Joint latent class model: Simulation study of model properties and application to amyotrophic lateral sclerosis disease.
Amyotrophic lateral sclerosis
Joint model
Latent classes
Linear mixed model
MLE properties
Monte Carlo simulations
Survival analysis
Journal
BMC medical research methodology
ISSN: 1471-2288
Titre abrégé: BMC Med Res Methodol
Pays: England
ID NLM: 100968545
Informations de publication
Date de publication:
30 09 2021
30 09 2021
Historique:
received:
07
04
2021
accepted:
20
08
2021
entrez:
1
10
2021
pubmed:
2
10
2021
medline:
3
11
2021
Statut:
epublish
Résumé
In many clinical applications, evolution of a longitudinal marker is censored by an event occurrence, and, symmetrically, event occurrence can be influenced by the longitudinal marker evolution. In such frameworks joint modeling is of high interest. The Joint Latent Class Model (JLCM) allows to stratify the population into groups (classes) of patients that are homogeneous both with respect to the evolution of a longitudinal marker and to the occurrence of an event; this model is widely employed in real-life applications. However, the finite sample-size properties of this model remain poorly explored. In the present paper, a simulation study is carried out to assess the impact of the number of individuals, of the censoring rate and of the degree of class separation on the finite sample size properties of the JLCM. A real-life application from the neurology domain is also presented. This study assesses the precision of class membership prediction and the impact of covariates omission on the model parameter estimates. Simulation study reveals some departures from normality of the model for survival sub-model parameters. The censoring rate and the number of individuals impact the relative bias of parameters, especially when the classes are weakly distinguished. In real-data application the observed heterogeneity on individual profiles in terms of a longitudinal marker evolution and of the event occurrence remains after adjusting to clinically relevant and available covariates; CONCLUSION: The JLCM properties have been evaluated. We have illustrated the discovery in practice and highlights the usefulness of the joint models with latent classes in this kind of data even with pre-specified factors. We made some recommendations for the use of this model and for future research.
Sections du résumé
BACKGROUND
In many clinical applications, evolution of a longitudinal marker is censored by an event occurrence, and, symmetrically, event occurrence can be influenced by the longitudinal marker evolution. In such frameworks joint modeling is of high interest. The Joint Latent Class Model (JLCM) allows to stratify the population into groups (classes) of patients that are homogeneous both with respect to the evolution of a longitudinal marker and to the occurrence of an event; this model is widely employed in real-life applications. However, the finite sample-size properties of this model remain poorly explored.
METHODS
In the present paper, a simulation study is carried out to assess the impact of the number of individuals, of the censoring rate and of the degree of class separation on the finite sample size properties of the JLCM. A real-life application from the neurology domain is also presented. This study assesses the precision of class membership prediction and the impact of covariates omission on the model parameter estimates.
RESULTS
Simulation study reveals some departures from normality of the model for survival sub-model parameters. The censoring rate and the number of individuals impact the relative bias of parameters, especially when the classes are weakly distinguished. In real-data application the observed heterogeneity on individual profiles in terms of a longitudinal marker evolution and of the event occurrence remains after adjusting to clinically relevant and available covariates; CONCLUSION: The JLCM properties have been evaluated. We have illustrated the discovery in practice and highlights the usefulness of the joint models with latent classes in this kind of data even with pre-specified factors. We made some recommendations for the use of this model and for future research.
Identifiants
pubmed: 34592944
doi: 10.1186/s12874-021-01377-9
pii: 10.1186/s12874-021-01377-9
pmc: PMC8482570
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
198Informations de copyright
© 2021. The Author(s).
Références
Hypertension. 2016 Jun;67(6):1126-32
pubmed: 27160196
Eur J Epidemiol. 2014 Mar;29(3):211-9
pubmed: 24577561
Stat Med. 2016 Feb 10;35(3):382-98
pubmed: 26376900
J Transplant. 2019 Apr 8;2019:7245142
pubmed: 31093367
Amyotroph Lateral Scler Other Motor Neuron Disord. 2000 Dec;1(5):293-9
pubmed: 11464847
Biometrics. 2016 Dec;72(4):1123-1135
pubmed: 27123856
Stat Methods Med Res. 2014 Feb;23(1):74-90
pubmed: 22517270
Epidemiology. 2019 Jan;30(1):38-47
pubmed: 30499863
J Am Heart Assoc. 2019 Feb 5;8(3):e010628
pubmed: 30686107
Amyotroph Lateral Scler. 2009 Oct-Dec;10(5-6):310-23
pubmed: 19922118
Eur J Neurol. 2014 Mar;21(3):529-36
pubmed: 24447620
Psychiatry Res. 2019 Aug;278:19-26
pubmed: 31132572
Comput Methods Programs Biomed. 2005 May;78(2):165-73
pubmed: 15848271
J Clin Neurosci. 2016 Oct;32:51-6
pubmed: 27401224
Biometrics. 1999 Jun;55(2):463-9
pubmed: 11318201
Stat Med. 2016 Sep 30;35(22):3933-48
pubmed: 27090611
Biometrics. 1997 Mar;53(1):330-9
pubmed: 9147598
Sci Rep. 2019 May 23;9(1):7760
pubmed: 31123290
Kidney Int. 2019 Jan;95(1):178-187
pubmed: 30415941
Lancet Neurol. 2011 May;10(5):481-90
pubmed: 21511200