoms. These longitudinal outcomes can be used to establish the temporal order of relevant biological processes and their association with the onset of clinical symptoms. Univariate change point models have been used to model various clinical endpoints, suchasCD4 count in studying the progression of HIV infection and cognitive function in the elderly. We propose to use bivariate change point models for two longitudinal outcomes with a focus on the correlation betweenthetwo change points. We consider three types of change point models in the bivariate model setting: the broken-stick model, the Bacon-Watts model, and the smooth polynomial model. We adopt a Bayesianapproachusing a Markov chain Monte Carlo sampling method for parameter estimation and inference. We assess the proposed methods in simulation studies and demonstrate the methodology using data from a longitudinal study of dementia.

KW - Bayesian method

KW - Longitudinal bivariate outcomes

KW - Random change point model

UR - http://www.scopus.com/inward/record.url?scp=84873999324&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84873999324&partnerID=8YFLogxK

U2 - 10.1002/sim.5557

DO - 10.1002/sim.5557

M3 - Article

VL - 32

SP - 1038

EP - 1053

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 6

ER -