Physical Therapy in Sport xxx (2013) 1e7
Contents lists available at ScienceDirect
Physical Therapy in Sport journal homepage: www.elsevier.com/ptsp
Original research
Functional movement screen test: A reliable screening test for young elite ice hockey players Elizabeth Parenteau-G a, Nathaly Gaudreault a, *,1, Stéphane Chambers a, Caroline Boisvert a, Alexandre Grenier a, Geneviève Gagné a, Frédéric Balg b a
School of Rehabilitation, Faculty of Medicine and Health Sciences, Université de Sherbrooke, 3001, 12e Avenue Nord, Sherbrooke, QC, Canada J1H 5N4 Departement of Orthopaedic Surgery, Faculty of Medicine and Health Sciences, Université de Sherbrooke, 3001, 12e Avenue Nord, Sherbrooke, QC, Canada J1H5N4 b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 7 March 2013 Received in revised form 9 August 2013 Accepted 3 October 2013
Objectives: To determine inter-rater and intra-rater reliability of the Functional Movement Screen (FMS) test among young elite hockey players. Design: Reliability study. Setting: Inter-rater reliability was evaluated by two raters in the field. All performances were videotaped. Two other raters evaluated the videos once and then again 6 weeks later to determine intra-rater reliability. A weighted kappa statistic was used to analyze intra-rater and inter-rater reliability of each FMS sub-test, while an intra-class correlation coefficient (ICC) was calculated for the total score. Participants: Twenty-eight male hockey players aged 13e16. Main Outcome: FMS total and sub-tests scores. Results: The video raters demonstrated excellent intra-rater reliability for the total score, with an ICC of 0.96 (95% CI; 0.92e0.98) and 0.96 (95% CI; 0.91e0.98). The field raters achieved excellent inter-rater reliability for the total score, with an ICC of 0.96 (95% CI; 0.92e0.98). Sub-test analysis showed good agreement among all four raters for five of the seven main sub-tests. Conclusion: FMS is a reliable test for young elite hockey players. Further research should be done to assess the predictive validity of the FMS test within this population so that physiotherapists may eventually use it as an injury prevention tool. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Functional test Ice hockey Reliability
1. Introduction Ice hockey is one of the most popular sports practiced by teenagers in Canada (Governement of Canada - Canadian Heritage, 2010) and young hockey players aged 12e17 are highly vulnerable to injuries (Hostetler, Xiang, & Smith, 2004). Although the majority of ice hockey injuries are caused by contact, nearly 15% of injuries are non-contact musculoskeletal (MSK) injuries (Agel, Dompier, Dick, & Marshall, 2007) caused by overuse or poor movement patterns (Emery, Meeuwisse, & Powell, 1999; Lorentzon, Wedren, Pietila, & Gustavsson, 1988). Non-contact injuries in hockey players seem to mostly affect their pelvic and abdominal muscles and tendons (Emery et al., 1999; Lorentzon et al., 1988). Most of
* Corresponding author. Tel.: þ1 819 820 6868x12910; fax: þ1 819 820 6864. E-mail address: [email protected] (N. Gaudreault). 1 The corresponding author agrees to allow the publication of her contact information.
these non-contact injuries are related to the high velocity movements executed by the players, such as slap shots and skating. Performance of these movements not only requires significant pelvic and abdominal core strength (static stability) but coordinated recruitment patterns of these muscles when performing dynamic peripheral movements (dynamic stability) (Hodges & Richardson, 1997; Leetun, Ireland, Willson, Ballantyne, & Davis, 2004; Nadler, Malanga, Bartoli, Feinberg, Prybicien, & Deprince, 2002). According to Leetun et al. (2004), pelvic and abdominal muscle injuries in hockey players might be reduced if core strength and movement pattern deficits were detected earlier. Core strength and muscle activity can be evaluated in many ways. Measures by dynamometer testing (Hanada, Hubley-Kozey, McKeon, & Gordon, 2008), ultrasonography (Brown & McGill, 2010) and by electromyographic studies (Hibbs, Thompson, French, Hodgson, & Spears, 2011) are precise but are not clinically suitable because of high costs and the equipment required. In fact, clinicians are more likely to use standardized tests such as the Front Abdominal Power Test (Cowley & Swensen, 2008), the Double Leg
1466-853X/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ptsp.2013.10.001
Please cite this article in press as: Parenteau-G, E., et al., Functional movement screen test: A reliable screening test for young elite ice hockey players, Physical Therapy in Sport (2013), http://dx.doi.org/10.1016/j.ptsp.2013.10.001
2
E. Parenteau-G et al. / Physical Therapy in Sport xxx (2013) 1e7
Lowering Manœuvre (DLLM) (Ladeira, Hess, Galin, Fradera, & Harkness, 2005) or the Functional Movement Screen (FMS). Firstly, the Front Abdominal Power Test assesses abdominal strength in the sagittal plane. This test has shown good reproducibility (ICC 0.95) but might show some limitations in the evaluation of hockey players since it does not address stability in multiple planes. Secondly, the DLLM also has good reproducibility (ICC 0.93) but does not appear to be a valid measure of strength and stability when compared to EMG studies (Ladeira et al., 2005). The FMS was developed to evaluate trunk stability and motion quality of fundamental movement patterns (Cook, Burton, & Hoogenboom, 2006; Hanada et al., 2008) with the ultimate goal of predicting non-contact injuries in athletes. Fundamental movements are basic movements that can simultaneously assess range of motion, stability and balance. This test is currently used in many pre-season screening protocols in different sports (Cook et al., 2006; Kiesel, Plisky, & Butler, 2011; Kiesel, Plisky, & Voight, 2007). The FMS assesses range of motion, symmetry and stability during seven main functional movements, and is rated on a 21 point scale. It has been shown that athletes who obtain a score less than or equal to 14/21 are at higher risk of injury (Kiesel et al., 2007). Recent studies with civilian (Onate et al., 2012), military (Teyhen et al., 2012), professional football players (Minick, Kiesel, Burton, Taylor, Plisky, & Butler, 2010) and elite soccer players (Frohm, Heijne, Kowalski, Svensson, & Myklebust, 2012) report substantial to excellent reliability of the FMS, with intra-class correlation coefficients (ICC) for the total score ranging from 0.76 to 0.98. Although these results appear to be promising in identifying athletes at risk of non-contact injuries due to poor stability and movement pattern deficits, further research is required with respect to FMS clinimetric properties. The reliability of the FMS test has never been documented in a population of young hockey players, despite the popularity of this sport. FMS reliability has also never been assessed for field rating, which in this study refers to a context very similar to a pre-season evaluation set-up for this population. Therefore, the aim of our study is to determine interrater (field rating) and intra-rater (video rating) reliability of the FMS test among young elite ice hockey players. 2. Materials and methods 2.1. Study design Inter-rater and intra-rater studies were conducted to determine the reliability of the FMS performed on young hockey players. 2.2. Participants A sample of 30 male hockey players was recruited from the hockey sports study program of a high school located in Sherbrooke, Quebec. The first step was to contact the program coordinator to explain the research project. Flyers describing the purpose and procedures of the study were then distributed to all hockey players involved in Bantam AA (13e14 years old) and Midget Espoir (15e17 years old) leagues. In Quebec, these categories consist of the best players and are now the cornerstone for identifying players expected to be selected for national teams and to be noticed by scouts for professional teams. A letter was also sent to their parents. Players who wanted to participate were invited to contact the project coordinator who verified their study eligibility. Thirty participants satisfied the inclusion and exclusion criteria defined in Table 1. Institutional ethical approval was obtained from the Etienne-Le Bel Clinical Research Centre Review Board, the 30 eligible participants and their parents willingly signed the consent form and the rights of the participants were protected. However,
two of the participants were sick on the day the data was collected. Therefore, 28 players participated in the study. 2.3. Description of the FMS The FMS test mainly evaluates trunk stability, range of motion, motion quality and symmetry during the performance of basic functional movements. The test includes the seven sub-tests depicted in Fig. 1 and three clearing exams: deep squat, hurdle step (right and left), active straight leg raise (right and left), rotary stability (right and left) and spinal flexion pain test (clearing exam); in-line lunge (right and left), trunk stability push-up and back extension pain test (clearing exam); shoulder mobility (right and left) and shoulder pain test (clearing exam). The sub-tests are evaluated using a 0e3 ordinal scale: 0 ¼ pain during the movement; 1 ¼ participant is unable to execute the correct movement; 2 ¼ participant executes the movement with compensations; and 3 ¼ participant executes correct movement without pain or compensations. The clearing exams were scored using a dichotomous variable: presence of pain (yes or no). The participants were given three attempts at each sub-test, and the evaluator had to rate only the best performance. A positive clearing exam (pain during the movement) automatically gave a score of 0 for the related sub-test. A summed total score was then calculated. 2.4. Data collection The data were collected at three different sessions as illustrated in Fig. 2: the first session allowed field inter-rater reliability data to be collected; the second and third sessions served to collect data for the intra-rater reliability at time 1 (T1) and time 2 (T2). The examiners (n ¼ 4) consisted of three senior physiotherapy students enrolled in their last year of a professional Master’s degree physiotherapy training program and one physiotherapist. Each evaluator had obtained official FMS training certification and had practiced the FMS prior to data collection. 2.5. First session: inter-rater reliability To collect data for the field inter-rater reliability study, all participants were invited to come to the Université de Sherbrooke’s rehabilitation clinic. Prior to the FMS evaluation, each participant completed a questionnaire to collect data on their sports experience and history of musculoskeletal (MSK) non-contact injuries. These data served to describe the participants’ characteristics. Each participant was then asked to perform the seven FMS sub-tests following standardized instructions (described in the FMS user manual). A random number generating software was used to randomly assigned the order of the sub-tests for each participant. One randomly selected pair of examiners (i.e., field raters) independently rated the participants during their performance of each sub-test using a scoring sheet. Examiners were blinded from each other’s scores. The scores of those two examiners were then used for inter-rater analysis. A front view camera and two side view
Table 1 Eligibility criteria. Inclusion
Exclusion
Male Member of an elite hockey sport-study program (Bantam AA or Midget Espoir leagues) 13e17 years old
Symptomatic (acute) musculoskeletal injury that may interfere with their ability to participate in regular hockey training/playing at the time of the FMS evaluation
Please cite this article in press as: Parenteau-G, E., et al., Functional movement screen test: A reliable screening test for young elite ice hockey players, Physical Therapy in Sport (2013), http://dx.doi.org/10.1016/j.ptsp.2013.10.001
E. Parenteau-G et al. / Physical Therapy in Sport xxx (2013) 1e7
3
Fig. 1. Illustration of the FMS sub-tests: 1) deep squat; 2) hurdle step; 3) active straight leg raise; 4) rotary stability; 5) in-line lunge; 6) trunk shoulder mobility and 7) stability push-up.
cameras (Canon ixia hfs200, Mississauga, Ontario, Canada) were installed during the test to capture the participants’ performance. 2.6. Second session: intra-rater reliability T1 The videos recorded during the first session were displayed so that the front view and both side views could be viewed simultaneously on a computer screen. This allowed the other two examiners (video rater 1 and 2) to independently rate each participant’s performance using the FMS sub-tests. The videos were projected in a closed room with a research assistant who made sure that there was no communication between the raters so that they were blinded from each other’s scores. This session was held over the course of one day.
Fig. 2. The data were collected at three different sessions: the first session allowed field inter-rater reliability data to be collected; the second and third sessions served to collect data for the intra-rater reliability at time 1 (T1) and time 2 (T2). A six-week time period separated T1 and T2 sessions.
2.7. Third session: intra-rater reliability T2 Six weeks following the first video rating session, the same pair of examiners rated the same 28 videos showing each participant performing the FMS sub-tests under the same condition as T1. The time elapsed was sufficient to control for recall bias. The scores obtained during T1 and T2 were used for intra-rater analysis. The data obtained at T2 were not used to document inter-rater reliability because inter-rater reliability has already been documented using video rating (Kiesel et al., 2011; Minick et al., 2010). The use of two video raters was to increase the variability in the study design for intra-rater reliability.
3. Statistical analysis Data were analyzed using SPSS (Version 18) and StatXact (Version 6) software. Descriptive statistics were used to describe the participant sample: mean and standard deviations were calculated for age, hockey experience and FMS sub-tests and total scores; percentage was used to depict the proportion of players who reported having a past history of MSK injury and those who were identified as being at risk of a non-contact MSK injury by having an FMS total score lower than 14/21 (O’Connor, Deuster, Davis, Pappas, & Knapik, 2011). To estimate inter- and intra-rater reliability (see Fig. 2), the percentage of agreement was calculated and quadratic weighted Cohen kappa coefficients were used for each FMS sub-test score. Using a quadratic weighted Cohen kappa coefficient is stronger than calculating the percentage of agreement between raters because it takes random agreement into account. The reliability results for each sub-test were classified as follows based on the work of Landis and Koch (1977): almost perfect reliability, 0.81e1.0; substantial reliability, 0.61e0.80; moderate reliability, 0.41e60; fair reliability, 0.21e0.40; and slight reliability,
Please cite this article in press as: Parenteau-G, E., et al., Functional movement screen test: A reliable screening test for young elite ice hockey players, Physical Therapy in Sport (2013), http://dx.doi.org/10.1016/j.ptsp.2013.10.001
4
E. Parenteau-G et al. / Physical Therapy in Sport xxx (2013) 1e7 Table 2 Participant characteristics.
Table 4 Inter-rater reliability.
Characteristics
FMS sub-tests Mean (SD) 14.89 0.99 63.8 10.3 171.0 5.4 10.28 1.21
Age (years) Weight (kg) Height (cm) Hockey experience (years) n
%
Players who reported having a history of a non-contact MSK injury - In the last year 14 50% - More than a year ago 14 50% Players who had an FMS score lower than 14 17 60.7%
0.00e0.20. However, weighted Cohen Kappa coefficients could not be used for the clearing exams because of the dichotomous nature of the score (presence of pain; yes or no) and the fact that a nonweighted Kappa test was used for these sub-tests. Then, intrarater and inter-rater reliability for the FMS total score was determined by using intra-class correlation coefficient (ICC) analysis. Interpretation of ICC analysis was based on the criteria proposed by Onate and Dewey (Onate et al., 2012): high reliability, 0.90e0.99; good reliability, 0.80e0.89; fair reliability, 0.70e0.79; and poor reliability, 0.00e0.69.
4. Results The characteristics of the participants who performed the evaluation are described in Table 2 and descriptive analysis for the FMS scores are presented in Table 3. As observed, the mean FMS total score was 12.64 (with scores ranging from 5 to 19). Moreover, close to two-thirds of the participants (60.7%) were identified as being at risk of a non-contact MSK injury according to Kiesel’s FMS risk injury threshold score of 14/21 on the basis that they had a total FMS score lower than 14 (Kiesel et al., 2007; O’Connor et al., 2011).
4.1. FMS field rating inter-rater reliability Table 4 shows inter-rater percentage of agreement and weighted kappa coefficient values for each sub-test and the ICC value for the FMS total score. The percentage of agreement for each FMS sub-test ranged from 67.9 to 100%, whereas the quadratic weighted kappa coefficient value ranged from 0.26 to 0.99. The quadratic weighted Cohen kappa coefficient was not calculated for the spinal flexion test (clearing exam) because the percentage agreement was 100%. Fig. 3 shows the correlation for the total FMS score obtained by the two field raters. Table 3 Descriptive analysis for FMS sub-tests and total scores. An analysis of field rater 1’s score sheet revealed the following data. FMS sub-tests
Deep squat Hurdle step Active straight leg raise Rotary stability In-line lunge Trunk stability push-up Shoulder mobility FMS total score
Number of players per score category 0
1
2
3
6 1 5 1 4 15 6
1 0 9 1 3 0 4
13 12 11 24 9 5 10
8 15 3 2 12 8 8
Mean SD FMS sub-test scores
1.82 1.09 2.46 0.69 1.43 0.92 1.96 0.51 2.04 1.07 1.21 1.37 1.71 1.11 12.643.65
Agreement (%) Kappaa (95% CI)
Deep squat 82.10 Hurdle step 67.90 Active straight leg raise 82.10 Rotary stability 89.30 Spinal flexion pain test (clearing exam) 100.00 In-line lunge 67.90 Trunk stability push-up 78.60 Back extension pain test (clearing exam) 96.40 Shoulder mobility 96.40 Shoulder pain test (clearing exam) 96.40 FMS total scoreb e
0.75 (0.46e1.00) 0.55 (0.30e0.80) 0.97 (0.90e1.00) 0.26 (0.18 to 0.67) e 0.77 (0.57e0.96) 0.90 (0.80e0.99) 0.92 (0.77e1.00) 0.99 (0.96e1.00) 0.87 (0.62e1.00) ICC (IC 95%) 0.96 (0.92e0.98)
a The kappa value corresponds to the quadratic weighted kappa coefficient, except for the clearing exams, for which non-weighted kappa coefficient tests were used. b ICC was used only for the FMS total score.
4.2. FMS video rating intra-rater reliability Intra-rater percentage of agreement and weighted kappa coefficient values for each sub-test and the ICC value for the FMS total score are presented in Table 5. For video rater 1, the agreement percentage ranged from 64.30 to 100.00% for each FMS sub-test, whereas the kappa value ranged from 0.65 to 1.00. For video rater 2, the agreement percentage ranged from 71.40 to 100.00% for each FMS sub-test, while the kappa value ranged from 0.76 to 1.00. The quadratic weighted Cohen kappa coefficient was not calculated for two clearing exams (spinal flexion test and back extension pain test) because the percentage of agreement was 100%. 5. Discussion As previously mentioned, the FMS was developed to evaluate trunk stability and motion quality while performing functional movements, with the ultimate goal of predicting non-contact injuries in athletes. Our results showed that half of the hockey players in our sample had already suffered a non-contact MSK injury, either in the past year or since they started playing hockey. This result can hardly be compared with existing evidence due to the paucity of scientific literature on the prevalence of non-contact injuries among young hockey players. This lack of information is quite surprising given that children and adolescents are particularly at risk of suffering injuries due to the immaturity of their musculoskeletal systems (Micheli & Klein, 1991; Wilkins, 1980). In adult hockey players, while most injuries occur during contact, the fact remains that almost 15% of injuries are not related to contact (Agel et al., 2007; Pettersson & Lorentzon, 1993). These results highlight the need to document the prevalence of non-contact injuries in young hockey players with more robust epidemiological studies. What is more troubling is that 60.7% of the young hockey players in this study had a total score lower than 14/21 (Kiesel et al., 2007; O’Connor et al., 2011) which means that 2 out of 3 players are more likely to suffer a noncontact MSK injury or reoccurrence of a previous injury. Positive effects of screening athletes who might be at risk of a non-contact injury with the FMS test and providing them with a specific exercise program has already been demonstrated (Kiesel et al., 2011). This leads us to believe that prevention of non-contact injuries with early screening and training programs adjustments is a promising avenue for young hockey players to consider. 5.1. Inter-rater reliability According to Onate and Dewey (Onate et al., 2012), we can say that with an ICC of 0.96, the total FMS score is highly reliable. When
Please cite this article in press as: Parenteau-G, E., et al., Functional movement screen test: A reliable screening test for young elite ice hockey players, Physical Therapy in Sport (2013), http://dx.doi.org/10.1016/j.ptsp.2013.10.001
E. Parenteau-G et al. / Physical Therapy in Sport xxx (2013) 1e7
5
5.2. Intra-rater reliability
Fig. 3. The high correlation for the total FMS score obtained by field rater 1 (on the abscissa) and field rater 2 (on the ordinate) is depicted here.
analyzing the reliability of the FMS sub-tests, the active straight leg raise test, the trunk stability push-up test, the shoulder mobility test and two of the clearing exams (back extension pain test, shoulder pain test) are considered as almost perfectly reliable. The deep squat and in-line lunge tests yielded substantial reliability, the Hurdle step test showed moderate reliability and the rotary stability test demonstrated fair reliability. These results are consistent with Minick’s inter-rater reliability study (Minick et al., 2010), which also revealed the rotary stability test to be a less reliable subtest when scored by novice raters (weighted Kappa ¼ 0.54). In the present study, we believe that the score given when pain was present could explain the moderate and low inter-rater reliability for some of the sub-tests. Although there is a clear Pain Criterion Checklist in the official FMS Manual, the raters in this study agreed to say the concept of “Discomfort” (which should be scored the same as pain) described in the Manual remains unclear. According to the raters, it seems to be hard to dissociate a feeling of actual discomfort from uneasiness to perform a movement that the participant is not used to doing. If the raters did not agree on the presence of pain when sensation of discomfort was reported, this would significantly decrease the kappa value because the presence of pain automatically gave a score of 0, even if the subject successfully completed the task. Thus, it would be interesting to clarify the concept of discomfort in the FMS training in a goal of enhancing reliability of the test.
Both video raters obtained an ICC value of 0.96 for the total FMS score, which is considered to be highly reliable. The active straight leg raise, trunk stability push-up, shoulder mobility subtests and one clearing exam (shoulder pain test) are considered almost perfectly reliable. All the other sub-tests (deep squat, inline lunge, hurdle step, rotary stability) obtained substantial reliability values. Our results were slightly different from Teyhen’s study (Teyhen et al., 2012), which revealed the rotary stability intra-rater scoring as poorly reliable (weighted kappa ¼ 0.29). These differences could be explained by the use of videotaping in our study, which removed the participants’ learning bias because the raters scored the same performance at T1 and T2. Furthermore, the six-week delay between both rating times enhances the power of our results. The presence of pain could also partly explain the differences in our results compared to Tehyen’s results, in which none of the 64 subjects scored a 0 on any sub-test. The kappa values were generally higher in the intra-rater analysis in this study. This finding can be explained by the score attributed to the presence of pain, a subjective concept that is consistent for one individual yet varies between different people (Marchand, 2009). Moreover, the level of inter- and intra-rater reliability is different among the FMS’s sub-tests. This could be attributed to 1) the fact that the criteria for some of the sub-tests are more objective than that of the other sub-tests; 2) the subjectivity of pain for the clearing exams; and 3) the difficulty in viewing all planes of movement simultaneously. Firstly, some of the functional tasks of the FMS were more difficult to score. The in-line lunge and hurdle step do not have induced compensations (e.g., board under heel for deep squat) or objective criteria (e.g., measurement for shoulder mobility) which help the rater to draw the line between a score of 3, 2 or 1. In those sub-tests, the difference between a score of 2 and 1 is more subjective because a score of 2 is described as acceptable compensation, and a score of 1 is too much compensation. There might be differences between what two evaluators consider as an acceptable compensation. Interestingly, the in-line lunge and hurdle step subtests showed the lowest reliability. One way to improve the scoring of the in-lunge sub-test would be to have vertical landmarks on the back wall as to define an area in which the athlete should remain when evaluating the athlete from the front. Using a 100 points scale as proposed by Butler, Plisky, and Kiesel (2012) could be another interesting option to improve scoring of those subtests (Butler et al., 2012).
Table 5 Intra-rater reliability. FMS sub-tests
Deep squat Hurdle step Active straight leg raise Rotary stability Spinal flexion pain (clearing exam) In-line lunge Trunk stability push-up Back extension pain (clearing exam) Shoulder mobility Shoulder pain test (clearing exam) FMS total scoreb a b
Video rater 1
Video rater 2 a
Agreement (%)
Kappa (95% CI)
Agreement (%)
Kappa (95% CI)
78.60 64.30 96.40 75.00 100.00 71.40 82.10 100.00 96.40 96.40
0.78 (0.51e1.00) 0.65 (0.37e0.93) 0.98 (0.93e1.00) 0.65 (0.35e0.96) e 0.70 (0.47e0.92) 0.85 (0.68e1.00) 1.00 0.98 (0.95e1.00) 0.87 (0.62e1.12) ICC (IC 95%) 0.96 (0.92e0.98)
82.1 71.4 85.7 85.7 100.0 78.6 89.3 100.0 96.4 90.0
0.91 (0.81e1.00) 0.76 (0.52e0.99) 0.91 (0.81e1.00) 0.77 (0.51e1.00) e 0.79 (0.61e0.97) 0.93 (0.83e1.00) 1.00 0.86 (0.60e1.00) 0.90 (0.71e1.00) ICC (95% CI) 0.96 (0.92e0.98)
The kappa value corresponds to the quadratic weighted kappa coefficient, except for the clearing exams, for which non-weighted kappa coefficient tests were used. ICC was used only for the FMS total score.
Please cite this article in press as: Parenteau-G, E., et al., Functional movement screen test: A reliable screening test for young elite ice hockey players, Physical Therapy in Sport (2013), http://dx.doi.org/10.1016/j.ptsp.2013.10.001
6
E. Parenteau-G et al. / Physical Therapy in Sport xxx (2013) 1e7
Secondly, the results of the clearing exams (back extension pain test, spinal flexion test, shoulder pain test) were not as good as originally expected. Theoretically, because the variable of interest was the presence of pain (yes or no), the percentage of agreement should have been 100%. However, the agreement percentages for the shoulder pain test were 96.4% (inter-rater reliability), 96.4% (intra-rater, video rater 1) and 90.0% (intra-rater, video rater 2), and the inter-rater reliability of the back extension pain test was 96.4%. As previously mentioned, since pain is subjective for both raters’ observations and participants’ perceptions, it can influence the score, and ultimately the reliability (especially inter-rater reliability). Thirdly, it is difficult to evaluate three planes of movement (sagittal, transverse and frontal planes) during the performance of the same sub-test. The subject was allowed to perform each subtest three times but movement patterns might not be identically performed during each trial and the rater could miss information in one plane of movement while evaluating another, affecting the score given by the rater. This can have a greater impact on FMS subtests that have more criteria to evaluate in different planes (e.g., hurdle step, in-line lunge, rotary stability and deep squat test). 5.3. Study limitations This study has some limitations. Firstly, the rating sessions (field and video) were designed in a way that 14 participants in a row were rated first, following by a second group of 14 participants. This could have caused fatigue among the raters, affecting the scoring of the last few participants. However, this was taken into account by allowing a period of rest once the first group of participants was evaluated. Secondly, all the raters followed the same FMS course and had the same experience using the FMS test which could potentially enhance our reliability results and limit external validity. 6. Conclusion This study demonstrates that the FMS is a reliable test for young elite hockey players. Physiotherapists as well as other health professionals involved with young hockey players could integrate the FMS into their clinical exam in order to suggest interventions that focus on movement pattern deficits observed during the sub-tests. It appears that pain can be an issue for the clearing exam scoring. Pain can also have an effect on whether or not a score of 0 should be attributed to the performance of the other sub-tests. This impact could be lessened by adding standardized questions which require clear yes or no answers. Interesting research avenues remain to be explored. The predictive validity of the FMS test within this population could be assessed. Moreover, further studies could investigate psychometric properties of the FMS in other populations vulnerable to overuse injuries due to a movement pattern deficit like workers performing physically demanding duties. This could promote the FMS test as an evaluation tool for physical therapy practice among various populations. Conflict of interest None declared. Ethical approval Institutional ethical approval (# 10-222) was obtained from the Étienne-Le Bel Clinical Research Centre Review Board and all participants and their parents signed the informed consent.
Funding None declared. Acknowledgments We would like to thank Phil Snider for his involvement in the FMS training and Louis-Simon Allaire (physiotherapist) who kindly agreed to participate in the data collection. References Agel, J., Dompier, T. P., Dick, R., & Marshall, S. W. (2007). Descriptive epidemiology of collegiate men’s ice hockey injuries: national collegiate athletic association injury surveillance system, 1988e1989 through 2003e2004. Journal of Athletic Training, 42(2), 241e248. Brown, S. H., & McGill, S. M. (2010). A comparison of ultrasound and electromyography measures of force and activation to examine the mechanics of abdominal wall contraction. Clinical Biomechanics, 25(2), 115e123. Butler, R. J., Plisky, P. J., & Kiesel, K. B. (2012). Interrater reliability of videotaped performance on the functional movement screen using the 100-point scoring scale. Athletic Training & Sports Health Care, 4(3). Cook, G., Burton, L., & Hoogenboom, B. (2006). Pre-participation screening: the use of fundamental movements as an assessment of function e part 1. North American Journal of Sports Physical Therapy, 1(2), 62e72. Cowley, P. M., & Swensen, T. C. (2008). Development and reliability of two core stability field tests. The Journal of Strength & Conditioning Research, 22(2), 619e624. Emery, C. A., Meeuwisse, W. H., & Powell, J. W. (1999). Groin and abdominal strain injuries in the National Hockey League. Clinical Journal of Sport Medicine, 9(3), 151e156. Frohm, A., Heijne, A., Kowalski, J., Svensson, P., & Myklebust, G. (2012). A nine-test screening battery for athletes: a reliability study. Scandinavian Journal of Medicine and Science in Sports, 22(3), 306e315. Governement of Canada e Canadian Heritage. (2010). Youth participation in sport. Fact sheetact sheet. Retrieved October, 2012, from http://publications.gc.ca/ collections/collection_2013/pc-ch/CH24-1-2012-eng.pdf. Hanada, E., Hubley-Kozey, C., McKeon, M., & Gordon, S. (2008). The feasibility of measuring the activation of the trunk muscles in healthy older adults during trunk stability exercises. BMC Geriatrics, 8(1), 33. Hibbs, A. E., Thompson, K. G., French, D. N., Hodgson, D., & Spears, I. R. (2011). Peak and average rectified EMG measures: which method of data reduction should be used for assessing core training exercises? Journal of Electromyography and Kinesiology, 21(1), 102e111. Hodges, P. W., & Richardson, C. A. (1997). Contraction of the abdominal muscles associated with movement of the lower limb. Physical Therapy, 77(2), 132e142. Hostetler, S. G., Xiang, H., & Smith, G. A. (2004). Characteristics of ice hockey-related injuries treated in US emergency departments, 2001e2002. Pediatrics, 114(6), e661e666. Kiesel, K., Plisky, P., & Butler, R. (2011). Functional movement test scores improve following a standardized off-season intervention program in professional football players. Scandinavian Journal of Medicine & Science in Sports, 21(2), 287e292. Kiesel, K., Plisky, P. J., & Voight, M. L. (2007). Can serious injury in professional football be predicted by a preseason functional movement screen? North American Journal of Sport Physical Therapy, 2(3), 147e158. Ladeira, C. E., Hess, L. W., Galin, B. M., Fradera, S., & Harkness, M. A. (2005). Validation of an abdominal muscle strength test with dynamometry. The Journal of Strength & Conditioning Research, 19(4), 925e930. Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33(1), 159e174. Leetun, D. T., Ireland, M. L., Willson, J. D., Ballantyne, B. T., & Davis, I. M. (2004). Core stability measures as risk factors for lower extremity injury in athletes. Medicine & Science in Sports & Exercise, 36(6), 926e934. Lorentzon, R., Wedren, H., Pietila, T., & Gustavsson, B. (1988). Injuries in international ice hockey. A prospective, comparative study of injury incidence and injury types in international and Swedish elite ice hockey. American Journal of Sports Medicine, 16(4), 389e391. Marchand, S. (2009). Le phénomène de la douleur (2e édition). Montréal: Éditions de la Chenelière. Micheli, L. J., & Klein, J. D. (1991). Sports injuries in children and adolescents. British Journal of Sports Medicine, 25(1), 6e9. Minick, K. I., Kiesel, K. B., Burton, L., Taylor, A., Plisky, P., & Butler, R. J. (2010). Interrater reliability of the functional movement screen. The Journal of Strength & Conditioning Research, 24(2), 479e486. Nadler, S. F., Malanga, G. A., Bartoli, L. A., Feinberg, J. H., Prybicien, M., & Deprince, M. (2002). Hip muscle imbalance and low back pain in athletes: influence of core strengthening. Medicine & Science in Sports & Exercise, 34(1), 9e16. O’Connor, F. G., Deuster, P. A., Davis, J., Pappas, C. G., & Knapik, J. J. (2011). Functional movement screening: predicting injuries in officer candidates. Medicine and Science in Sports and Exercise, 43(12), 2224e2230.
Please cite this article in press as: Parenteau-G, E., et al., Functional movement screen test: A reliable screening test for young elite ice hockey players, Physical Therapy in Sport (2013), http://dx.doi.org/10.1016/j.ptsp.2013.10.001
E. Parenteau-G et al. / Physical Therapy in Sport xxx (2013) 1e7 Onate, J. A., Dewey, T., Kollock, R. O., Thomas, K. S., Van Lunen, B. L., DeMaio, M., et al. (2012). Real-time intersession and interrater reliability of the functional movement screen. The Journal of Strength & Conditioning Research, 26(2), 408e415. Pettersson, M., & Lorentzon, R. (1993). Ice hockey injuries: a 4-year prospective study of a Swedish elite ice hockey team. British Journal of Sports Medicine, 27(4), 251e254.
7
Teyhen, D. S., Shaffer, S. W., Lorenson, C. L., Halfpap, J. P., Donofry, D. F., Walker, M. J., et al. (2012). The functional movement screen: a reliability study. Journal of Orthopaedic & Sports Physical Therapy, 42(6), 530e540. Wilkins, K. E. (1980). The uniqueness of the young athlete: musculoskeletal injuries. American Journal of Sports Medicine, 8(5), 377e382.
Please cite this article in press as: Parenteau-G, E., et al., Functional movement screen test: A reliable screening test for young elite ice hockey players, Physical Therapy in Sport (2013), http://dx.doi.org/10.1016/j.ptsp.2013.10.001
Contents lists available at ScienceDirect
Physical Therapy in Sport journal homepage: www.elsevier.com/ptsp
Original research
Functional movement screen test: A reliable screening test for young elite ice hockey players Elizabeth Parenteau-G a, Nathaly Gaudreault a, *,1, Stéphane Chambers a, Caroline Boisvert a, Alexandre Grenier a, Geneviève Gagné a, Frédéric Balg b a
School of Rehabilitation, Faculty of Medicine and Health Sciences, Université de Sherbrooke, 3001, 12e Avenue Nord, Sherbrooke, QC, Canada J1H 5N4 Departement of Orthopaedic Surgery, Faculty of Medicine and Health Sciences, Université de Sherbrooke, 3001, 12e Avenue Nord, Sherbrooke, QC, Canada J1H5N4 b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 7 March 2013 Received in revised form 9 August 2013 Accepted 3 October 2013
Objectives: To determine inter-rater and intra-rater reliability of the Functional Movement Screen (FMS) test among young elite hockey players. Design: Reliability study. Setting: Inter-rater reliability was evaluated by two raters in the field. All performances were videotaped. Two other raters evaluated the videos once and then again 6 weeks later to determine intra-rater reliability. A weighted kappa statistic was used to analyze intra-rater and inter-rater reliability of each FMS sub-test, while an intra-class correlation coefficient (ICC) was calculated for the total score. Participants: Twenty-eight male hockey players aged 13e16. Main Outcome: FMS total and sub-tests scores. Results: The video raters demonstrated excellent intra-rater reliability for the total score, with an ICC of 0.96 (95% CI; 0.92e0.98) and 0.96 (95% CI; 0.91e0.98). The field raters achieved excellent inter-rater reliability for the total score, with an ICC of 0.96 (95% CI; 0.92e0.98). Sub-test analysis showed good agreement among all four raters for five of the seven main sub-tests. Conclusion: FMS is a reliable test for young elite hockey players. Further research should be done to assess the predictive validity of the FMS test within this population so that physiotherapists may eventually use it as an injury prevention tool. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Functional test Ice hockey Reliability
1. Introduction Ice hockey is one of the most popular sports practiced by teenagers in Canada (Governement of Canada - Canadian Heritage, 2010) and young hockey players aged 12e17 are highly vulnerable to injuries (Hostetler, Xiang, & Smith, 2004). Although the majority of ice hockey injuries are caused by contact, nearly 15% of injuries are non-contact musculoskeletal (MSK) injuries (Agel, Dompier, Dick, & Marshall, 2007) caused by overuse or poor movement patterns (Emery, Meeuwisse, & Powell, 1999; Lorentzon, Wedren, Pietila, & Gustavsson, 1988). Non-contact injuries in hockey players seem to mostly affect their pelvic and abdominal muscles and tendons (Emery et al., 1999; Lorentzon et al., 1988). Most of
* Corresponding author. Tel.: þ1 819 820 6868x12910; fax: þ1 819 820 6864. E-mail address: [email protected] (N. Gaudreault). 1 The corresponding author agrees to allow the publication of her contact information.
these non-contact injuries are related to the high velocity movements executed by the players, such as slap shots and skating. Performance of these movements not only requires significant pelvic and abdominal core strength (static stability) but coordinated recruitment patterns of these muscles when performing dynamic peripheral movements (dynamic stability) (Hodges & Richardson, 1997; Leetun, Ireland, Willson, Ballantyne, & Davis, 2004; Nadler, Malanga, Bartoli, Feinberg, Prybicien, & Deprince, 2002). According to Leetun et al. (2004), pelvic and abdominal muscle injuries in hockey players might be reduced if core strength and movement pattern deficits were detected earlier. Core strength and muscle activity can be evaluated in many ways. Measures by dynamometer testing (Hanada, Hubley-Kozey, McKeon, & Gordon, 2008), ultrasonography (Brown & McGill, 2010) and by electromyographic studies (Hibbs, Thompson, French, Hodgson, & Spears, 2011) are precise but are not clinically suitable because of high costs and the equipment required. In fact, clinicians are more likely to use standardized tests such as the Front Abdominal Power Test (Cowley & Swensen, 2008), the Double Leg
1466-853X/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ptsp.2013.10.001
Please cite this article in press as: Parenteau-G, E., et al., Functional movement screen test: A reliable screening test for young elite ice hockey players, Physical Therapy in Sport (2013), http://dx.doi.org/10.1016/j.ptsp.2013.10.001
2
E. Parenteau-G et al. / Physical Therapy in Sport xxx (2013) 1e7
Lowering Manœuvre (DLLM) (Ladeira, Hess, Galin, Fradera, & Harkness, 2005) or the Functional Movement Screen (FMS). Firstly, the Front Abdominal Power Test assesses abdominal strength in the sagittal plane. This test has shown good reproducibility (ICC 0.95) but might show some limitations in the evaluation of hockey players since it does not address stability in multiple planes. Secondly, the DLLM also has good reproducibility (ICC 0.93) but does not appear to be a valid measure of strength and stability when compared to EMG studies (Ladeira et al., 2005). The FMS was developed to evaluate trunk stability and motion quality of fundamental movement patterns (Cook, Burton, & Hoogenboom, 2006; Hanada et al., 2008) with the ultimate goal of predicting non-contact injuries in athletes. Fundamental movements are basic movements that can simultaneously assess range of motion, stability and balance. This test is currently used in many pre-season screening protocols in different sports (Cook et al., 2006; Kiesel, Plisky, & Butler, 2011; Kiesel, Plisky, & Voight, 2007). The FMS assesses range of motion, symmetry and stability during seven main functional movements, and is rated on a 21 point scale. It has been shown that athletes who obtain a score less than or equal to 14/21 are at higher risk of injury (Kiesel et al., 2007). Recent studies with civilian (Onate et al., 2012), military (Teyhen et al., 2012), professional football players (Minick, Kiesel, Burton, Taylor, Plisky, & Butler, 2010) and elite soccer players (Frohm, Heijne, Kowalski, Svensson, & Myklebust, 2012) report substantial to excellent reliability of the FMS, with intra-class correlation coefficients (ICC) for the total score ranging from 0.76 to 0.98. Although these results appear to be promising in identifying athletes at risk of non-contact injuries due to poor stability and movement pattern deficits, further research is required with respect to FMS clinimetric properties. The reliability of the FMS test has never been documented in a population of young hockey players, despite the popularity of this sport. FMS reliability has also never been assessed for field rating, which in this study refers to a context very similar to a pre-season evaluation set-up for this population. Therefore, the aim of our study is to determine interrater (field rating) and intra-rater (video rating) reliability of the FMS test among young elite ice hockey players. 2. Materials and methods 2.1. Study design Inter-rater and intra-rater studies were conducted to determine the reliability of the FMS performed on young hockey players. 2.2. Participants A sample of 30 male hockey players was recruited from the hockey sports study program of a high school located in Sherbrooke, Quebec. The first step was to contact the program coordinator to explain the research project. Flyers describing the purpose and procedures of the study were then distributed to all hockey players involved in Bantam AA (13e14 years old) and Midget Espoir (15e17 years old) leagues. In Quebec, these categories consist of the best players and are now the cornerstone for identifying players expected to be selected for national teams and to be noticed by scouts for professional teams. A letter was also sent to their parents. Players who wanted to participate were invited to contact the project coordinator who verified their study eligibility. Thirty participants satisfied the inclusion and exclusion criteria defined in Table 1. Institutional ethical approval was obtained from the Etienne-Le Bel Clinical Research Centre Review Board, the 30 eligible participants and their parents willingly signed the consent form and the rights of the participants were protected. However,
two of the participants were sick on the day the data was collected. Therefore, 28 players participated in the study. 2.3. Description of the FMS The FMS test mainly evaluates trunk stability, range of motion, motion quality and symmetry during the performance of basic functional movements. The test includes the seven sub-tests depicted in Fig. 1 and three clearing exams: deep squat, hurdle step (right and left), active straight leg raise (right and left), rotary stability (right and left) and spinal flexion pain test (clearing exam); in-line lunge (right and left), trunk stability push-up and back extension pain test (clearing exam); shoulder mobility (right and left) and shoulder pain test (clearing exam). The sub-tests are evaluated using a 0e3 ordinal scale: 0 ¼ pain during the movement; 1 ¼ participant is unable to execute the correct movement; 2 ¼ participant executes the movement with compensations; and 3 ¼ participant executes correct movement without pain or compensations. The clearing exams were scored using a dichotomous variable: presence of pain (yes or no). The participants were given three attempts at each sub-test, and the evaluator had to rate only the best performance. A positive clearing exam (pain during the movement) automatically gave a score of 0 for the related sub-test. A summed total score was then calculated. 2.4. Data collection The data were collected at three different sessions as illustrated in Fig. 2: the first session allowed field inter-rater reliability data to be collected; the second and third sessions served to collect data for the intra-rater reliability at time 1 (T1) and time 2 (T2). The examiners (n ¼ 4) consisted of three senior physiotherapy students enrolled in their last year of a professional Master’s degree physiotherapy training program and one physiotherapist. Each evaluator had obtained official FMS training certification and had practiced the FMS prior to data collection. 2.5. First session: inter-rater reliability To collect data for the field inter-rater reliability study, all participants were invited to come to the Université de Sherbrooke’s rehabilitation clinic. Prior to the FMS evaluation, each participant completed a questionnaire to collect data on their sports experience and history of musculoskeletal (MSK) non-contact injuries. These data served to describe the participants’ characteristics. Each participant was then asked to perform the seven FMS sub-tests following standardized instructions (described in the FMS user manual). A random number generating software was used to randomly assigned the order of the sub-tests for each participant. One randomly selected pair of examiners (i.e., field raters) independently rated the participants during their performance of each sub-test using a scoring sheet. Examiners were blinded from each other’s scores. The scores of those two examiners were then used for inter-rater analysis. A front view camera and two side view
Table 1 Eligibility criteria. Inclusion
Exclusion
Male Member of an elite hockey sport-study program (Bantam AA or Midget Espoir leagues) 13e17 years old
Symptomatic (acute) musculoskeletal injury that may interfere with their ability to participate in regular hockey training/playing at the time of the FMS evaluation
Please cite this article in press as: Parenteau-G, E., et al., Functional movement screen test: A reliable screening test for young elite ice hockey players, Physical Therapy in Sport (2013), http://dx.doi.org/10.1016/j.ptsp.2013.10.001
E. Parenteau-G et al. / Physical Therapy in Sport xxx (2013) 1e7
3
Fig. 1. Illustration of the FMS sub-tests: 1) deep squat; 2) hurdle step; 3) active straight leg raise; 4) rotary stability; 5) in-line lunge; 6) trunk shoulder mobility and 7) stability push-up.
cameras (Canon ixia hfs200, Mississauga, Ontario, Canada) were installed during the test to capture the participants’ performance. 2.6. Second session: intra-rater reliability T1 The videos recorded during the first session were displayed so that the front view and both side views could be viewed simultaneously on a computer screen. This allowed the other two examiners (video rater 1 and 2) to independently rate each participant’s performance using the FMS sub-tests. The videos were projected in a closed room with a research assistant who made sure that there was no communication between the raters so that they were blinded from each other’s scores. This session was held over the course of one day.
Fig. 2. The data were collected at three different sessions: the first session allowed field inter-rater reliability data to be collected; the second and third sessions served to collect data for the intra-rater reliability at time 1 (T1) and time 2 (T2). A six-week time period separated T1 and T2 sessions.
2.7. Third session: intra-rater reliability T2 Six weeks following the first video rating session, the same pair of examiners rated the same 28 videos showing each participant performing the FMS sub-tests under the same condition as T1. The time elapsed was sufficient to control for recall bias. The scores obtained during T1 and T2 were used for intra-rater analysis. The data obtained at T2 were not used to document inter-rater reliability because inter-rater reliability has already been documented using video rating (Kiesel et al., 2011; Minick et al., 2010). The use of two video raters was to increase the variability in the study design for intra-rater reliability.
3. Statistical analysis Data were analyzed using SPSS (Version 18) and StatXact (Version 6) software. Descriptive statistics were used to describe the participant sample: mean and standard deviations were calculated for age, hockey experience and FMS sub-tests and total scores; percentage was used to depict the proportion of players who reported having a past history of MSK injury and those who were identified as being at risk of a non-contact MSK injury by having an FMS total score lower than 14/21 (O’Connor, Deuster, Davis, Pappas, & Knapik, 2011). To estimate inter- and intra-rater reliability (see Fig. 2), the percentage of agreement was calculated and quadratic weighted Cohen kappa coefficients were used for each FMS sub-test score. Using a quadratic weighted Cohen kappa coefficient is stronger than calculating the percentage of agreement between raters because it takes random agreement into account. The reliability results for each sub-test were classified as follows based on the work of Landis and Koch (1977): almost perfect reliability, 0.81e1.0; substantial reliability, 0.61e0.80; moderate reliability, 0.41e60; fair reliability, 0.21e0.40; and slight reliability,
Please cite this article in press as: Parenteau-G, E., et al., Functional movement screen test: A reliable screening test for young elite ice hockey players, Physical Therapy in Sport (2013), http://dx.doi.org/10.1016/j.ptsp.2013.10.001
4
E. Parenteau-G et al. / Physical Therapy in Sport xxx (2013) 1e7 Table 2 Participant characteristics.
Table 4 Inter-rater reliability.
Characteristics
FMS sub-tests Mean (SD) 14.89 0.99 63.8 10.3 171.0 5.4 10.28 1.21
Age (years) Weight (kg) Height (cm) Hockey experience (years) n
%
Players who reported having a history of a non-contact MSK injury - In the last year 14 50% - More than a year ago 14 50% Players who had an FMS score lower than 14 17 60.7%
0.00e0.20. However, weighted Cohen Kappa coefficients could not be used for the clearing exams because of the dichotomous nature of the score (presence of pain; yes or no) and the fact that a nonweighted Kappa test was used for these sub-tests. Then, intrarater and inter-rater reliability for the FMS total score was determined by using intra-class correlation coefficient (ICC) analysis. Interpretation of ICC analysis was based on the criteria proposed by Onate and Dewey (Onate et al., 2012): high reliability, 0.90e0.99; good reliability, 0.80e0.89; fair reliability, 0.70e0.79; and poor reliability, 0.00e0.69.
4. Results The characteristics of the participants who performed the evaluation are described in Table 2 and descriptive analysis for the FMS scores are presented in Table 3. As observed, the mean FMS total score was 12.64 (with scores ranging from 5 to 19). Moreover, close to two-thirds of the participants (60.7%) were identified as being at risk of a non-contact MSK injury according to Kiesel’s FMS risk injury threshold score of 14/21 on the basis that they had a total FMS score lower than 14 (Kiesel et al., 2007; O’Connor et al., 2011).
4.1. FMS field rating inter-rater reliability Table 4 shows inter-rater percentage of agreement and weighted kappa coefficient values for each sub-test and the ICC value for the FMS total score. The percentage of agreement for each FMS sub-test ranged from 67.9 to 100%, whereas the quadratic weighted kappa coefficient value ranged from 0.26 to 0.99. The quadratic weighted Cohen kappa coefficient was not calculated for the spinal flexion test (clearing exam) because the percentage agreement was 100%. Fig. 3 shows the correlation for the total FMS score obtained by the two field raters. Table 3 Descriptive analysis for FMS sub-tests and total scores. An analysis of field rater 1’s score sheet revealed the following data. FMS sub-tests
Deep squat Hurdle step Active straight leg raise Rotary stability In-line lunge Trunk stability push-up Shoulder mobility FMS total score
Number of players per score category 0
1
2
3
6 1 5 1 4 15 6
1 0 9 1 3 0 4
13 12 11 24 9 5 10
8 15 3 2 12 8 8
Mean SD FMS sub-test scores
1.82 1.09 2.46 0.69 1.43 0.92 1.96 0.51 2.04 1.07 1.21 1.37 1.71 1.11 12.643.65
Agreement (%) Kappaa (95% CI)
Deep squat 82.10 Hurdle step 67.90 Active straight leg raise 82.10 Rotary stability 89.30 Spinal flexion pain test (clearing exam) 100.00 In-line lunge 67.90 Trunk stability push-up 78.60 Back extension pain test (clearing exam) 96.40 Shoulder mobility 96.40 Shoulder pain test (clearing exam) 96.40 FMS total scoreb e
0.75 (0.46e1.00) 0.55 (0.30e0.80) 0.97 (0.90e1.00) 0.26 (0.18 to 0.67) e 0.77 (0.57e0.96) 0.90 (0.80e0.99) 0.92 (0.77e1.00) 0.99 (0.96e1.00) 0.87 (0.62e1.00) ICC (IC 95%) 0.96 (0.92e0.98)
a The kappa value corresponds to the quadratic weighted kappa coefficient, except for the clearing exams, for which non-weighted kappa coefficient tests were used. b ICC was used only for the FMS total score.
4.2. FMS video rating intra-rater reliability Intra-rater percentage of agreement and weighted kappa coefficient values for each sub-test and the ICC value for the FMS total score are presented in Table 5. For video rater 1, the agreement percentage ranged from 64.30 to 100.00% for each FMS sub-test, whereas the kappa value ranged from 0.65 to 1.00. For video rater 2, the agreement percentage ranged from 71.40 to 100.00% for each FMS sub-test, while the kappa value ranged from 0.76 to 1.00. The quadratic weighted Cohen kappa coefficient was not calculated for two clearing exams (spinal flexion test and back extension pain test) because the percentage of agreement was 100%. 5. Discussion As previously mentioned, the FMS was developed to evaluate trunk stability and motion quality while performing functional movements, with the ultimate goal of predicting non-contact injuries in athletes. Our results showed that half of the hockey players in our sample had already suffered a non-contact MSK injury, either in the past year or since they started playing hockey. This result can hardly be compared with existing evidence due to the paucity of scientific literature on the prevalence of non-contact injuries among young hockey players. This lack of information is quite surprising given that children and adolescents are particularly at risk of suffering injuries due to the immaturity of their musculoskeletal systems (Micheli & Klein, 1991; Wilkins, 1980). In adult hockey players, while most injuries occur during contact, the fact remains that almost 15% of injuries are not related to contact (Agel et al., 2007; Pettersson & Lorentzon, 1993). These results highlight the need to document the prevalence of non-contact injuries in young hockey players with more robust epidemiological studies. What is more troubling is that 60.7% of the young hockey players in this study had a total score lower than 14/21 (Kiesel et al., 2007; O’Connor et al., 2011) which means that 2 out of 3 players are more likely to suffer a noncontact MSK injury or reoccurrence of a previous injury. Positive effects of screening athletes who might be at risk of a non-contact injury with the FMS test and providing them with a specific exercise program has already been demonstrated (Kiesel et al., 2011). This leads us to believe that prevention of non-contact injuries with early screening and training programs adjustments is a promising avenue for young hockey players to consider. 5.1. Inter-rater reliability According to Onate and Dewey (Onate et al., 2012), we can say that with an ICC of 0.96, the total FMS score is highly reliable. When
Please cite this article in press as: Parenteau-G, E., et al., Functional movement screen test: A reliable screening test for young elite ice hockey players, Physical Therapy in Sport (2013), http://dx.doi.org/10.1016/j.ptsp.2013.10.001
E. Parenteau-G et al. / Physical Therapy in Sport xxx (2013) 1e7
5
5.2. Intra-rater reliability
Fig. 3. The high correlation for the total FMS score obtained by field rater 1 (on the abscissa) and field rater 2 (on the ordinate) is depicted here.
analyzing the reliability of the FMS sub-tests, the active straight leg raise test, the trunk stability push-up test, the shoulder mobility test and two of the clearing exams (back extension pain test, shoulder pain test) are considered as almost perfectly reliable. The deep squat and in-line lunge tests yielded substantial reliability, the Hurdle step test showed moderate reliability and the rotary stability test demonstrated fair reliability. These results are consistent with Minick’s inter-rater reliability study (Minick et al., 2010), which also revealed the rotary stability test to be a less reliable subtest when scored by novice raters (weighted Kappa ¼ 0.54). In the present study, we believe that the score given when pain was present could explain the moderate and low inter-rater reliability for some of the sub-tests. Although there is a clear Pain Criterion Checklist in the official FMS Manual, the raters in this study agreed to say the concept of “Discomfort” (which should be scored the same as pain) described in the Manual remains unclear. According to the raters, it seems to be hard to dissociate a feeling of actual discomfort from uneasiness to perform a movement that the participant is not used to doing. If the raters did not agree on the presence of pain when sensation of discomfort was reported, this would significantly decrease the kappa value because the presence of pain automatically gave a score of 0, even if the subject successfully completed the task. Thus, it would be interesting to clarify the concept of discomfort in the FMS training in a goal of enhancing reliability of the test.
Both video raters obtained an ICC value of 0.96 for the total FMS score, which is considered to be highly reliable. The active straight leg raise, trunk stability push-up, shoulder mobility subtests and one clearing exam (shoulder pain test) are considered almost perfectly reliable. All the other sub-tests (deep squat, inline lunge, hurdle step, rotary stability) obtained substantial reliability values. Our results were slightly different from Teyhen’s study (Teyhen et al., 2012), which revealed the rotary stability intra-rater scoring as poorly reliable (weighted kappa ¼ 0.29). These differences could be explained by the use of videotaping in our study, which removed the participants’ learning bias because the raters scored the same performance at T1 and T2. Furthermore, the six-week delay between both rating times enhances the power of our results. The presence of pain could also partly explain the differences in our results compared to Tehyen’s results, in which none of the 64 subjects scored a 0 on any sub-test. The kappa values were generally higher in the intra-rater analysis in this study. This finding can be explained by the score attributed to the presence of pain, a subjective concept that is consistent for one individual yet varies between different people (Marchand, 2009). Moreover, the level of inter- and intra-rater reliability is different among the FMS’s sub-tests. This could be attributed to 1) the fact that the criteria for some of the sub-tests are more objective than that of the other sub-tests; 2) the subjectivity of pain for the clearing exams; and 3) the difficulty in viewing all planes of movement simultaneously. Firstly, some of the functional tasks of the FMS were more difficult to score. The in-line lunge and hurdle step do not have induced compensations (e.g., board under heel for deep squat) or objective criteria (e.g., measurement for shoulder mobility) which help the rater to draw the line between a score of 3, 2 or 1. In those sub-tests, the difference between a score of 2 and 1 is more subjective because a score of 2 is described as acceptable compensation, and a score of 1 is too much compensation. There might be differences between what two evaluators consider as an acceptable compensation. Interestingly, the in-line lunge and hurdle step subtests showed the lowest reliability. One way to improve the scoring of the in-lunge sub-test would be to have vertical landmarks on the back wall as to define an area in which the athlete should remain when evaluating the athlete from the front. Using a 100 points scale as proposed by Butler, Plisky, and Kiesel (2012) could be another interesting option to improve scoring of those subtests (Butler et al., 2012).
Table 5 Intra-rater reliability. FMS sub-tests
Deep squat Hurdle step Active straight leg raise Rotary stability Spinal flexion pain (clearing exam) In-line lunge Trunk stability push-up Back extension pain (clearing exam) Shoulder mobility Shoulder pain test (clearing exam) FMS total scoreb a b
Video rater 1
Video rater 2 a
Agreement (%)
Kappa (95% CI)
Agreement (%)
Kappa (95% CI)
78.60 64.30 96.40 75.00 100.00 71.40 82.10 100.00 96.40 96.40
0.78 (0.51e1.00) 0.65 (0.37e0.93) 0.98 (0.93e1.00) 0.65 (0.35e0.96) e 0.70 (0.47e0.92) 0.85 (0.68e1.00) 1.00 0.98 (0.95e1.00) 0.87 (0.62e1.12) ICC (IC 95%) 0.96 (0.92e0.98)
82.1 71.4 85.7 85.7 100.0 78.6 89.3 100.0 96.4 90.0
0.91 (0.81e1.00) 0.76 (0.52e0.99) 0.91 (0.81e1.00) 0.77 (0.51e1.00) e 0.79 (0.61e0.97) 0.93 (0.83e1.00) 1.00 0.86 (0.60e1.00) 0.90 (0.71e1.00) ICC (95% CI) 0.96 (0.92e0.98)
The kappa value corresponds to the quadratic weighted kappa coefficient, except for the clearing exams, for which non-weighted kappa coefficient tests were used. ICC was used only for the FMS total score.
Please cite this article in press as: Parenteau-G, E., et al., Functional movement screen test: A reliable screening test for young elite ice hockey players, Physical Therapy in Sport (2013), http://dx.doi.org/10.1016/j.ptsp.2013.10.001
6
E. Parenteau-G et al. / Physical Therapy in Sport xxx (2013) 1e7
Secondly, the results of the clearing exams (back extension pain test, spinal flexion test, shoulder pain test) were not as good as originally expected. Theoretically, because the variable of interest was the presence of pain (yes or no), the percentage of agreement should have been 100%. However, the agreement percentages for the shoulder pain test were 96.4% (inter-rater reliability), 96.4% (intra-rater, video rater 1) and 90.0% (intra-rater, video rater 2), and the inter-rater reliability of the back extension pain test was 96.4%. As previously mentioned, since pain is subjective for both raters’ observations and participants’ perceptions, it can influence the score, and ultimately the reliability (especially inter-rater reliability). Thirdly, it is difficult to evaluate three planes of movement (sagittal, transverse and frontal planes) during the performance of the same sub-test. The subject was allowed to perform each subtest three times but movement patterns might not be identically performed during each trial and the rater could miss information in one plane of movement while evaluating another, affecting the score given by the rater. This can have a greater impact on FMS subtests that have more criteria to evaluate in different planes (e.g., hurdle step, in-line lunge, rotary stability and deep squat test). 5.3. Study limitations This study has some limitations. Firstly, the rating sessions (field and video) were designed in a way that 14 participants in a row were rated first, following by a second group of 14 participants. This could have caused fatigue among the raters, affecting the scoring of the last few participants. However, this was taken into account by allowing a period of rest once the first group of participants was evaluated. Secondly, all the raters followed the same FMS course and had the same experience using the FMS test which could potentially enhance our reliability results and limit external validity. 6. Conclusion This study demonstrates that the FMS is a reliable test for young elite hockey players. Physiotherapists as well as other health professionals involved with young hockey players could integrate the FMS into their clinical exam in order to suggest interventions that focus on movement pattern deficits observed during the sub-tests. It appears that pain can be an issue for the clearing exam scoring. Pain can also have an effect on whether or not a score of 0 should be attributed to the performance of the other sub-tests. This impact could be lessened by adding standardized questions which require clear yes or no answers. Interesting research avenues remain to be explored. The predictive validity of the FMS test within this population could be assessed. Moreover, further studies could investigate psychometric properties of the FMS in other populations vulnerable to overuse injuries due to a movement pattern deficit like workers performing physically demanding duties. This could promote the FMS test as an evaluation tool for physical therapy practice among various populations. Conflict of interest None declared. Ethical approval Institutional ethical approval (# 10-222) was obtained from the Étienne-Le Bel Clinical Research Centre Review Board and all participants and their parents signed the informed consent.
Funding None declared. Acknowledgments We would like to thank Phil Snider for his involvement in the FMS training and Louis-Simon Allaire (physiotherapist) who kindly agreed to participate in the data collection. References Agel, J., Dompier, T. P., Dick, R., & Marshall, S. W. (2007). Descriptive epidemiology of collegiate men’s ice hockey injuries: national collegiate athletic association injury surveillance system, 1988e1989 through 2003e2004. Journal of Athletic Training, 42(2), 241e248. Brown, S. H., & McGill, S. M. (2010). A comparison of ultrasound and electromyography measures of force and activation to examine the mechanics of abdominal wall contraction. Clinical Biomechanics, 25(2), 115e123. Butler, R. J., Plisky, P. J., & Kiesel, K. B. (2012). Interrater reliability of videotaped performance on the functional movement screen using the 100-point scoring scale. Athletic Training & Sports Health Care, 4(3). Cook, G., Burton, L., & Hoogenboom, B. (2006). Pre-participation screening: the use of fundamental movements as an assessment of function e part 1. North American Journal of Sports Physical Therapy, 1(2), 62e72. Cowley, P. M., & Swensen, T. C. (2008). Development and reliability of two core stability field tests. The Journal of Strength & Conditioning Research, 22(2), 619e624. Emery, C. A., Meeuwisse, W. H., & Powell, J. W. (1999). Groin and abdominal strain injuries in the National Hockey League. Clinical Journal of Sport Medicine, 9(3), 151e156. Frohm, A., Heijne, A., Kowalski, J., Svensson, P., & Myklebust, G. (2012). A nine-test screening battery for athletes: a reliability study. Scandinavian Journal of Medicine and Science in Sports, 22(3), 306e315. Governement of Canada e Canadian Heritage. (2010). Youth participation in sport. Fact sheetact sheet. Retrieved October, 2012, from http://publications.gc.ca/ collections/collection_2013/pc-ch/CH24-1-2012-eng.pdf. Hanada, E., Hubley-Kozey, C., McKeon, M., & Gordon, S. (2008). The feasibility of measuring the activation of the trunk muscles in healthy older adults during trunk stability exercises. BMC Geriatrics, 8(1), 33. Hibbs, A. E., Thompson, K. G., French, D. N., Hodgson, D., & Spears, I. R. (2011). Peak and average rectified EMG measures: which method of data reduction should be used for assessing core training exercises? Journal of Electromyography and Kinesiology, 21(1), 102e111. Hodges, P. W., & Richardson, C. A. (1997). Contraction of the abdominal muscles associated with movement of the lower limb. Physical Therapy, 77(2), 132e142. Hostetler, S. G., Xiang, H., & Smith, G. A. (2004). Characteristics of ice hockey-related injuries treated in US emergency departments, 2001e2002. Pediatrics, 114(6), e661e666. Kiesel, K., Plisky, P., & Butler, R. (2011). Functional movement test scores improve following a standardized off-season intervention program in professional football players. Scandinavian Journal of Medicine & Science in Sports, 21(2), 287e292. Kiesel, K., Plisky, P. J., & Voight, M. L. (2007). Can serious injury in professional football be predicted by a preseason functional movement screen? North American Journal of Sport Physical Therapy, 2(3), 147e158. Ladeira, C. E., Hess, L. W., Galin, B. M., Fradera, S., & Harkness, M. A. (2005). Validation of an abdominal muscle strength test with dynamometry. The Journal of Strength & Conditioning Research, 19(4), 925e930. Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33(1), 159e174. Leetun, D. T., Ireland, M. L., Willson, J. D., Ballantyne, B. T., & Davis, I. M. (2004). Core stability measures as risk factors for lower extremity injury in athletes. Medicine & Science in Sports & Exercise, 36(6), 926e934. Lorentzon, R., Wedren, H., Pietila, T., & Gustavsson, B. (1988). Injuries in international ice hockey. A prospective, comparative study of injury incidence and injury types in international and Swedish elite ice hockey. American Journal of Sports Medicine, 16(4), 389e391. Marchand, S. (2009). Le phénomène de la douleur (2e édition). Montréal: Éditions de la Chenelière. Micheli, L. J., & Klein, J. D. (1991). Sports injuries in children and adolescents. British Journal of Sports Medicine, 25(1), 6e9. Minick, K. I., Kiesel, K. B., Burton, L., Taylor, A., Plisky, P., & Butler, R. J. (2010). Interrater reliability of the functional movement screen. The Journal of Strength & Conditioning Research, 24(2), 479e486. Nadler, S. F., Malanga, G. A., Bartoli, L. A., Feinberg, J. H., Prybicien, M., & Deprince, M. (2002). Hip muscle imbalance and low back pain in athletes: influence of core strengthening. Medicine & Science in Sports & Exercise, 34(1), 9e16. O’Connor, F. G., Deuster, P. A., Davis, J., Pappas, C. G., & Knapik, J. J. (2011). Functional movement screening: predicting injuries in officer candidates. Medicine and Science in Sports and Exercise, 43(12), 2224e2230.
Please cite this article in press as: Parenteau-G, E., et al., Functional movement screen test: A reliable screening test for young elite ice hockey players, Physical Therapy in Sport (2013), http://dx.doi.org/10.1016/j.ptsp.2013.10.001
E. Parenteau-G et al. / Physical Therapy in Sport xxx (2013) 1e7 Onate, J. A., Dewey, T., Kollock, R. O., Thomas, K. S., Van Lunen, B. L., DeMaio, M., et al. (2012). Real-time intersession and interrater reliability of the functional movement screen. The Journal of Strength & Conditioning Research, 26(2), 408e415. Pettersson, M., & Lorentzon, R. (1993). Ice hockey injuries: a 4-year prospective study of a Swedish elite ice hockey team. British Journal of Sports Medicine, 27(4), 251e254.
7
Teyhen, D. S., Shaffer, S. W., Lorenson, C. L., Halfpap, J. P., Donofry, D. F., Walker, M. J., et al. (2012). The functional movement screen: a reliability study. Journal of Orthopaedic & Sports Physical Therapy, 42(6), 530e540. Wilkins, K. E. (1980). The uniqueness of the young athlete: musculoskeletal injuries. American Journal of Sports Medicine, 8(5), 377e382.
Please cite this article in press as: Parenteau-G, E., et al., Functional movement screen test: A reliable screening test for young elite ice hockey players, Physical Therapy in Sport (2013), http://dx.doi.org/10.1016/j.ptsp.2013.10.001
