Browsing Kinesiology by Subject "ROC curve"
Now showing items 1-1 of 1
Association of Functional Screening Tests and Noncontact Injuries in Division I Women Student-AthletesTo determine the association between functional screening tests and lower-body, noncontact injuries in Division I women basketball, soccer, and volleyball student-athletes (SA). Sixty-eight injury-free women SA (age19.1 ± 1.1 years, height171.3 ± 8.7 cm, and mass68.4 ± 9.5 kg) were tested preseason with single hop (SH), triple hop (TH), and crossover hop (XH) for distance, and isometric hip strength (abduction, extension, and external rotation) in randomized order. The first lower-body (spine and lower extremity), noncontact injury requiring intervention by the athletic trainer was abstracted from the electronic medical record. Receiver operating characteristic and area under the curve (AUC) were calculated to determine cut-points for each hopping test from the absolute value of between-limb difference. Body mass–adjusted strength was categorized into tertiles. Logistic regression determined the odds of injury with each functional screening test using the hopping tests cut-points and strength categories, adjusting for previous injury. Fifty-two SA were injured during the sport season. The cut-point for SH was 4 cm (sensitivity = 0.77, specificity = 0.43, and AUC = 0.53), and for TH and XH was 12 cm (sensitivity = 0.75 and 0.67, specificity = 0.71 and 0.57, AUC = 0.59 and 0.41, respectively). A statistically significant association with TH and injuries (adjusted odds ratio = 6.50 [95% confidence interval1.69–25.04]) was found. No significant overall association was found with SH or XH, nor with the strength tests. Using a clinically relevant injury definition, the TH showed the strongest predictive ability for noncontact injuries. This hopping test may be a clinically useful tool to help identify increased risk of injury in women SA participating in high-risk sports.