D. Lee, L. Zumeta Olaskoaga, A. Bender
We propose a flexible modeling framework to analyze time-varying exposures and recurrent events in team sports injuries. This framework leverages a piece-wise exponential additive mixed model to capture the cumulative and potentially complex effects of past exposures, such as high-intensity training loads, on current injury risk. To determine the optimal time window during which past exposures influence risk, we introduce a penalty-based approach.
A simulation study is conducted to assess the performance of the proposed model under various scenarios, including different underlying weight functions and varying levels of heterogeneity across recurrent events. Finally, we demonstrate the application of this approach through a case study involving an elite male football team competing in Spain's LaLiga. The cohort includes time-loss injury data and external training load measures tracked via Global Positioning System (GPS) devices over the 2017–2018 and 2018–2019 seasons.
Keywords: Time-varying exposures, Sports Injuries, Piece-wise Exponential Additive Models
Scheduled
Sports analytics
June 12, 2025 7:00 PM
MR 3