CMStatistics 2016: Start Registration
View Submission - CMStatistics
B1090
Title: Regression approaches for competing risks models with time-dependent covariates Authors:  Manuela Cattelan - University of Padova (Italy) [presenting]
Giuliana Cortese - University of Padua (Italy)
Abstract: In the study of access to pediatric intensive care unit for children who underwent hematopoietic stem cell transplantation, the event may not be observed because of the patients death. In such instances, models for competing risks should be employed and the additional presence of internal time-dependent covariates should be considered. Indeed, children may undergo a second hematopoietic stem cell transplantation. In competing risks models there are different approaches to account for time-dependent covariates, an example is landmark analysis. The advantages and disadvantages of different models and methods for accounting for time-dependent covariates will be pointed out and a novel approach for mean residual life models in presence of competing risks will be presented. The methodology will be illustrated through an application to access to intensive care unit of children affected by oncohematologic diseases who underwent hematopoietic stem cell transplantation in the HSCT Unit of Paduas Pediatric Oncohematology Clinic in the last 15 years.