Joint modeling of longitudinal data and competing risks survival data with cause and time dependent masking
Principal Investigator
Name
Vikas Barnwal
Degrees
Ph.D.
Institution
Banaras Hindu University
Position Title
Research Scholar
Email
About this CDAS Project
Study
PLCO
(Learn more about this study)
Project ID
PLCO-972
Initial CDAS Request Approval
Apr 25, 2022
Title
Joint modeling of longitudinal data and competing risks survival data with cause and time dependent masking
Summary
This is statistical theory development work for publication in a suitable statistical journal. We request the data as an application to apply the methodology we would develop. In many clinical trials, along with repeated measurements on potential biomarkers for a disease of interest, time to event data is also generated. In a competing risks setup, the subjects are at risk due to more than one possible cause of failure simultaneously. The failure of the subject is related to the longitudinal measurements and in this case, the censoring is informative. There is always a demand to jointly analyze the longitudinal data and competing risks data. Many times, the cause of failure is not known to some of the individuals and hence the incomplete data with respect to the cause of failure is known as masked data in competing risks setup. There is always a possibility that the probability of masking is not identical for all the causes. Hence, to deal with the cause dependent masking situation in joint modelling, we develop new methodologies utilizing the techniques to deal with different types of masking for the survival model. The new methods and techniques will be derived under the Bayesian paradigm.
Aims
1. To develop the methods for dealing with the cause dependent masking in the survival sub-model in a joint analysis of longitudinal and competing risks data.
2. The developed methodology will also extend to handle the time-dependent masking for the competing risks data
3. To evaluate the effect of longitudinal responses on the masking due to specific causes.
Collaborators
Dr. M S Panwar, Assistant Professor, Department of Statistics, BHU, Varanasi, India.