Bayesian joint modeling of longitudinal patient-reported outcomes and survival: an application to chronic obstructive pulmonary disease

2026-09-24

AI Summary

This study addresses the two-stage estimation bias in joint modeling of longitudinal patient-reported outcomes and survival data when responses are discrete, bounded, and overdispersed. A Bayesian joint model is proposed that couples a beta-binomial mixed-effects submodel with a Weibull proportional hazards submodel through individual response probabilities, employing a simultaneous estimation strategy to replace conventional two-stage approaches. Applied to a COPD cohort, the method substantially reduces bias in longitudinal slope estimates and yields unbiased association parameters. Compared with existing methods, it identifies more prognostic associations across SF-36 and SGRQ dimensions while achieving accurate dynamic survival prediction.

Abstract

Questionnaire-based patient-reported outcomes (PROs) are discrete, bounded and overdispersed, yet joint models relating them to survival may ignore these features or estimate both processes sequentially. We propose a Bayesian joint model combining a beta-binomial mixed-effects submodel with a Weibull proportional hazards submodel, linked through the subject-specific response probability. Simulations show that simultaneous estimation reduces bias in the longitudinal slope and yields practically unbiased association estimates, unlike two-stage estimation. In a cohort of 543 patients with chronic obstructive pulmonary disease, the model identified associations for all eight SF-36 dimensions and for two of three SGRQ dimensions, including several associations not detected by the two-stage approach, and provided dynamic survival predictions.

Problem

• patient-reported outcomes
• joint modeling
• survival analysis
• longitudinal data
• overdispersion

Innovation

• Bayesian joint model
• beta-binomial mixed-effects
• patient-reported outcomes
• simultaneous estimation
• dynamic survival prediction