Assessing the Calibration of Subdistribution Hazard Models in Discrete Time

Abstract: The generalization performance of a risk prediction model can be evaluated by its calibration, which measures the agreement between predicted and observed outcomes on external validation data. Here, methods for assessing the calibration of discrete time-to-event models in the presence of competing risks are proposed. The methods are designed for the class of discrete subdistribution hazard models, which directly relate the cumulative incidence function of one event of interest to a set of covariates. Simulation studies show that the methods are strong tools for calibration assessment even in scenarios with a high censoring rate and/or a large number of discrete time points. The proposed approaches are illustrated by an analysis of nosocomial pneumonia.