A Prospective Study on the Effect of Sport Specialization on Lower Extremity Injury Rates in High School Athletes

1,544 athletes; 167,349 athlete-exposures

US high-school athletes across 29 schools; 2,843 athlete-seasons

Prospective cohort (Level 2), 29 Wisconsin high schools, athletic-trainer-verified injury surveillance with athlete-exposure denominators — the strongest design in this literature. Specialization graded at baseline on the 3-point scale (low 59.5%, moderate 27.1%, high 13.4%). Still observational: not randomized, with residual confounding by year-round training volume, sport type and prior injury.

The best-designed specialization-injury study: prospective, trainer-verified injuries, exposure denominators. High specializers had ~1.85x the lower-extremity injury hazard of low specializers with a monotonic dose-response (1.00/1.51/1.85). Confidence intervals are wide with lower bounds near 1.0-1.1, so the true effect may be modest. Because specialization and annual volume are inseparable observationally, this bounds the harm as real but likely smaller than headline framings imply.

Heritable tissue and maturational phenotypes drive injury susceptibility; highly specialized athletes are also positively selected on ability, so specialization is not a purely imposed exposure.

Consistent with Bell 2018 meta (RR 1.81) and Jayanthi 2020 (OR 1.41-1.46); later single-sport prospective work (e.g. HS basketball) reports weaker or null specialization effects once volume is modelled.