Predicting Musculoskeletal Injury in National Collegiate Athletic Association Division II Athletes From Asymmetries and Individual-Test Versus Composite Functional Movement Screen Scores

Musculoskeletal injuries (MSIs) are inherent in intercollegiate athletes, with an overall rate of 63.1 per 1000 athlete-exposures for both contact and noncontact injuries.¹ Furthermore, the likelihood of subsequent MSI is high.^2–7 Although our current understanding of injury causation is limited, researchers have identified both extrinsic and intrinsic risk factors for MSI in intercollegiate athletes. Extrinsic risk factors are external or environmental factors⁸ such as footwear or playing surface. Intrinsic risk factors are characteristics of the individual athlete that increase injury disposition,⁹ such as inadequate strength or high body mass index. Recent attention has focused on the relationship between athletic injury and intrinsic risk factors, including history of previous injury,^10,11 core dysfunction,^9,12 adiposity,¹³ landing and cutting biomechanics,¹⁴ quadriceps : hamstrings ratio,⁸ and sex.¹⁴ In populations such as recreational runners,¹⁵ professional American football players,¹⁶ elite track and field athletes,¹⁷ netballers,¹⁸ and high school basketball players,¹⁹ asymmetries in movement patterns have been shown to increase injury risk. Movement-related dysfunctions are of particular interest because they are considered modifiable risk factors that can be targeted by intervention programs, which may decrease injury risk.

Identifying dysfunctional movement patterns that contribute to MSI is an important component in managing active populations.²⁰ However, in practical terms, this does not come easily. The preparticipation physical examination (PPE) is the litmus test for identifying conditions that may lead to MSI in athletes. The standard PPE includes a medical and family history, orthopaedic examination (joint and muscle specific), and general medical screen (eg, cardiorespiratory system, vision). The PPE typically does not include any assessment of movement patterns. Therefore, the PPE may fall short in identifying and preventing injuries caused by improper movement patterns. If movement-pattern screening yields substantive information for injury prediction, then it may be a consideration for inclusion in the PPE.

The Functional Movement Screen (FMS) allows for assessment of movement patterns involving strength, mobility, motor control, and core stability.²¹ Asymmetries or insufficiencies in 7 fundamental movement patterns are rated on a 4-point scale and summed to provide a score out of 21 possible points. Moderate evidence supports the use of FMS summed scores to predict future injury in athletes. Specifically, National Football League players,²² National Collegiate Athletic Association (NCAA) Division II female athletes,²³ and Marine officer candidates²⁴ who scored ≤14 points were 1.89, 4.12, and 1.91 times, respectively, more likely to sustain an MSI than those who scored lower. Lehr et al²⁵ tested an algorithm that used demographic information, previous injury history, presence of pain, and Lower Quarter Y-Balance scores along with FMS scores (composite and individual) to predict MSI in 183 Division III athletes. Participants with moderate or substantial risk, which they delineated as high risk, were 3.4 times more likely to be injured (95% confidence interval [CI] = 2.0, 6.0). Thus, researchers and clinicians have adopted the FMS in managing athletic populations. However, the variation in relative risk (RR) in these aforementioned studies should be considered when interpreting FMS results. High specificity but low sensitivity was reported by authors^16,22–26 using composite FMS scores for injury prediction (specificity = 0.71–0.94, sensitivity = 0.12–0.67), which suggests a high number of false-negative results for those scoring >14. Garrison et al²⁶ noted an increase in specificity (from 0.73 to 0.89) and a reduction in sensitivity (from 0.67 to 0.65) when combining history of previous injury with the composite score in Division I athletes. The use of the composite score may still be questionable. Furthermore, for a composite score on a test composed of individual items to be valid, each individual item is assumed to measure the same latent variable. Factor analysis has been used to test validation in other assessment tools related to athletic injuries, such as the Landing Error Scoring System²⁷ and the Star Excursion Balance Test.²⁸ Kazman et al²⁹ examined the internal consistency and factor structure of the FMS in testing 877 Marine officer candidates and demonstrated a Cronbach α of 0.39; values ≤0.60 are considered unacceptable for scales.³⁰ Exploratory factor analysis revealed very weak correlations between the individual tests (≤0.26). Also, varimax-rotated factor loading revealed 2 components that were significantly higher in 2 factors (pain and no-pain groups): the shoulder-mobility (0.74) and squat tests (0.87). The authors concluded that the FMS composite score cannot be considered a unitary construct. The individual tests were not found to measure a common latent variable; thus, the ability of the FMS composite score to measure injury proneness should be questioned. Furthermore, individual movement patterns are likely more informative than the composite score.

The presence of asymmetries and low scores on individual tests may provide additional injury-predictive value to the FMS, strengthening the usefulness of this tool beyond the summed scores. Of the 7 fundamental movements tested in the FMS, 5 compare movement quality between sides, thereby offering an opportunity to observe movement asymmetry. Wiese et al³¹ failed to predict injury risk across a variety of injury stratifications in 144 NCAA Division I American football players using the summed score. They suggested using the FMS within athletic populations to screen for functional asymmetries. Asymmetries in other types of movement tests such as hop tests,³² lower extremity functional tests,³² and dynamic balance tests¹⁹ have been associated with athletic injuries. More recently, the presence of asymmetry and low individual test scores on the FMS were identifiable risk factors for time loss due to injury in professional American football players.¹⁶ Low scores on the individual movement tests indicate an inability to complete a basic movement pattern, even with significant compensation. Because functional movement deficiencies may be modifiable intrinsic risk factors, examining them is crucial for constructing appropriate intervention programs. Therefore, the purpose of our study was to assess the predictive value of movement asymmetry and a low individual test score on the FMS by investigating preseason scores and subsequent injury over a competitive season in Division II athletes. We hypothesized that, when compared with the FMS composite score, asymmetry or a score of 1 on an individual test would be associated with a greater likelihood of MSI.

METHODS

RESULTS

Descriptive statistics for the FMS summed and individual scores are shown in Table 1. Thirty-eight athletes (45.2%) sustained a total of 94 MSIs. Contact and noncontact MSIs represented 30.9% (29 of 94) and 69.1% (65 of 94), respectively, of total MSIs. Lower extremity injuries were the most frequent type in this group. An injury summary by body region and frequency is shown in Table 2.

Table 1. Functional Movement Screen Scores (N = 84)^a

^a A score of 3 = movement completed as instructed and free of compensation or pain, 2 = movement completed pain free but with some level of compensation, 1 = unable to complete movement as instructed, and 0 = pain with movement or during a clearing test designed to provoke pain and identify injury.

View Fullscreen

Table 2. Summary of Injuries by Body Region

View Fullscreen

The association between injury occurrence and summed FMS score, asymmetry, or score of 1 on an individual test is illustrated in Table 3. Athletes with composite scores of ≤14 were no more likely to sustain an injury than those with higher scores: χ²₁ = 2.07, P = .15. The mean summed FMS score for the injured group was slightly lower than that of the uninjured group but not statistically different (15.75 ± 1.79 versus 15.99 ± 1.71, P > .05). Using the contingency values in Table 4, we calculated sensitivity (26.3%) and specificity (58.7%) for the composite scores. However, athletes who displayed at least 1 asymmetry or limited movement pattern (score = 1) on any of the individual tests were statistically more likely to sustain an injury than those who did not (χ²₁ = 11.39, P = .001). The RR of injury to this group was 2.73 (95% CI = 1.36, 5.44; P = .001), and the odds ratio was 5.27 (95% CI = 1.93, 14.40; P = .001). Sensitivity and specificity were 81.5% and 54.3%, respectively (Table 5). Finally, the Figure depicts the ROC curve for the entire sample. The cut-point score was maximized at 16. The RR and odds ratio calculated for participants with scores ≤16 were 0.29 (95% CI = 0.09, 0.91) and 0.58 (95% CI = 0.38, 0.88; P = .03), respectively. Results of the ROC curve analysis for FMS composite score in predicting MSI are shown in Table 6.

Table 3. Association Between Injury Risk and Functional Movement Screen Measures

Abbreviation: NA, not applicable.

View Fullscreen

Table 4. 2 × 2 Contingency Table Model for Functional Movement Screen Summed Score and Injury

View Fullscreen

Table 5. 2 × 2 Contingency Table Model for Asymmetry or Score of 1 on Functional Movement Screen Individual Test and Injury

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

Table 6. Results From the Receiver Operating Curve Analysis

View Fullscreen

DISCUSSION

Participation in intercollegiate athletics comes with an inherent risk for injury. Being able to identify modifiable factors related to injury has significant value for athletic health care. We sought to determine the utility of examining limited and asymmetric movement patterns from the FMS as factors that predisposed a collegiate athlete to an MSI. A key finding of this study was that the summed score (≤14) did not predict the occurrence of an MSI but that asymmetry or a limited movement pattern on an individual test did.

Most researchers investigating the usefulness of the FMS to predict injury have assessed the composite score and identified a cutoff score of either 14 points^{16,22–24,26,33} or 17 points.³¹ With a cutoff score of 14, we found that these Division II athletes were not more likely to sustain an MSI (contact or noncontact) than those with higher scores. The mean summed score of 15.84 ± 1.73 for all athletes was higher than that reported by Chorba et al²³ for a similar group of participants (NCAA Division II females: 14.30 ± 1.77) but lower than that reported for professional American football players (16.9 ± 1.70).¹⁶ Only 29 of 84 (34.5%) of our athletes had scores of ≤14 versus 16 of 38 (42.1%) in the study of Chorba et al,²³ who noted that low FMS summed scores (≤14) did predict injury. The ROC curve analysis determined a cutoff score of 16 for our sample, which is closer to the finding of 17 from Weise et al.³¹ The area under the curve was 0.363, which reflected a less than 50% chance of predicting injury with the composite FMS score. Furthermore, the RR associated with this cutoff score was 0.29, indicating that those with composite scores of ≤16 were less likely to become injured. The composite score does not appear to be a good predictor of MSIs for this sample.

Sensitivity and specificity for the composite scores were 26% and 59%, respectively, for all participants. Our results contrast with the calculated sensitivity of 58% and specificity of 74% for the Chorba et al²³ study. Both investigations showed that the FMS composite score was better at ruling in injury when the sum score was ≤14 than it was at ruling out injury when the sum score was >14. Sensitivity in our study increased 81.5% when asymmetries and low scores on individual tests were considered independently of summed scores. Asymmetries or low individual movement test scores in the FMS may be present in athletes who have a composite score >14. This may be a reason for the high number of false-negative results associated with FMS composite scores greater than 14. Another consideration for the low sensitivity and the lower specificity associated with the FMS composite score in our sample is that injury risk is likely multifactorial. Movement dysfunction is likely not the only factor predisposing an athlete to injury. Other variables, such as history of previous injury, training load, body composition, fitness levels, and extrinsic factors, also contribute to MSI risk.^{1–7,9–12,27,28,32,39} Investigators²⁵ who combined FMS scores with Y-Balance scores and a history of previous MSI were able to identify Division III athletes at high risk for noncontact lower extremity injury. Previous authors who reported that a composite score of ≤14 was a predictor of MSI accounted for previous injury²⁷ or studied different cohorts (professional National Football League athletes,^22,33 females only,²³ or male Marine officer candidates²⁴), which may have been responsible for the different outcomes. Furthermore, our definition of MSI needs to be acknowledged. We included MSIs resulting from both contact and noncontact mechanisms, which represented 30.9% (29 of 94) and 69.1% (65 of 94), respectively, of total MSIs. Our rationale for including both was that not all contact injuries are out of the athlete's control and that he or she may, in part, be in a position to be contacted or to fall because of a faulty underlying movement pattern. For example, a player who falls on an outstretched hand and sprains an elbow while slide tackling in soccer may have asymmetries in the inline lunge and active straight-leg raise that could have put the player at risk. We do not know if dysfunction on the FMS tests is related to proficiency in sport-specific skills such as cutting, running, or landing.

Recent attention has been given to performance on the individual tests that compose the FMS and their relationship to injury.^16,29 Kazman et al²⁹ showed through measures of internal consistency and factor analysis that the meaning of the summed score is actually unclear because the individual FMS tests do not appear to represent a unitary construct. They suggested that sports medicine professionals focus more on the individual movement scores, as was originally intended by the developers of the screen.²⁹ This was the aim of our study, to examine the role of asymmetries and limited movement patterns (scores of 1) in the individual movements on injury risk. In a study involving 238 American professional football players over 1 season, Kiesel et al¹⁶ reported that players who had at least 1 asymmetry on the FMS were 1.8 times more likely to sustain an MSI than those who did not. Asymmetries in other functional movement tasks, such as dynamic balance (Star Excursion Balance Test),¹⁹ running biomechanics (eg, step length, impact peak, loading rate),¹⁵ jump-landing biomechanics (knee valgus, knee rotational moments),⁴⁰ and single-leg postural stability,⁴⁰ have been linked to MSI occurrence. However, these tasks were examined for their predictive ability regarding only lower extremity injuries. When all 7 tests are performed, the FMS involves total body patterns and therefore may be better in predicting a wider range of MSIs.

Our inclusion of limited movement patterns (score = 1) may account for the larger odds ratio (5.27). We included limited movement patterns because we hypothesized that significant impairments must be present to prevent the body from moving through a basic functional movement pattern. Movement dysfunctions in athletic populations have been associated with injury.^41–43 If movement in basic patterns is dysfunctional, then the higher demands of athletic movements may also be impaired and could contribute to injury potential. Additionally, 2 of the individual FMS tests, the stability push-up and overhead squat, are not evaluated for asymmetry, as no side-to-side comparisons are made. Our findings demonstrated that limited basic movement patterns were associated with athletic injury.

This study is not without limitations. First, not including a history of previous MSI as a variable may have affected the generalizability of the results. We had some control over this because we excluded data from participants who were experiencing ongoing or recurrent injuries. However, the role of previous injury history in those who were included is unknown. This limitation may not have detracted from the overall finding that the presence of asymmetric or limited movement patterns on the individual tests are better predictors of MSI than the composite FMS score. Second, the sample size of 84 was small and included athletes from only 3 sports (rowing, soccer, and volleyball) because of time constraints during the PPEs. Thus, the risk estimates should be interpreted with caution in comparison with studies of larger samples. However, as with the previous limitation, this may not have affected the overall finding supporting the usefulness of scores on the individual FMS movement patterns.

Functional movement is the ability to produce and maintain a balance between mobility and stability along the kinetic chain while performing fundamental movement patterns with accuracy and efficiency.⁴⁴ The FMS is one method of identifying movement deviations that are base level and, although not sport specific, underlie sport movements and tasks. Identifying movement deviations can be critical not only in recognizing an individual's risk for injury but also in designing intervention programs for that individual. Performance on the FMS appears to be modifiable. For example, Bodden et al⁴⁵ and Kiesel et al³³ improved FMS scores (increased the composite score in mixed martial artists and decreased asymmetries in professional American football players, respectively) using a standardized corrective exercise program. The customary precautionary measure for recognizing any preexisting condition that may lead to injury is the PPE. The PPE includes a medical and family history, orthopaedic (joint- and muscle-specific) examination, and general medical screen (eg, cardiorespiratory system, vision) in an attempt to identify conditions that may disqualify the athlete or predispose him or her to injury or illness. However, the PPE may fall short in the identification and prevention of injuries caused by limited functional movements. The FMS can fill that critical gap between the PPE and performance training.

CONCLUSIONS

We examined the presence of asymmetrical or limited (score = 1) movement patterns on the individual tests of the FMS. Division II athletes with summed scores of ≤14 were not at greater risk of MSI than those with higher scores. However, those with an asymmetry or score of 1 on any of the 7 individual FMS tests were at 2.73 times greater risk of MSI. Performance on the FMS is an independent factor that should be considered when assessing injury risk from a multifactorial perspective.