Randomized trial evaluates association between lung function and mortality in women, indicating improved modeling of health indicators.
Key Points
The study aims to develop a Bayesian joint model that evaluates the association between longitudinal data and time-to-event outcomes by addressing the limitations of existing methods.
Developed a joint model for time-to-event and multivariate longitudinal data.
Employed a mixed-effect location-scale model to analyze longitudinal biomarkers and their within-individual variability.
Applied the model on a dataset from the UK cystic fibrosis registry to evaluate associations.
Demonstrated improved hazard ratio estimates compared to traditional models due to reduced regression dilution.
The joint model effectively quantified within-individual variability, showing significant associations between lung function, malnutrition, and mortality.