Joint Modelling of Longitudinal and Competing Risks Data in Clinical Research
DOI:
https://doi.org/10.57805/revstat.v17i2.267Keywords:
competing risks, joint modelling, longitudinal data, peritoneal dialysis, time-to-event dataAbstract
Joint modelling of longitudinal and survival data has received much attention in the recent years and is becoming increasingly used in clinical studies. When the longitudinal outcome and survival endpoints are associated, the many well-established models with different specifications proposed to analyse separately longitudinal and time-to-event outcomes are not suitable to analyse such data and a joint modelling approach is required. Although some joint models were adapted in order to allow for competing endpoints, this methodology has not been widely disseminated. The present study has as main objective to model jointly longitudinal and survival data in a competing risk context, discussing the different parameterisations of systematic implementations of these models in the R, using a real data set as an example for the comparison between the different model approaches. The relevance of this issue is associated with the need to draw attention of the users of this statistical software to the different interpretations of model parameters when fitting these models. To reinforce the relevance of these models in clinical research, we give an example of a data set on peritoneal dialysis that was analysed in this context, where death/transfer to haemodialysis was the event of interest and renal transplant was the competing event. Joint modelling results were also compared to separate analysis for these data.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2019 REVSTAT-Statistical Journal
This work is licensed under a Creative Commons Attribution 4.0 International License.