Abstract
Often, in follow‐up studies, patients experience intermediate events, such as reinterventions or adverse events, which directly affect the shapes of their longitudinal profiles. Our work is motivated by two studies in which such intermediate events have been recorded during follow‐up. In both studies, we are interested in the change of the longitudinal evolutions after the occurrence of the intermediate event and in utilizing this information to improve the accuracy of dynamic prediction of their risk. To achieve so, we propose a flexible joint modeling framework for longitudinal and time‐to‐event data, which includes features of the intermediate event as time‐varying covariates in both the longitudinal and survival submodels. We consider a set of joint models that postulate different effects of the intermediate event in the longitudinal profile and the risk of the clinical endpoint, with different formulations for the association structure while allowing its functional form to change after the occurrence of the intermediate event. Based on these models, we derive dynamic predictions of conditional survival probabilities which are adaptive to different scenarios with respect to the occurrence of the intermediate event. We evaluate the predictive accuracy of these predictions with a simulation study using the time‐dependent area under the receiver operating characteristic curve and the expected prediction error adjusted to our setting. The results suggest that accounting for the changes in the longitudinal profiles and the instantaneous risk for the clinical endpoint is important, and improves the accuracy of the dynamic predictions.
Highlights
Nowadays, there is great interest in the medical field for predictive tools that facilitate precision medicine
Our goal is to investigate the association between the pulmonary gradient and the risk of death, how reoperation as an intermediate event changes the evolution of the pulmonary gradient and the instantaneous risk for death, and to utilize this information to derive individualized dynamic predictions under different scenarios with respect to a future time of reoperation
The main goal of the simulation study is to show the benefit in the accuracy of the individualized dynamic predictions when assuming that the intermediate event changes both the risk for the event of interest and the longitudinal trajectory against the case of assuming that the intermediate event only changes the risk for the event of interest while the longitudinal trajectory is extrapolated and the case where the longitudinal data are not taken into account
Summary
There is great interest in the medical field for predictive tools that facilitate precision medicine. In the context of follow-up studies, in which patients are monitored with several longitudinally measured parameters and biomarkers, physicians are interested in utilizing this information for predicting clinical endpoints. Often, in follow-up studies, intermediate events occur in some patients that directly affect the shapes of their longitudinal evolutions These may include events that are either directly in the control of the investigators, such as additional reinterventions, or maybe not, such as adverse events that the patients may experience. Only Sène et al[7] and Taylor et al[8] investigated this topic in the context of prostate cancer recurrence and radiotherapy as an intermediate event In their approach, they only considered the biomarker trajectories up to the occurrence of the intermediate event assuming extrapolation of the longitudinal profile thereafter. Our goal is to show that utilizing the whole longitudinal trajectory, while capturing the changes to its shape due to the occurrence of intermediate events, can considerably improve the accuracy of such predictions
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call