A0613
Title: Testing similarity of parametric competing risks models for identifying potentially similar pathways in healthcare
Authors: Kathrin Moellenhoff - University of Cologne, Faculty of Medicine and University Hospital, Cologne, Germany (Germany) [presenting]
Nadine Binder - University of Freiburg (Germany)
Holger Dette - Ruhr-Universitaet Bochum (Germany)
Abstract: The identification of similar patient pathways is a crucial task in healthcare analytics. A flexible tool to address this issue is parametric competing risks models, where transition intensities may be specified by a variety of parametric distributions, thus, in particular, being possibly time-dependent. The similarity between two such models is assessed by examining the transitions between different health states. A method is introduced to measure the maximum differences in transition intensities over time, leading to the development of a test procedure for assessing similarity. A parametric bootstrap approach is proposed for this purpose, and proof to confirm the validity of this procedure is provided. The performance of the proposed method is evaluated through a simulation study, considering a range of sample sizes, differing amounts of censoring, and various thresholds for similarity. Finally, the practical application of the approach is demonstrated with a case study from urological clinical routine practice, which inspired this research.
