Modeling Training Loads and Injuries: The Dangers of Discretization

To evaluate common modeling strategies in training load and injury risk research when modeling continuous variables and interpreting continuous risk estimates; and present improved modeling strategies.

Workload data were pooled from Australian football (n = 2550) and soccer (n = 23,742) populations to create a representative sample of acute:chronic workload ratio observations for team sports. Injuries were simulated in the data using three predefined risk profiles (U-shaped, flat and S-shaped). One-hundred data sets were simulated with sample sizes of 1000 and 5000 observations. Discrete modeling methods were compared with continuous methods (spline regression and fractional polynomials) for their ability to fit the defined risk profiles. Models were evaluated using measures of discrimination (area under receiver operator characteristic [ROC] curve) and calibration (Brier score, logarithmic scoring).

Discrete models were inferior to continuous methods for fitting the true injury risk profiles in the data. Discrete methods had higher false discovery rates (16%-21%) than continuous methods (3%-7%). Evaluating models using the area under the ROC curve incorrectly identified discrete models as superior in over 30% of simulations. Brier and logarithmic scoring was more suited to assessing model performance with less than 6% discrete model selection rate.

Many studies on the relationship between training loads and injury that have used regression modeling have significant limitations due to improper discretization of continuous variables and risk estimates. Continuous methods are more suited to modeling the relationship between training load and injury. Comparing injury risk models using ROC curves can lead to inferior model selection. Measures of calibration are more informative judging the utility of injury risk models.