Can Vertical Countermovement Jump Force-Time Metrics Predict Non-Contact Lower Body Injury in NCAA Division-I Female Athletes?

Abstract

Vertical countermovement jump (CMJ) assessments on force plate systems have been purported to screen for musculoskeletal injury risk but with little to no scientific support. Thus, the purpose of this study was to identify associations and non-contact lower body injury predictability using machine learning algorithms with demographic and CMJ force-time metrics in female athletes. The study entailed a retrospective analysis of routine injury and performance monitoring from 148 female National Collegiate Athletics Association Division I athletes. Medical staff recorded non-contact lower-body injuries that occurred as result of competition or training within three months following CMJ testing (two maximal effort, no arm-swing, jumps on dual force plates). Of the 46 documented injuries, majority occurred in Field Hockey (47.8%), followed by Soccer (26.1%), Lacrosse (21.7%), and Ice Hockey (4.3%). Univariate generalized estimating equation models (GEE) found increased risk of injury as a result of previous injury (Odds Ratio= 6.52±0.29), younger age (0.80±0.10), lower Eccentric Mean Power (0.69±0.15) and Impulse (0.98±0.01), and participation in field hockey (13.50±0.78), lacrosse (6.25±0.78), and soccer (5.27±0.72) compared to ice hockey. Multivariate GEE, controlling for age, sport, and previous injury, revealed no significant injury associations with CMJ metrics. Random forest classification (RFC) resulted in low and unbalanced accuracy (54%). The RFC sensitivity was low (7.2%) and specificity was high (95.0%), meaning this model was more accurate at predicting non-injured athletes than injured athletes. Thus, CMJ force-time metrics from one timepoint may not be useful for non-contact lower body musculoskeletal injury screening or predictability in NCAA female athletes.

Emaly Vatne

Assistant Sport Scientist at The Ohio State University Department of Athletics and Human Performance Collaborative

My research interests include applied sport science, effects of recovery interventions, and data analysis, visualization, and engineering.