Joint Analysis of Survival Time and Longitudinal Categorical Outcomes

Jaeun Choi, Jianwen Cai, Donglin Zeng, Andrew F. Olshan

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

In biomedical or public health research, it is common for both survival time and longitudinal categorical outcomes to be collected for a subject, along with the subject's characteristics or risk factors. Investigators are often interested in finding important variables for predicting both survival time and longitudinal outcomes which could be correlated within the same subject. Existing approaches for such joint analyses deal with continuous longitudinal outcomes. New statistical methods need to be developed for categorical longitudinal outcomes. We propose to simultaneously model the survival time with a stratified Cox proportional hazards model and the longitudinal categorical outcomes with a generalized linear mixed model. Random effects are introduced to account for the dependence between survival time and longitudinal outcomes due to unobserved factors. The Expectation-Maximization (EM) algorithm is used to derive the point estimates for the model parameters, and the observed information matrix is adopted to estimate their asymptotic variances. Asymptotic properties for our proposed maximum likelihood estimators are established using the theory of empirical processes. The method is demonstrated to perform well in finite samples via simulation studies. We illustrate our approach with data from the Carolina Head and Neck Cancer Study (CHANCE) and compare the results based on our simultaneous analysis and the separately conducted analyses using the generalized linear mixed model and the Cox proportional hazards model. Our proposed method identifies more predictors than by separate analyses.

Original languageEnglish (US)
Pages (from-to)1-29
Number of pages29
JournalStatistics in Biosciences
DOIs
StateAccepted/In press - 2013
Externally publishedYes

Fingerprint

Survival Time
Survival Analysis
Categorical
Joints
Proportional Hazards Models
Linear Models
Cox Proportional Hazards Model
Generalized Linear Mixed Model
Hazards
Head and Neck Neoplasms
Observed Information
Point Estimate
Information Matrix
Public Health
Empirical Process
Research Personnel
Expectation-maximization Algorithm
Asymptotic Variance
Public health
Risk Factors

Keywords

  • EM algorithm
  • Generalized linear mixed model
  • Maximum likelihood estimator
  • Random effect
  • Simultaneous modeling
  • Stratified Cox proportional hazards model

ASJC Scopus subject areas

  • Statistics and Probability
  • Biochemistry, Genetics and Molecular Biology (miscellaneous)

Cite this

Joint Analysis of Survival Time and Longitudinal Categorical Outcomes. / Choi, Jaeun; Cai, Jianwen; Zeng, Donglin; Olshan, Andrew F.

In: Statistics in Biosciences, 2013, p. 1-29.

Research output: Contribution to journalArticle

Choi, Jaeun ; Cai, Jianwen ; Zeng, Donglin ; Olshan, Andrew F. / Joint Analysis of Survival Time and Longitudinal Categorical Outcomes. In: Statistics in Biosciences. 2013 ; pp. 1-29.
@article{e6f982d0df4741f28ef1d081b76ef704,
title = "Joint Analysis of Survival Time and Longitudinal Categorical Outcomes",
abstract = "In biomedical or public health research, it is common for both survival time and longitudinal categorical outcomes to be collected for a subject, along with the subject's characteristics or risk factors. Investigators are often interested in finding important variables for predicting both survival time and longitudinal outcomes which could be correlated within the same subject. Existing approaches for such joint analyses deal with continuous longitudinal outcomes. New statistical methods need to be developed for categorical longitudinal outcomes. We propose to simultaneously model the survival time with a stratified Cox proportional hazards model and the longitudinal categorical outcomes with a generalized linear mixed model. Random effects are introduced to account for the dependence between survival time and longitudinal outcomes due to unobserved factors. The Expectation-Maximization (EM) algorithm is used to derive the point estimates for the model parameters, and the observed information matrix is adopted to estimate their asymptotic variances. Asymptotic properties for our proposed maximum likelihood estimators are established using the theory of empirical processes. The method is demonstrated to perform well in finite samples via simulation studies. We illustrate our approach with data from the Carolina Head and Neck Cancer Study (CHANCE) and compare the results based on our simultaneous analysis and the separately conducted analyses using the generalized linear mixed model and the Cox proportional hazards model. Our proposed method identifies more predictors than by separate analyses.",
keywords = "EM algorithm, Generalized linear mixed model, Maximum likelihood estimator, Random effect, Simultaneous modeling, Stratified Cox proportional hazards model",
author = "Jaeun Choi and Jianwen Cai and Donglin Zeng and Olshan, {Andrew F.}",
year = "2013",
doi = "10.1007/s12561-013-9091-z",
language = "English (US)",
pages = "1--29",
journal = "Statistics in Biosciences",
issn = "1867-1764",
publisher = "Springer New York",

}

TY - JOUR

T1 - Joint Analysis of Survival Time and Longitudinal Categorical Outcomes

AU - Choi, Jaeun

AU - Cai, Jianwen

AU - Zeng, Donglin

AU - Olshan, Andrew F.

PY - 2013

Y1 - 2013

N2 - In biomedical or public health research, it is common for both survival time and longitudinal categorical outcomes to be collected for a subject, along with the subject's characteristics or risk factors. Investigators are often interested in finding important variables for predicting both survival time and longitudinal outcomes which could be correlated within the same subject. Existing approaches for such joint analyses deal with continuous longitudinal outcomes. New statistical methods need to be developed for categorical longitudinal outcomes. We propose to simultaneously model the survival time with a stratified Cox proportional hazards model and the longitudinal categorical outcomes with a generalized linear mixed model. Random effects are introduced to account for the dependence between survival time and longitudinal outcomes due to unobserved factors. The Expectation-Maximization (EM) algorithm is used to derive the point estimates for the model parameters, and the observed information matrix is adopted to estimate their asymptotic variances. Asymptotic properties for our proposed maximum likelihood estimators are established using the theory of empirical processes. The method is demonstrated to perform well in finite samples via simulation studies. We illustrate our approach with data from the Carolina Head and Neck Cancer Study (CHANCE) and compare the results based on our simultaneous analysis and the separately conducted analyses using the generalized linear mixed model and the Cox proportional hazards model. Our proposed method identifies more predictors than by separate analyses.

AB - In biomedical or public health research, it is common for both survival time and longitudinal categorical outcomes to be collected for a subject, along with the subject's characteristics or risk factors. Investigators are often interested in finding important variables for predicting both survival time and longitudinal outcomes which could be correlated within the same subject. Existing approaches for such joint analyses deal with continuous longitudinal outcomes. New statistical methods need to be developed for categorical longitudinal outcomes. We propose to simultaneously model the survival time with a stratified Cox proportional hazards model and the longitudinal categorical outcomes with a generalized linear mixed model. Random effects are introduced to account for the dependence between survival time and longitudinal outcomes due to unobserved factors. The Expectation-Maximization (EM) algorithm is used to derive the point estimates for the model parameters, and the observed information matrix is adopted to estimate their asymptotic variances. Asymptotic properties for our proposed maximum likelihood estimators are established using the theory of empirical processes. The method is demonstrated to perform well in finite samples via simulation studies. We illustrate our approach with data from the Carolina Head and Neck Cancer Study (CHANCE) and compare the results based on our simultaneous analysis and the separately conducted analyses using the generalized linear mixed model and the Cox proportional hazards model. Our proposed method identifies more predictors than by separate analyses.

KW - EM algorithm

KW - Generalized linear mixed model

KW - Maximum likelihood estimator

KW - Random effect

KW - Simultaneous modeling

KW - Stratified Cox proportional hazards model

UR - http://www.scopus.com/inward/record.url?scp=84878007966&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878007966&partnerID=8YFLogxK

U2 - 10.1007/s12561-013-9091-z

DO - 10.1007/s12561-013-9091-z

M3 - Article

SP - 1

EP - 29

JO - Statistics in Biosciences

JF - Statistics in Biosciences

SN - 1867-1764

ER -