Abstract
OBJECTIVE: Reducing the incidence of hamstring strain injuries (HSIs) is a priority for soccer clubs. However, robust multifactorial predictive models are lacking and potential predictors such as sprint kinematics, performance fatigability, and psychological variables have been overlooked. Thus, the aim of this study was to develop a preliminary parsimonious multifactorial model to predict players at risk of HSI through preseason screening. MATERIALS AND METHOD: Psychological, physiological, kinematic, performance fatigability and health-related variables were collected for 120 regional and national soccer players during the 2022 preseason. HSIs were prospectively recorded over the entire soccer season. After variable selection, logistic regressions with the Wald backward stepwise method were used to refine the model. The predictive abilities of the model and of the individual variables were determined using the area under the receiver operating characteristic curve (AUC). RESULTS: Twenty-nine players sustained an HSI during the follow-up period. The final model included eight variables: age, sex, HSI history, knee flexor performance fatigability, sprint performance (best sprint time and maximal theoretical velocity V(0)), perceived vulnerability to injury, and subjective norms in soccer. While its model was preliminary, it showed good fit indices and strong predictive performance (true positive rate: 79%, AUC = .82). None of the variables evaluated independently demonstrated satisfactory performance in predicting HSI (AUC≤.65). CONCLUSION: Using a multidisciplinary approach and measurements of only a few variables during preseason screening, the current model tends to demonstrate high accuracy in identifying soccer players at risk of HSI.