RELATIONSHIP BETWEEN FUNCTIONAL MOVEMENT SCREENING SCORE AND HISTORY OF INJURY

Abstract

Background and Aim:

The Functional Movement Screen (FMS™) is a screening instrument that evaluates selective fundamental movement patterns. The main aim of this study was to investigate the relationship between the FMS™ score and history of injury, and attempt to determine which active students are prone to injury.

Methods:

One hundred physically active (50 females and 50 males) students, between 18 and 25 years of age, with no recent (<6 weeks) history of musculoskeletal injury were recruited. All participants performed the FMS™ and were scored using the previously established standardized FMS™ criteria. The chi square, independent t‐test, one‐way analysis of variance, and POSTHOC Bonferroni tests were used for data analysis with a preset alpha value of p < 0.05.

Results:

Of the 100 subjects, 35 suffered an acute lower extremity (ankle = 20, knee = 15) injury in practice or competition. An odds ratio was calculated at 4.70, meaning that an athlete had an approximately 4.7 times greater chance of suffering a lower extremity injury during a regular competitive season if they scored less than 17 on the FMS™. There were statistical differences between the pre‐season FMS™ scores of the injured and non‐injured groups, the ankle injury, knee injury, and non‐injured groups, and also between contact injury, non‐contact injury, and non‐injured groups.

Discussion and Conclusion:

This cross‐sectional study provides FMS™ reference values for physically active students, which will assist in the interpretation of individual scores when screening athletes for musculoskeletal injury and performance factors. More research is still necessary before implementing the FMS™ into a pre‐participation physical examination (PPE) for athletics, but due to the low cost and its simplicity to implement, it should be considered by clinicians and researchers in the future.

Level of Evidence:

2B

Introduction

Athletics has always carried an inherent risk for injuries. Large numbers of participants in sport can result in a high number of injuries. Determining an individual’s ability to participate in sporting events requires careful evaluation of the rigors and demands on the athlete within his/her designated sport.1 One of the primary responsibilities of the sports medicine staff at all levels of athletics is to attempt to prevent injury.2 Several authors have evaluated risk factors that contribute to overall injury rates in athletics such as previous injury, body mass index, muscle flexibility, and biomechanics during athletic movements. Intrinsic risk factors include: agonist/antagonist muscle ratios for strength and endurance, structural musculoskeletal abnormalities, neuromuscular control, core weakness, contra‐lateral muscular imbalances.1‐3 Most research in this area has investigated these factors individually, but recently attention has been directed at the multifactorial influence of several risk factors. The potential to screen athletes for risk of injury during a pre‐participation physical examination (PPE) could be extremely helpful and important. One instrument that may be useful for these purposes is the Functional Movement Screen™ (FMS™).2,3

The FMS™ is designed to evaluate a variety of functional movements that are proposed to be necessary to participate in higher‐level functions such as sport or recreation. The FMS™ requires the ability to move through up to three planes of movement during the assessment movements. The FMS™ is assessed by means of qualitative and quantitative information regarding specialized motions related to functional activities. The tests are often utilized for assessing the athlete’s pain, muscle strength, lower extremity joint stability in multiple planes of movement, muscle flexibility, balance, and proprioception.4

The FMS™ is comprised of a series of movements designed to assess the quality of fundamental movement patterns and presumably identify an individual’s functional limitations or asymmetries. Previous small studies have demonstrated that low FMS™ scores (≤14) may associated with serious injury in American football players and that FMS™ scores can be improved following a standardized intervention.2,5 In addition, a large interventional study in firefighters suggested that FMS™ assessment followed by an eight week program designed to enhance functional movement reduced time lost to injury by 62% when compared with historical injury rates.6 The assessment of fundamental movements is an attempt to pinpoint deficient areas of mobility and stability that may be overlooked in the asymptomatic active population. The difficulty in preventing injury seems to be directly related to the inability to consistently determine those athletes who are predisposed to injuries. Meeuwisse suggested that unless specific markers are identified for each individual, determining who is predisposed to injuries would be very difficult.7 Numerous sports medicine professionals have suggested the need for specific assessment techniques that utilize a more functional approach in order to identify movement deficits.7‐10

The FMS™ may be included in the pre‐placement/PPE or be used as a stand‐alone assessment technique utilized to determine deficits that may be overlooked during the traditional medical and performance evaluations.12 In many cases, muscle flexibility and strength imbalances may not be identified during the traditional assessment methods. These problems, previously acknowledged as significant risk factors, can be identified using the FMS™.2 This movement‐based assessment serves to pinpoint functional deficits (or biomarkers) related to proprioceptive, mobility, and stability weaknesses.7‐11 Therefore, the main aim of this study was to examine the relationship between FMS™ score with history of injury, and attempt to determine which active students are prone to injury.

Methods

Subjects

The study employed a prospective cross‐sectional design and a reliability component. A convenience sample of one hundred physically active students (50 females and 50 males), weight 69.44± 5.84 kg, height 172.69± 7.26 cm, and age 22.56± 2.99 years who participated in soccer, handball, and basketball for at least for 5 years were recruited from a tertiary student population. Subjects were included in the study if they participated in sports at a competitive or recreational level regularly (>1.5 h/week) in a sport for at least 3 years. Exclusion criteria included the use of a mobility aid or a prophylactic device (e.g. knee brace) or if they had reported a recent (within the past 6 weeks) musculoskeletal or head injury that was likely to affect their motor performance on the FMS™. Also, subjects with known physical impairments were excluded. The University of Tehran Human Ethics Committee approved the study and written informed consent was obtained from all subjects prior to data collection. The subjects were required to read and sign consent forms approved by The University of Kharazmi Institutional Review Board.

Instrumentation

Instruments utilized included the FMS™ Kit which includes a measuring device, hurdle step, stretch bands, measuring stick necessary for the deep squat, hurdle step, in‐line lunge, shoulder mobility, active straight leg raise, trunk stability push up, rotary stability tests.

Data Collection Procedures

The data were collected by two members of the research team, both physical therapists. A pilot study was conducted with 20 participants in order to achieve a reliable level of agreement between the two test raters, which resulted in Kappa values >.80 for all tests that comprise the FMS™.

The average inter‐tester reliability between tester was high for FMS™ tests (ICC = .877‐.932).

The FMS™, developed by Cook and Burton, was used in the study. The subjects tested in the study were evaluated on the FMS™ using the standard 0‐3 ordinal system. A score of 3 was given for performing the specific movement perfectly, a 2 was given when the movement was completed with some compensatory movements observed, a score of 1 was given when the subject could not complete the movement, and a score of 0 was given if pain was present during the movement. The FMS™ includes seven movement tests: the deep squat, hurdle step, in‐line lunge, shoulder mobility, active straight leg raise, trunk stability push‐up, and rotary stability tests.

The composite score for all seven movements of the FMS™ was recorded and then compared with the injury documentation and tracking of the lower extremity that occurred throughout the season, which was achieved by the teams’ specific athletic trainer and sports medicine staff. The injury documentation was completed after each team exposure, where an exposure was considered one athlete per practice or game (based on the time of session/practice/game). Any acute lower extremity injury that occurred and kept the athlete out of participation for one or more full consecutive exposures was counted as an injury. If an athlete suffered multiple or repeated acute injuries during the competition season, only the first injury incident was included in this analysis. Therefore, an athlete could not appear more than once in the “injured” group’s analysis.

Data analysis

To determine if there was a significant difference in FMS™ scores between athletes that were injured and athletes that were not injured during the regular competitive seasons, independent t‐tests were performed. To determine if there was a significant difference between sports, body parts of injured subjects, and mechanism of injury, one‐way analyses of variance were used. To determine cut‐off scores, a receiver‐operator characteristic (ROC) curve was used to plot sensitivity (true positives) versus 1‐specificity (false positives) for the screening test.^A A 2x2 contingency table was produced in order to dichotomize the athletes that suffered an injury and those who did not, as well as those who were above or below the specified cutoff score. From the table, odds ratios, likelihood ratios, sensitivity and specificity were calculated. Chi‐square tests were used to evaluate if there were any significant differences between males and females in the distribution of scores for the different tests. The Intra‐class Correlation Coefficient (ICC model 3,1) was used to establish the inter‐rater reliability for the FMS™ composite score, and the unweighted Kappa statistic was used to establish the inter‐rater reliability measurement for each item. The inter‐rater reliability data were interpreted according to the categories defined by Landis and Koch. A Kappa value over 86% represents excellent agreement. All calculations were performed using SPSS (version 16.0) and the a priori level of significance was set at p ≤ 0.05.

Results

One hundred subjects participated in the study, 50 females and 50 males. Table 1 presents the subject’s demographic information. Table 2 presents the inter‐rater reliability results for the individual FMS™ tests, with levels of agreement ranging from substantial to excellent. The inter‐rater reliability (ICC) of the composite score for both testers was .92, which indicates excellent reliability.

Table 1.

Demographic characteristics of subjects.

Sport	Weight (kg)	Height (cm)	Age (yrs)	* Baecke Sport Score
Basketball (Nfemale=14, Nmale=14)	69.19±6.16	174.14±7.13	23.21±2.12	21±1.04
Soccer (Nfemale=18, Nmale=17)	68.12±7.22	172.42±6.23	22.56±3.54	22±2.03
Handball (Nfemale=18, Nmale=19)	71.03±4.15	171.52±8.44	21.92±3.33	22±2.12

^*The optimal score for Baecke Sport Score is 21 to 23; persons scoring below 21 are not considered active, while persons scoring above 21 are considered to be active. For the purpose of this study, the authors used subjects with scores from 21‐23 to denote “active” participants.

Table 2.

Inter-rater reliability of individual FMS™ tests.

Variable	Agreement %	Kappa	Level of agreement
Deep squat	100	1.00	Excellent
Hurdle step R	88	0.85	Substantial
Hurdle step L	92	0.86	Substantial
Hurdle step Final	96	0.91	Excellent
Lunge R	90	0.86	Substantial
Lunge L	93	0.90	Excellent
Lunge Final	96	0.91	Excellent
Shoulder mobility R	95	0.91	Excellent
Shoulder mobility L	96	0.93	Excellent
Shoulder mobility Final	98	0.92	Excellent
ASLR R	92	0.86	Substantial
ASLR L	93	0.89	Substantial
ASLR Final	96	0.94	Excellent
Trunk Stability Push‐up	100	1.00	Excellent
Rotary stability R	100	1.00	Excellent
Rotary stability L	100	1.00	Excellent
Rotary stability Final	100	1.00	Excellent

R= Right; L= Left

The composite mean scores on the FMS™ for females, males and the entire sample were 16.3± 1.2, 16.9± 1.9, and 16.7± 1.8 respectively. These scores are presented in Table 3.

Table 3.

FMS™ individual test scores for males and females.

Variable/Test	Males	Females	p‐values
Deep squat	2.25±0.22	2.28±0.5	0.091
Hurdle step	2.04±0.1	2.12±0.31	0.213
Lunge	2.18±0.19	2.17±0.28	0.612
Shoulder mobility	2.18±0.1	2.56	0.021¥
ASLR	2.01±0.5	2.44±0.3	0.013¥
Trunk Stability Push‐up	2.32±0.3	2±0.17	0.027¥
Rotary stability	2.49±0.36	2.17±0.48	0.023¥

^¥Significant differences.

Differences observed between males and females in trunk stability push‐up, the rotary stability, active SLR, and shoulder mobility tests were significant. There were significant differences between football, handball, and basketball sport groups. Basketball players had lower scores in all seven FMS™ tests.

For all subjects, a cut‐off score of 17 was used that maximized sensitivity (0.645) and specificity (0.780). These findings resulted in a positive likelihood ratio (Sensitivity/1‐Specificity) of 2.46 and a negative likelihood ratio (1‐Sensitivity/Specificity) of 0.621 (Table 4). An overall odds ratio was calculated at 4.70, meaning that an athlete has an approximately 4.7 time greater chance of suffering a lower extremity injury during a regular season by scoring less than 17 on the FMS™. By using the cut‐off score of 17 a 2× 2 contingency table was created to dichotomize the subjects by their FMS™ score and injury status after the regular competitive season (Table 5).

Table 4.

Odds Ratio and Sensitivity/Specificity calculations by FMS Scores.

	OR (CI)	Sensitivity	Specificity
No injury	1.4 (1.1–2.1)	51.4	82.3
Knee injury	2.6 (0.5–4.1)	14.3	93.4
Ankle injury	3.0 (0.7–5.1)	13.4	94.8

Table 5.

2x2 Contingency Table for FMS™ score data

	Injured	Non‐injured
FMS™ score ≤ 17	22	24
FMS™ score > 17	20	34

There was a statistically significant difference between the pre‐season FMS™ scores of the injured and the non‐injured groups (${\text{t}}_{100}=3.60$; p = .005).

A one‐way ANOVA revealed a statistically significant difference between the ankle injury group, knee injury group, and no injury group ($F_{\left(2,99\right)}=3.43$; p = .030). The Bonfferoni post hoc testing demonstrated that the differences existed between the ankle injury group and no injury group (p = .021), as well as between the knee injury group and no injury group (P = .030); but not between the ankle injury group and knee injury group (p = .101).

The one‐way ANOVA did reveal statistically significant differences between the groups with a contact injury, non‐contact injury or no injury ($F_{\left(2,99\right)}=2.11$; p = 0.010). The Bonfferoni post hoc testing demonstrated that differences existed between the non‐contact injury and no injury groups (p = .032), as well as between the contact injury and no injury groups (p = .013); but not between the contact and non‐contact injury groups (p = .217).

Discussion

This study was designed to determine if a battery of functional assessment tests relating to athletic performance could be used to predict lower extremity injury risk in a select group of subjects. Because of compensations which may occur along the kinetic chain during movements, isolation of individual body parts may be necessary to determine if the subject is at, above, or below average in a certain area. The most difficult challenge will be to determine which tests are the most appropriate to use during the screening process.

The composite score for all seven components of the FMS™ test was recorded and then compared with the injury documentation and tracking of the lower extremity injuries that occurred throughout the season by the teams’ specific athletic trainer and sports medicine staff. The mean composite score reported in this study is lower than that reported for a group of professional male football players2. It might be expected that professional football players score better than the average athlete due to their intensive training regimens, however, in a subsequent study on a similar cohort the mean pre‐intervention composite score was 11.8 for “lineman” and non‐lineman.5 The difference may relate to the cohort studied, the specific training regimens undertaken by each team or familiarity with the FMS™ testing procedures. Cowen14 studied male and female firefighters whose mean baseline FMS™ score was also lower than the current study at 13.25. In the latter two studies the composite FMS™ score significantly increased following an exercise‐based intervention.

Based on this study, males were on average better on the trunk stability push‐up and the rotary stability tests than females, and females performed better on the active straight leg raise and the shoulder mobility items. The trunk stability push‐up is associated with upper body strength and stability (including core stability in the sagittal plane), the rotary stability test with transverse plane (rotational) core stability, the active straight leg raise with flexibility in the hamstring muscles, and the shoulder mobility test with range of motion in the shoulder complex and thoracic spine.12 The sex differential finding was supported by Kibler et al in a study that investigated 2107 athletes from a variety of sports inclusive of junior high to college levels.13 The rotary stability test demands trunk stability in the sagittal and transverse planes during asymmetric movement of the upper and lower extremities.12 The FMS™ training manual comments that it is difficult to obtain a score of 3 (only 1 subject did so in the present study) but it is included to capture elite performance. The authors believe that the proper and perfect implementation of this test is not applicable for all, and maybe only some professional athletes are able to perform this test without error. The rotary stability test does however provide the potential to measure change following a specific exercise‐training program targeted at asymmetric or multi‐planar trunk stability.

In this study 27% of the participants had a score of 14 or less which might indicate a potentially higher risk of injury. This is in comparison to the 22% of the professional football players in the Kiesel et al study2 and 89% in the subsequent study by Kiesel et al5 who scored below a 14 and were statistically deemed to be more likely to be injured. The cutoff score of 14 was determined in a study on 46 professional football players but, because of the small sample size and the fact that the target group didn’t represent a general athletic population, the authors of the current study suggest that this cutoff value should be used with caution. Further studies need to be conducted on other athletic and occupational groups before determining a substantiated cutoff value.6,15,16

Preliminary studies with the FMS™ have attempted to examine risk of injury in a small number of NFL football players.2,13 Kiesel et al retrospectively analyzed the relationship between FMS™ scores for National Football League (NFL) football players and the likelihood of serious injury.2 FMS scores were obtained before the start of the season for 46 NFL players, and a score of ≤14 was found to positively predict serious injury with a specificity of 0.91 and sensitivity of 0.54; the odds of sustaining a serious injury was 11.7 times higher in those with an FMS score ≤14 compared with those with a score >14. Kiesel et al also noted lower scores among those who had been injured compared to those without injury.2 In the present study, when the authors compared entry FMS™ scores by no injury versus any injury, the scores were the same and the odds ratio for sustaining a serious injury was 2.0; the sensitivity and specificity were 0.67 and 0.90, respectively. Interestingly, two groups have reported that FMS™ did not predict injury: one study was with 60 marathon runners17 and another was on 112 basketball players.18 Hoover et al.17 reported 8.3% sensitivity and 94.5% specificity for marathon runners, whereas Sorenson’s18 data yielded a sensitivity and specificity of 53.8% and 52.3%, respectively, for basketball players. The low sensitivity is problematic because sensitivity above 50% is desirable so those predisposed to injury can be identified early and potentially rehabilitated before injury. Although some reports of specificity are high, this is in large part explained by the small proportion of the cohort with scores ≤14.2,4,6,17‐19

The current study displays the need for additional research of this nature to be conducted. Athletics has transformed into a business at the collegiate and professional levels. There seem to be unlimited possibilities as far as pre‐season screening tests and collection of injury data in multiple sports at the collegiate and professional level.

Conclusion

The results of the current study demonstrated that pre‐season FMS™ scores show a relationship with injury in Kharazmi University athletes. Furthermore, those who scored less than 17 on the FMS were 4.7 times more likely to sustain an injury of the lower extremity. More research is still necessary before implementing the FMS™ into a PPE for athletics, but due to the low cost and simplicity of implementation, it should be considered as a screening tool by clinicians and researchers in the future. As more evidence becomes available on the FMS™, it could be an effective tool to use when screening athletes and determining potential risk for injury.