Predictive Validity of the Functional Movement ScreenTM in a Population of Recreational Runners Training for a Half Marathon

Abstract

PURPOSE: The Functional Movement Screen™ (FMS) has been touted as an easily administered 7-test battery that predicts risk of injury in non-specified athletic populations. A cut score of 14 on the 21 point instrument has been cited as a predictor of basic movement pattern dysfunction and related risk of injury. This study tested the predictive validity of the recommend cut score for the FMS in a sample of recreational runners training for a half-marathon. METHODS: Participants for this study were recruited from the 35,000 individuals registered to participate in the 2006 Indianapolis Mini-Marathon. Sixty individuals were evaluated using the FMS as described. The participants then completed a weekly online survey as a means of tracking training and injury status during their preparation and participation in the event. In each survey participants answered questions related to training variables, subjective pain ratings, and symptoms which may have changed training or prompted participants to seek medical attention. Data was collected for a four-month period leading up to the completion of the half-marathon. RESULTS: FMS scores ranged from 11 to 20 on the 21 point instrument. Forty- nine of the 60 participants (82%) completed the weekly questionnaires for the duration of the study. Of these 49 individuals, 12 reported symptoms consistent with overuse injuries, making the total incidence of injury 24.49%. Only 1 of the 12 injured individuals scored below the suggested cut score of 14, which generated 8.3% sensitivity and 94.5% specificity, respectively. A cut score of 20 generated the highest sensitivity (100%), identifying all injured participants but resulting in 27.0% specificity. The cut score with the highest specificity was 11 (97.2%), but it had a sensitivity of 0%. The cut score of 17 showed the highest combination for specificity (41.6%) and sensitivity (56.7%). Neither linear nor logistic regression analyses identified FMS scores as predictors of musculoskeletal injury. CONCLUSIONS: Based on the results of this study, the cut scores on the FMS did not positively predict musculoskeletal injury, indicating this instrument lacks predictive validity in this population of recreational runners. These findings suggest further study is needed to establish the predictive validity of the FMS in athletic populations.