Joint modelling of complex longitudinal outcomes and time-to-event data : with a focus on quality of life outcomes in cancer clinical trials

Doms, Hortense
(2026)

Files

thesis_hortensedoms.pdf
  • Open Access
  • Adobe PDF
  • 3.82 MB
Download

Details

Authors
  • Doms, Hortenseorcid-logoInstitut de Statistique, Biostatistique et Sciences Actuarielles, Université catholique de Louvain, Louvain-la-Neuve, Belgium
    author
Supervisors
Legrand, Catherine  
 ;
Lambert, Philippe  
Abstract
(en) In clinical trials, longitudinal measurements and time-to-event outcomes are collected to assess treatment efficacy and disease progression. Joint models provide a natural framework to analyse these data simultaneously, but their use becomes challenging with complex longitudinal structures. In oncology, health-related quality of life (HRQoL) outcomes are increasingly considered alongside survival endpoints. However, these data are often multidimensional, ordinal, and subject to informative dropout, which may lead to biased inference if not properly addressed. This thesis develops joint modelling approaches tailored to these challenges, with a focus on HRQoL data. The proposed work builds on latent variable models to capture the underlying structure of HRQoL outcomes over time. In particular, a joint framework is introduced for the analysis of multiple longitudinal ordinal outcomes in the presence of informative and competing dropouts. The model links a latent HRQoL process to cause-specific dropout hazards, explicitly accounting for dropout mechanisms in the longitudinal component. To address computational challenges, two-stage approaches are also investigated as efficient alternatives to full joint estimation. A bias-corrected strategy is proposed for multidimensional latent trait joint models, enabling the analysis of multiple HRQoL domains with reduced computational burden. In addition, more flexible specifications of the survival submodel are explored to improve model applicability. The proposed methods are illustrated using clinical trial data in glioblastoma.
Affiliations

Citations

  • APA
  • Chicago
  • FWB

Doms, H. (2026). Joint modelling of complex longitudinal outcomes and time-to-event data : with a focus on quality of life outcomes in cancer clinical trials [Sage Publications Ltd.]. https://hdl.handle.net/2078.5/276562