Construction and Practical Research on Risk‑Screening System for Urban Adolescent Physical Fitness Training

Abstract

With the continuous expansion of paid commercial adolescent physical‑fitness training, training‑related musculoskeletal injuries occur frequently. Most existing international screening tools are developed based on cohorts of European and American white adolescents, showing defects in population‑ and scenario‑adaptation for East‑Asian urban adolescents. Based on a retrospective real‑world stratified cohort study carried out in three high‑end domestic commercial fitness studios, this research enrolled 427 Asian adolescents aged 11‑17 years (GBD 2021 Risk Factors Collaborators, 2023). LASSO‑penalized conditional logistic regression was adopted to screen independent injury predictors, construct a weighted risk‑screening scoring model, and evaluate injury‑prevention and‑control effects after embedding the screening system into business workflows. The results showed that the ROC‑AUC of the final model in the internal validation set reached 0.812 (95%CI: 0.735‑0.874), and the P‑value of the Hosmer‑Lemeshow test was 0.417, indicating favorable model calibration performance. After the implementation of the screening system, the incidence of clinically significant training‑related injuries within 12 months decreased from 28.34% to 11.72% (Cook G, Burton L & Hoogenboom B., 2006). After adjusting for confounding factors, the odds ratio (OR) for injury occurrence was 0.33 (95%CI: 0.19‑0.57, P<0.001) (von Elm E, Altman D G, Egger M, Pocock S J, Gøtzsche P C & Vandenbroucke J P., 2007). This study confirms that the risk‑screening system containing eight predictors including prior injury history, postural asymmetry and screen exposure can effectively realize individualized risk stratification of training injuries among commercially‑enrolled adolescents. It possesses practical real‑world business implementation value and can provide quantitative tool support for safe physical‑fitness training of adolescents in China.