Latent transition regression for mixed outcomes

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

Health status is a complex outcome, often characterized by multiple measures. When assessing changes in health status over time, multiple measures are typically collected longitudinally. Analytic challenges posed by these multivariate longitudinal data are further complicated when the outcomes are combinations of continuous, categorical, and count data. To address these challenges, we propose a fully Bayesian latent transition regression approach for jointly analyzing a mixture of longitudinal outcomes from any distribution. Health status is assumed to be a categorical latent variable, and the multiple outcomes are treated as surrogate measures of the latent health state, observed with error. Using this approach, both baseline latent health state prevalences and the probabilities of transitioning between the health states over time are modeled as functions of covariates. The observed outcomes are related to the latent health states through regression models that include subject-specific effects to account for residual correlation among repeated measures over time, and covariate effects to account for differential measurement of the latent health states. We illustrate our approach with data from a longitudinal study of back pain.

Original languageEnglish (US)
Pages (from-to)710-720
Number of pages11
JournalBiometrics
Volume59
Issue number3
DOIs
StatePublished - Sep 2003
Externally publishedYes

Fingerprint

health status
Health
Regression
Health Status
longitudinal studies
pain
Covariates
Back Pain
Multiple Outcomes
Longitudinal Studies
Categorical variable
Repeated Measures
Nominal or categorical data
Count Data
Longitudinal Study
Multivariate Data
Pain
Latent Variables
Longitudinal Data
Baseline

Keywords

  • Hierarchical model
  • Latent class analysis
  • Latent traits
  • Latent transition analysis
  • Longitudinal data
  • Markov chain Monte Carlo
  • Multivariate response

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)
  • Public Health, Environmental and Occupational Health
  • Agricultural and Biological Sciences (miscellaneous)
  • Applied Mathematics
  • Statistics and Probability

Cite this

Latent transition regression for mixed outcomes. / Miglioretti, Diana L.

In: Biometrics, Vol. 59, No. 3, 09.2003, p. 710-720.

Research output: Contribution to journalArticle

@article{1327e53bd7804bc0bb40d2bc2eba4feb,
title = "Latent transition regression for mixed outcomes",
abstract = "Health status is a complex outcome, often characterized by multiple measures. When assessing changes in health status over time, multiple measures are typically collected longitudinally. Analytic challenges posed by these multivariate longitudinal data are further complicated when the outcomes are combinations of continuous, categorical, and count data. To address these challenges, we propose a fully Bayesian latent transition regression approach for jointly analyzing a mixture of longitudinal outcomes from any distribution. Health status is assumed to be a categorical latent variable, and the multiple outcomes are treated as surrogate measures of the latent health state, observed with error. Using this approach, both baseline latent health state prevalences and the probabilities of transitioning between the health states over time are modeled as functions of covariates. The observed outcomes are related to the latent health states through regression models that include subject-specific effects to account for residual correlation among repeated measures over time, and covariate effects to account for differential measurement of the latent health states. We illustrate our approach with data from a longitudinal study of back pain.",
keywords = "Hierarchical model, Latent class analysis, Latent traits, Latent transition analysis, Longitudinal data, Markov chain Monte Carlo, Multivariate response",
author = "Miglioretti, {Diana L}",
year = "2003",
month = "9",
doi = "10.1111/1541-0420.00082",
language = "English (US)",
volume = "59",
pages = "710--720",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",
number = "3",

}

TY - JOUR

T1 - Latent transition regression for mixed outcomes

AU - Miglioretti, Diana L

PY - 2003/9

Y1 - 2003/9

N2 - Health status is a complex outcome, often characterized by multiple measures. When assessing changes in health status over time, multiple measures are typically collected longitudinally. Analytic challenges posed by these multivariate longitudinal data are further complicated when the outcomes are combinations of continuous, categorical, and count data. To address these challenges, we propose a fully Bayesian latent transition regression approach for jointly analyzing a mixture of longitudinal outcomes from any distribution. Health status is assumed to be a categorical latent variable, and the multiple outcomes are treated as surrogate measures of the latent health state, observed with error. Using this approach, both baseline latent health state prevalences and the probabilities of transitioning between the health states over time are modeled as functions of covariates. The observed outcomes are related to the latent health states through regression models that include subject-specific effects to account for residual correlation among repeated measures over time, and covariate effects to account for differential measurement of the latent health states. We illustrate our approach with data from a longitudinal study of back pain.

AB - Health status is a complex outcome, often characterized by multiple measures. When assessing changes in health status over time, multiple measures are typically collected longitudinally. Analytic challenges posed by these multivariate longitudinal data are further complicated when the outcomes are combinations of continuous, categorical, and count data. To address these challenges, we propose a fully Bayesian latent transition regression approach for jointly analyzing a mixture of longitudinal outcomes from any distribution. Health status is assumed to be a categorical latent variable, and the multiple outcomes are treated as surrogate measures of the latent health state, observed with error. Using this approach, both baseline latent health state prevalences and the probabilities of transitioning between the health states over time are modeled as functions of covariates. The observed outcomes are related to the latent health states through regression models that include subject-specific effects to account for residual correlation among repeated measures over time, and covariate effects to account for differential measurement of the latent health states. We illustrate our approach with data from a longitudinal study of back pain.

KW - Hierarchical model

KW - Latent class analysis

KW - Latent traits

KW - Latent transition analysis

KW - Longitudinal data

KW - Markov chain Monte Carlo

KW - Multivariate response

UR - http://www.scopus.com/inward/record.url?scp=0041833563&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0041833563&partnerID=8YFLogxK

U2 - 10.1111/1541-0420.00082

DO - 10.1111/1541-0420.00082

M3 - Article

C2 - 14601773

AN - SCOPUS:0041833563

VL - 59

SP - 710

EP - 720

JO - Biometrics

JF - Biometrics

SN - 0006-341X

IS - 3

ER -