Risk estimation and dynamic prediction using discrete-time joint models for longitudinal and multistate data with interval and state censoring.

Authors

Lu You, Falastin Salami, Carina Törn, Åke Lernmark, Kendra Vehik, Xiang Liu, Peihua Qiu, Roy Tamura

Published in

Biostatistics (Oxford, England). Volume 27. Issue 1. Jan 20, 2026.

Abstract

This paper presents a joint model of multivariate longitudinal data and multistate data with application to modeling and predicting autoantibody development in The Environmental Determinants of Diabetes in the Young (TEDDY) study. The model quantifies the risks of state transitions based on observed time-varying and non-time-varying risk factors. Based on the estimated model, a dynamic prediction approach is suggested to predict future state occupation probabilities using historical data. The proposed method can handle uncertainties in the observed data, due to measurement errors in the observed longitudinal data and interval censoring or missing information in the observed multistate data. For evaluating the predictions by the proposed approach, some performance metrics and their estimation are discussed. The proposed method is evaluated by some simulation studies. It is discussed in detail how this method can be used in analyzing the TEDDY data by properly handling the missing information and predicting future disease status using the proposed dynamic prediction algorithm.

PMID:
42410993
Bibliographic data and abstract were imported from PubMed on 07 Jul 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.