A Bayesian hierarchical model for prediction of latent health states from multiple data sources with application to active surveillance of prostate cancer

DOI10.1111/BIOM.12577zbMATH Open1372.62063arXiv1508.07511OpenAlexW2964114032WikidataQ31122903 ScholiaQ31122903MaRDI QIDQ5283323FDOQ5283323

Authors: Rebecca Yates Coley, Aaron J. Fisher, Mufaddal Mamawala, H. Ballentine Carter, Kenneth J. Pienta, Scott L. Zeger 

Publication date: 21 July 2017

Published in: Biometrics (Search for Journal in Brave)

Abstract: In this article, we present a Bayesian hierarchical model for predicting a latent health state from longitudinal clinical measurements. Model development is motivated by the need to integrate multiple sources of data to improve clinical decisions about whether to remove or irradiate a patient's prostate cancer. Existing modeling approaches are extended to accommodate measurement error in cancer state determinations based on biopsied tissue, clinical measurements possibly not missing at random, and informative partial observation of the true state. The proposed model enables estimation of whether an individual's underlying prostate cancer is aggressive, requiring surgery and/or radiation, or indolent, permitting continued surveillance. These individualized predictions can then be communicated to clinicians and patients to inform decision-making. We demonstrate the model with data from a cohort of low risk prostate cancer patients at Johns Hopkins University and assess predictive accuracy among a subset for whom true cancer state is observed. Simulation studies confirm model performance and explore the impact of adjusting for informative missingness on true state predictions. R code and simulated data available at https://github.com/rycoley/prediction-prostate-surveillance.

Full work available at URL: https://arxiv.org/abs/1508.07511

Recommendations

• Individual Prediction in Prostate Cancer Studies Using a Joint Longitudinal Survival–Cure Model
• Real-time individual predictions of prostate cancer recurrence using joint models
• Hierarchical Bayesian Neural Networks
• Modeling Disease Progression With Longitudinal Markers
• Latent Class Models for Joint Analysis of Longitudinal Biomarker and Event Process Data

zbMATH Keywords

missing datalatent class analysisprecision medicinerisk classificationprostate cancer prognosis

Mathematics Subject Classification ID

Applications of statistics to biology and medical sciences; meta analysis (62P10) Medical applications (general) (92C50)

Cites Work

• Title not available (Why is that?)
• Random-Effects Models for Longitudinal Data
• The Calculation of Posterior Distributions by Data Augmentation
• Efficient parametrisations for normal linear mixed models
• Title not available (Why is that?)
• Joint modelling of longitudinal measurements and event time data
• Missing Data in Longitudinal Studies
• Modelling Progression of CD4-Lymphocyte Count and Its Relationship to Survival Time
• Latent Class Models for Joint Analysis of Longitudinal Biomarker and Event Process Data
• Domain-level covariance analysis for multilevel survey data with structured nonresponse
• Modeling Disease Progression With Longitudinal Markers

Cited In (7)

• Integrative analysis using coupled latent variable models for individualizing prognoses
• Bayesian hierarchical rule modeling for predicting medical conditions
• A numerically stable algorithm for integrating Bayesian models using Markov melding
• A Bayesian latent class approach for EHR-based phenotyping
• Optimizing active surveillance for prostate cancer using partially observable Markov decision processes
• prediction-prostate-surveillance
• Hierarchical Bayesian Neural Networks

Uses Software

• R
• JAGS
• R2jags
• GitHub
• PRMLT

This page was built for publication: A Bayesian hierarchical model for prediction of latent health states from multiple data sources with application to active surveillance of prostate cancer

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5283323)