Predictive Value of the Functional Movement Screen for Sports-Related Injury in NCAA Division I Athletes. Issue 3 (31st March 2017)

• Record Type: Journal Article
• Authors: Wang, Dean; Chen, James; Lai, Wilson; Vail, Jeremy; Rugg, Caitlin Marie; Hame, Sharon L.
• Is Part Of: Orthopaedic journal of sports medicine. Volume 5:Issue 3(2017:Mar.)Supplement 3
• Publication Date: 2017-03-31
• Subjects: Sports medicine -- Periodicals; Orthopedics -- Periodicals; Arthroscopy -- Periodicals; Arthroplasty -- Periodicals; Knee -- Surgery -- Periodicals; 616.7
• DOI: 10.1177/2325967117S00132
• Languages: English
• ISSNs: 2325-9671

Abstract

Objectives: Assessing risk of injury among National Collegiate Athletic Association (NCAA) athletes remains a significant challenge for sports medicine professionals. The Functional Movement Screen (FMS) was created to identify persons at risk for sport-related injuries and is comprised of 7 functional tests (overhead deep squat, in-line lunge, hurdle step, active straight leg raise, shoulder mobility, trunk stability push up, and rotatory stability) involving locomotor, manipulative and stabilization actions that assess balance, mobility, and stability. The purpose of this study was to analyze the predictive value of the FMS for sports-related injury in a cohort of NCAA athletes. Methods: The FMS was administered to NCAA Division I athletes at a single institution (N = 315) during pre-participation physical examinations (PPE). Individual athlete data, including history of prior surgeries, sex, sport, and BMI, was collected during the PPE. Athletes were followed prospectively for an average of 19.3 months (range, 4.1-28.3 months). Clinically significant injuries, defined as those that caused an athlete to miss seven or more days of athletic participation, were recorded. Rate of injury was calculated per athlete-exposure. Predictor variables were first univariately analyzed and included in multivariate models with sex and sport if P < .10. Multivariate Cox regression and Poisson regression was performed to assess predictors for earlier and higher rates of injury, respectively. Receiver Operator Characteristic (ROC) analysis was used to determine the optimal predictive value cut score for the FMS as an injury screening tool. Results: Of the 315 athletes, 186 (59%) were male and 129 (41%) were female. Participants were classified into collision (27%), contact (30%), limited contact (21%) and non-contact (22%) sports. Twenty-eight athletes (9%) had undergone precollegiate lower extremity surgery. Athletes with an obese BMI had a lower mean FMS composite score (12.9) than those with a normal (14.0) and overweight (14.6) BMI (P = 0.03 and P < .01, respectively). An FMS score ≤ 11 was significantly associated with earlier injury (HR 1.98; 95% CI, 1.20-3.19) and higher rates of injury (RR 1.32; 95% CI, 1.23 -1.41) compared to an FMS score >11. Female sex, type of sport, higher BMI, and precollegiate lower extremity surgery were also independent predictors of injury (Table 1 ). The optimal cut point for injury screening as determined by ROC analysis was an FMS score of 13. Using this cut point, sensitivity was 48.1%, specificity was 62.4%, positive predictive value (PPV) was 50.7%, and negative predictive value (NPV) was 60.0%. Moving the cut point to 11 decreased sensitivity to 17.4%, increased specificity to 91.7%, increased PPV to 62.7%, and decreased NPV to 58.1%. The area under the curve (AUC) was 0.58. Table 1. Risk Factors for Rate of Injury Predictor Variable Rate Ratio (95% Cl) P Value Sex Male 1.00 Female 1.19(1.16-1.22) <.01 Sport Collision 1.48(1.36-1.61) <.01 Contact 1.00 Limited Contact 1.40(1.32-1.48) <.01 Non Contact 1.40(1.32-1.49) <.01 Body mass index (kg/m 2 ) <25.0 (underweight/normal) 1.00 25.0-29.9 (overweight) 1.28(1.22-1.34) <.01 ≥30 (obese) 1.76(1.60-1.94) <.01 Precollegiate lower extremity surgery Yes 1.51 (1.46-1.56) <.01 No 1.00 FMS composite score ≤11 1.32(1.23-1.41) <.01 12-14 1.04 (0.99-1.09) .10 ≥15 1.00 Bold values indicate statistical significance ( P < .05) FMS, Functional Movement Screen Conclusion: Although a low FMS score (≤11) indicated earlier and higher rates of injury, the low sensitivity, PPV, and AUC suggest that the FMS is not suitable as a solitary injury screening tool in collegiate athletes. This is likely due to the multifactorial and complex nature of sports-related injuries. Any functional measure must take into account a multitude of factors, such as sex, sport, BMI, and prior injuries/surgeries that are specific to the observed population in order to accurately assess injury risk. Thus, sports medicine professionals should be cautioned against using the FMS alone as an injury screening tool in NCAA athletes.