A1465
Title: Modelling competing risks data through clustering of cumulative incidence functions
Authors: Marta Sestelo - University of Vigo (Spain) [presenting]
Nora M Villanueva - University of Vigo (Spain)
Luis Machado - University of Minho (Portugal)
Javier Roca Pardinas - University of Vigo (Spain)
Abstract: The cumulative incidence function is the standard tool for estimating the marginal probability of a given event in the presence of competing risks. One basic but important goal in the analysis of competing risk data is the comparison of these curves, for which limited literature exists. An R package is presented that implements a procedure to not only test the equality of cumulative incidence curves but also cluster them when differences exist. The package automatically determines group composition and selects the optimal number of clusters. The applicability of the proposed method is illustrated using real data. This tool provides researchers with a practical and accessible framework for exploring and analyzing competing risks data.
