research.tees.ac.uk€¦ · Web viewWe aimed to evaluate the internal consistency and factor...
Transcript of research.tees.ac.uk€¦ · Web viewWe aimed to evaluate the internal consistency and factor...
Functional Movement ScreenTM total score does not present a gestalt measure of
movement quality in youth athletes.
Matthew David Wright1, Paul Chesterton2
1. Sport and Wellbeing, Department of Student and Library Services, Teesside
University, Middlesbrough, United Kingdom
2. Department of Sport, Exercise and Psychology, School of Social Sciences,
Humanities and Law, Teesside University, Middlesbrough, United Kingdom, e-
mail; [email protected]
Words: 3730
Tables: 3
Figures: 3
Matthew Wright
Sport and Wellbeing,
Teesside University,
Middlesbrough,
United Kingdom
TS1 3BX
+44 (0) 1642 342267
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Acknowledgements: We would like to thank all the young athletes who participated in
this study and to Tees Valley Sport for enabling a platform for talented young athletes
to access athletic development support in the Tees Valley.
23
24
25
Functional Movement ScreenTM total score does not present a gestalt measure of
movement quality in youth athletes.
We aimed to evaluate the internal consistency and factor structure of the
Functional Movement Screen (FMSTM) in youth athletes and quantify differences
between individual task score at different stages of maturation. FMSTM and
anthropometric variables were measured in 144 youth athletes (96 female, 48
male). Biological maturation was categorised as before- (<-0.5 years), at- (-0.49-
0.50 years) and after- peak height velocity [PHV] (>0.51 years). Internal
consistency was poor (Cronbach’s alpha; 0.53, ±90% confidence limit 0.10;
ordinal alpha 0.6, ±0.09). Principle component analysis extracted two
components, representing 47% of the total variance. Tasks loading highest on
component 1 required stability, while those loading highest on component 2
favoured mobility. ‘Likely’ decrements in component 1 tasks were observed
before-PHV. In-line lunge (effect size ±90% confidence limit; -0.47, ±0.49),
hurdle step (-0.38, ±0.49), and trunk stability push-up (-0.51, ±0.45), were lower
compared with athletes at-PHV and rotatory stability (-0.44, ±0.37) was lower
than those after-PHV. Boys’ scored ‘most likely’, higher (0.73, ±0.28) in trunk
stability push-up, and girls ‘likely’ higher in shoulder mobility (0.46, ±0.29). In
our population, the FMSTM is not uni-dimensional, thus total score should be
avoided. Clear maturation affects were observed in stability tasks.
Keywords: PCA; FMSTM; long-term athlete development
26
27
282930313233343536373839404142434445
46
Introduction
The Functional Movement Screen (FMSTM) consists of seven tasks, rated on a
scale of 0 to 3 and totalled to quantify movement quality (FMSTM total). The
FMSTM is the most common screening tool in top flight professional association
football clubs (McCall et al., 2014) and used widely with youth athletes (Lloyd et
al., 2015; Newton et al., 2017; Portas, Parkin, Roberts, & Batterham, 2016). This
is likely due to the importance of movement skill in youth athlete development,
which continues throughout the career span of the athlete (Ford et al., 2011; Lloyd
& Oliver, 2012). Furthermore, ‘foundational movements’ included within the
FMSTM, such as squats and lunges, directly impact upon the individuals’ capability
to be physically active (Hulteen, Morgan, Barnett, Stodden, & Lubans, 2018).
Whilst injury prediction was never the intended purpose of the FMSTM, some
studies have reported that a total score of 14 arbitrary units (AU) or under is a
valid predictor of injury (Bonazza, Smuin, Onks, Silvis, & Dhawan, 2017).
FMSTM total score does not predict injuries in adolescent association football
players (Newton et al., 2017), nor does it meet the criteria defined by (Bahr,
2016), to be a valid screen for injury risk.
Evaluating the FMSTM total score assumes it is a gestalt measure of movement
quality and therefore is uni-dimensional. Using this approach fails to consider that
the seven constituent tasks may measure different constructs (Kazman, Galecki,
Lisman, Deuster, & OʼConnor, 2014). Exploratory factor analysis has shown poor
internal consistency for the seven tasks (Kazman et al., 2014; Koehle, Saffer,
Sinnen, & MacInnis, 2016; Li, Wang, Chen, & Dai, 2015). Li et al., (2015) used
principal component analysis to explore the factor structure of the FMSTM in elite
1
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
athletes, identifying two main constructs. They concluded that each task has its
own unique value. The composite FMSTM score does not provide a general
representation of the function of the whole body, given different individuals can
achieve the same composite score with varying task profiles (Li et al, 2015).
Kelleher, Beach, Frost, Johnson, & Dickey (2018) also reported poor factor
congruity between different cohorts; firefighters, general populations and varsity
athletes, suggesting the factor structure may be population specific. The FMSTM
has not been subjective to such scrutiny in youth athlete populations. Given the
importance of movement skill development in these athletes (Lloyd and Oliver,
2012) and the wide use of the FMS in youth athletes (Newton et al., 2017; Portas
et al., 2016), understanding the factor structure in this cohort is relevant.
The FMSTM total score has been shown to be generally lower in youth populations
than in adults (Laurent, Masteller, & Sirard, 2018; Abraham, Sannasi, & Nair,
2015; Wright, Portas, Evans, & Weston, 2015). Younger athletes (~14 or under)
score, on average, around 12 AU, but this increases throughout maturation in boys
(Lloyd et al., 2015; Wright et al., 2015; Portas et al., 2016). Portas et al. (2016),
conducted a large observational study on 1163 adolescent association football
players, reporting a substantial effect for maturation on FMS™ performance.
Only players past their peak-height velocity (PHV) achieved ‘satisfactory’ scores,
with the stability tests most influenced by maturation. This may be due to the
maturation of the nervous system improving co-ordination and driving motor
development, with accelerated development in late childhood (Viru, Loko, Harro,
& Volver, 1999). Interestingly, Abraham et al., (2015) showed no correlation
between FMS™ total score and chronological age in 1005 physically active
children between 10 and 17 years of age. There is limited information pertaining
2
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
to how individual task score may change at different stages of maturation. As
such, further research is warranted, particularly as guidelines for athletic
development recommend a reduced emphasis upon fundamental movement skills
after their peak-height velocity (Lloyd and Oliver, 2012).
In light of a lack of research evaluating the factor structure of the FMSTM in youth
athletes and the impact of maturation on each individual task, the aims of this
research were twofold. Firstly, to identify the internal consistency and factor
structure within the seven tasks in youth athletes. Secondly, to quantify any
differences in individual task score between participants who were before-, at-, or
after- their adolescent growth spurt (peak-height velocity).
Methods
Participants
144 youth athletes (96 female, 48 male) between 8 and 18 years of age (mean 14.1
± standard deviation 2.3 years) who were accepted to participate in a youth athlete
support programme were recruited. The programme included girls’ association
football players from a centre of excellence and other athletes from a wide range
of sports that included track and field athletics, endurance sports (e.g. cycling and
swimming), team sports (e.g. basketball, volleyball, netball), racket sports (e.g.
tennis and table tennis), combat sports (e.g. taekwondo, judo) and water sports
(e.g. canoe / kayak, sailing). All participants were verified by their sport’s
National Governing Body (NGB) as competing at a minimum of “regional level”
competition. As such, they were involved in sports specific training and general
physical education, but were not exposed to structured strength and conditioning
3
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
training at the time of testing. The FMSTM was carried out as part of their initial
athlete induction and all participants were free from self-reported musculoskeletal
injury and medically fit to participate.
***** Table 1 near here *****
Ethics approval was granted from the University ethics committee
(SSSBLREC008) and written informed parental consent was obtained prior to
testing. Biological maturation was estimated from chronological age, body mass,
sitting and standing stature, using a non-invasive method that has been validated
for both boys and girls with a standard error of 0.57 and 0.59 years respectively
(Mirwald, Baxter-Jones, Bailey, & Beunen, 2002). Players were categorised as
before-PHV (<-0.5 years), at-PHV (-0.49-0.50 years) and after-PHV (>0.51
years). Participant characteristics for each maturation category are presented in
Table 1.
Procedures
Participants attended the University’s strength and conditioning laboratory and
completed the seven tasks of the FMSTM (deep squat, hurdle step, in-line lunge,
shoulder mobility, active straight-leg raise, trunk stability push-up and rotatory
stability) and the associated clearing tests without any specific warm-up (Cook,
Burton, & Hoogenboom, 2006a; 2006b). Verbal instructions following the FMSTM
guidelines (Cook, et al., 2006a; 2006b) were delivered to each participant
verbatim by a lead rater. No coaching was provided; however, participants were
provided feedback if they were not adhering specifically to the task instructions.
4
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
All tasks were scored in real time and agreed by two raters. 1) A lead rater, who
was a qualified sports therapist or strength and conditioning coach, with a
minimum of one years’ experience conducting movement assessments including
the FMSTM. 2) An undergraduate sport and exercise student. Raters were not
FMSTM certified, but had been familiarised with the FMSTM scoring criteria, as
detailed by Cook, et al. (2006a; 2006b), through an internal training session,
suggested by Smith, Chimera, Wright & Warren (2013). Level of experience and
certification have been shown not to effect the inter-rater reliability of the FMSTM
(Bonazza et al., 2017; Smith et al., 2013). Each task was scored on the 0-3 scale
proposed by Cook, et al., (2006b). A score of 0 indicated that the participant
experience pain anywhere in the body during the main tasks, or clearing tests. A
score of 1 indicated an inability to complete the test, or assume the position to
perform the movement. A score of 2 indicated the task could be completed with
some compensations. A score of 3 indicated that the task could be completed
correctly without compensations.
Our inter-rater reliability for each of the FMSTM tasks in adolescent populations,
has been shown to be ‘substantial’ to ‘almost perfect’ for the deep squat, hurdle
step left, shoulder mobility and active straight-leg raise, and ‘moderate’ for the in-
line lunge and hurdle step right (Wright et al., 2015). However, agreement of the
trunk stability press-up and rotatory stability ranged from only ‘slight’ to
‘moderate’ (Wright et al., 2015). In order to minimise any inter-rater error in this
study, each task was recorded in the frontal and sagittal plane using standard
video camera’s (HC-V770E; Panasonic, Tokyo, Japan) and the lead author
reviewed each video. Any clear disagreements with the live score were discussed
with a third rater before a final score was agreed via consensus.
5
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
Statistical Analysis
Each FMSTM task-score was converted to a z-score and analysed using SPSS
Statistic version 24 (SPSS, Armonk NY: IBM Corp.). A Pearson’s correlation
matrix was produced for all participants, and for boys and girls separately (Table
2). Sex differences were evaluated in a custom-made spreadsheet, with the
smallest-important difference defined as 0.1 (Hopkins, 2006). The internal
consistency of the seven tasks was evaluated by Cronbach’s alpha. As the FMSTM
task score is categorical data, the ordinal alpha was also calculated through the R
package “pscyh”. Before subjecting the data to principle component analysis
(PCA), its suitability was determined using the Kaiser-Meyer-Olikn measure of
sampling adequacy, and Bartlett’s test of sphericity. The Bartlett’s test of
sphericity was significant (Chi-squared = 87.5; p<0.01) and Kaiser-Meyer-Olikn
values were greater than 0.5 for each test (mean 0.63 ± standard deviation 0.05),
which indicate these data are suitable for PCA (Weaving, Marshall, Earle, Nevill,
& Abt, 2014). Principle component analysis reduces data to unique components
containing variables which correlate with each other, whilst the principle
components themselves do not correlate (Weaving et al., 2014). Principle
components were extracted when an eigenvalue was greater than 1. Visual
inspection of the screen plot confirmed those components extracted were above
the break point in the data, as recommended (Costello & Osborne, 2005).
A General Linear Model was used to evaluate differences between male and
female athletes, and between maturation groups for each FMSTM task. Uncertainty
of the estimate was expressed as 90 percent confidence intervals. Inferences were
based on the disposition of the confidence interval for the mean difference to the
6
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
smallest-important difference. It is common for authors to choose a mathematical
proxy (Reider, 2015) for the smallest-important difference, which is typically
recommended as 0.2 between-subject pooled standard deviations (Hopkins,
Marshall, Batterham, & Hanin, 2009). For example, this would be equivalent to
0.13 AU for our deep squat. However, in practice the minimal important
detectable change when assessing an individual on each of the seven FMSTM tasks
is 1 AU, which could be viewed as conservative when making comparisons at a
group level. For example, a difference of 1 AU for the deep squat in our data
would represent a large effect size (1.5 SDs). Thus, we present inferences based
upon both this conservative smallest-important difference of 1AU, and a more
liberal 0.2 between-athlete pooled standard deviations.
The probability that between-group differences in task score were positive,
negative or trivial, was identified according to the magnitude-based inferences
approach (Batterham & Hopkins, 2006). Descriptors were assigned using the
following scales: 0.5–5 % very unlikely; 5–25 % unlikely; 25–75 % possibly; 75–
95 % likely; 95–99.5 % very likely; > 99.5 % most likely. An effect was deemed
unclear if the confidence limits overlapped the smallest positive or negative
change by ≥ 5 % (Hopkins et al., 2009). The magnitude of responses was
evaluated through standardised differences in the means, using the following
thresholds: < 0.2 trivial; < 0.6 small; < 1.2 moderate; < 2 large; < 4 very large; ≥ 4
extremely large (Hopkins et al., 2009).
7
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
Results
The seven tasks of the FMSTM showed ‘poor’ to ‘questionable’ internal
consistency (Cronbach’s alpha, ±90% confidence limit; 0.53, ±0.10; Ordinal
alpha, 0.60, ±0.09), and weak correlations between tasks (Table 2). We observed a
‘very likely’, moderate sex difference in correlation between trunk-stability push-
up and active straight-leg raise (0.47 ±0.29). Here, trunk stability push-up had a
‘likely’ small positive relationship with active straight-leg raise in girls (0.25
±0.16) but a ‘likely’ small negative relationship in boys (-0.22 ±0.23).
***** Table 2 near here *****
Two components were extracted from the principle component analysis,
representing 47% of the total variance. Component 1 accounted for 28% of the
variance, and component 2 accounted for a further 19%. The highest loadings for
principle component 1 were the in-line lunge (0.71), rotatory stability (0.67) and
hurdle step (0.61) tasks. The highest loadings for principle component 2 were
shoulder mobility (0.73) and active straight-leg raise (0.66) tasks (Figure 1).
***** Figure 1 near here *****
Differences between athletes who were categorised as before-, at-, or after-PHV
are presented in figure 2, and sex differences in figure 3. We also present the
percentage likelihood that these differences were negative, trivial or positive in
these figures. Decrements in component 1 tasks were observed in athletes
categorised as before-, compared to those at-, and those after-PHV (Figure 2).
Compared to those at-PHV, these decrements were ‘likely’ small for in-line lunge
(-0.47, ±0.49), hurdle step (-0.38, ±0.49), and trunk-stability push-up (-0.51,
8
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
±0.45), and ‘possibly’ small, but also ‘possibly’ trivial, for rotator stability (-0.31,
±0.48). Compared to those after-PHV, decrements were ‘likely’ small for rotatory
stability (-0.35, ±0.37), ‘possibly’ small for the hurdle step (-0.35, ±0.37), and
‘possibly’ small, ‘possibly’ trivial for in-line lunge (-0.34, ±0.37), and trunk-
stability push-up (-0.30, ±0.34). ‘Possibly’ small, ‘possibly’ trivial increases were
observed in component 2 tasks; active straight-leg raise (0.20, ±0.36), and
shoulder mobility (0.19, ±0.36). ‘Possibly’ small, ‘possibly’ trivial improvements
were also observed in trunk stability push-up (0.21, ±0.38), and active straight-leg
raise (0.30, ±0.34), in athletes at-, compared with those after-PHV. Shoulder
mobility score was ‘likely’ higher in girls, but the differences were small (0.46,
±0.29). Conversely, trunk stability push-up score was ‘most likely’ moderately
higher (0.73, ±0.28), in boys (Figure 3). No observations were greater than the
minimal practical threshold of 1 AU.
***** Figures 2 and 3 near here *****
Discussion
The primary aim of this study was to examine the internal consistence and factor
structure within the seven FMSTM tasks in youth athletes. The factor structure of a
scale is a critical component of test validation when a measure is assumed to be
uni-dimensional (Gorsuch, 1983; Kelleher et al., 2018). FMSTM total score is
commonly used within youth athlete programmes but the internal consistency, and
factor structure had not been explored in this cohort. Our key findings indicate
that the FMSTM tasks exhibited ‘poor’ or ‘questionable’ internal consistence, and
principle component analysis revealed two interpretable components. Component
9
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
1 included tasks that rely heavily on stability, and component 2 clearly
represented joint mobility. Most correlations between tasks were either trivial or
small, except for the deep squat and in-line lunge, which were moderately
correlated. As such, the FMSTM total score cannot be viewed as a gestalt measure
for movement competence, and practitioners working with youth athletes should
consider each individual task element separately.
The internal consistency of FMSTM tasks was similar to previous research
(Kazman et al., 2014; Li et al., 2015). Our Chronbach’s alpha was greater than
that reported by Kazman et al., (2014) in marine officers (0.39), but similar to the
value reported in elite athletes (0.59) by Li et al., (2015). Chonbach’s, and ordinal
alpha were lower than reported in a general health care population (0.64 and 0.73
respectively) (Koehle et al., 2016). This likely reflects the highly complex multi-
dimensional nature of movement. The seven tasks appear to represent different
constructs, but their validity as individual movement assessment tools within
athletic development programmes are yet to be established, thus critical evaluation
of each task is recommended.
Principle component analysis is a method used for data reduction, for example,
Williams, Trewartha, Cross, Kemp, & Stokes, (2017). These authors considered
factor loadings exceeding ± 0.7, to be well defined. Using this criterion, our data
suggests the in-line lunge and shoulder mobility could be chosen to provide an
overview of functional movement, as defined by FMSTM total score. However,
given the poor internal consistency, weak correlations between tasks, and both
10
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
components only accounting for 47% of the total variance, we would support
previous recommendations (Li et al., 2015) that in its current form, there is limited
redundancy within the FMSTM.
Within component 1, the in-line lunge was the only task with loading exceeding
0.7. Factor loadings are essentially a correlation coefficient and more liberal
loading thresholds have been advocated, with ‘strong’ loadings exceeding 0.5
desirable (Costello & Osborne, 2005). Alongside the in-line lunge; rotatory
stability, hurdle step, and trunk stability push-up were loaded strongly to
component 1. The commonality between these tasks is the requirement of the
body to stabilise throughout the movement. Only two tasks, active straight-leg
raise, and shoulder mobility loaded strongly on component 2. These both require
joint mobility and the least degree of stability and control (Cook, Burton, Kiesel,
Bryant, & Torine, 2010). In summary, we show the FMSTM represents two clear
components in young athletes; tasks that require a high degree of stability and
tasks that require a high degree of mobility.
A second aim of our study was to evaluate the effect of maturation on individual
FMSTM task scores. Differences between maturation groups was clearest in those
tasks loading highest on component 1, suggesting stability is developed
throughout maturation. These findings support previous research in boys academy
football players (Lloyd et al., 2015; Portas et al., 2016).
In-line lunge and rotatory stability loaded highest on component 1, and have
previously shown strong relationships with physical performance tests in young
athletes (Lloyd et al., 2015; Okada, Huxel, & Nesser, 2011). These studies
11
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
correlate relatively homogenous data which could be open to errors and caution
should be taken when interpreting these relationships (Atkinson & Nevill, 2001).
Despite this, training interventions that target stability of both the trunk and
contra-lateral limb are likely to be important throughout maturation. Indeed, a
recent pilot study showed suspension training, which inherently challenges
stability is a promising modality for improving FMSTM and upper body muscular
performance in youth athletes (Laurent et al., 2018).
Trunk stability push-up task score was higher in athletes who were at-, or after-
PHV, when compared with those who were before-PHV. Interestingly, athletes
who were after-PHV, showed a ‘possibly’ small decrement in this task compared
to those at-PHV. This may be due to the greater proportion of female athletes
(67%) in our sample, which was most marked after-PHV (74%). Upper body
strength is a substantial contributor to success in the trunk stability push-up, and
boys have shown a natural development of strength post puberty, compared to a
plateaux in girls (Catley & Tomkinson, 2013; Tomkinson et al., 2017). This likely
explains why we observed moderately lower trunk stability push-up scores in
girls.
Significant sex differences have been shown previously for the trunk stability
push-up, as well as in-line lunge and rotatory stability in children (Abraham et al.,
2015). This is an important consideration when evaluating our findings, as the rate
of physical development is markedly slower in girls compared with boys during
the teenage years, with the exception of flexibility (Catley & Tomkinson, 2013;
Malina, Sławinska, Ignasiak, & Rożek, 2010; Tomkinson et al., 2017).
12
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
Girls who naturally develop flexibility post puberty (Catley & Tomkinson, 2013;
Tomkinson et al., 2017), were likely to score higher on shoulder mobility. There
was no clear difference between boys and girls for the active straight-leg raise.
This may reflect the active nature of a task that requires hip strength and stability
to raise and hold the limb. Indeed, the association between trunk stability push-up
and active straight-leg raise was very likely different between boys and girls.
Neither sex, nor maturation, had a substantial effect on FMSTM task performance,
when the smallest-important difference was set as the minimal practical value of 1
unit. This suggests that either the observations made here are not practically
meaningful, or that the 4-point categorical scale used may not be sensitive enough
to detect the small or moderate differences in individual tasks between sexes or
maturation groups. The validity of this scale for measurement has been questioned
recently (Philp et al., 2018), and some researchers have suggested using a more
detailed 100-point scale (Butler, Plisky, & Kiesel, 2011). Further research is
necessary to investigate alternative scoring systems (Kelleher et al., 2018).
There are limitations to our study which should be acknowledged. Firstly, the data
was collected in real time by different raters, which could increase noise. This
impact was limited by the lead author reviewing videos post screening to identify
differences of opinions before discussing and agreeing final scores. Secondly, the
prediction of maturational status using maturity offset may overestimate the
timing of peak-height velocity. However, the equations used are reliable (Mills,
Baker, Pacey, Wollin, & Drew, 2016), and alternatives based upon skeletal age or
secondary sex characteristics are impractical (Cumming, Lloyd, Oliver,
Eisenmann, & Malina, 2017). Finally, the sample used in the current study is
13
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
biased towards female athletes and those after-PHV. Given the clear sex
differences in correlations between certain tasks presented in table 2, principle
component analysis would ideally be performed separately for boys and girls.
However, a larger sample size would be desirable for such analyses.
The magnitude-based inference approach that we have taken is advantageous
when compared with null hypothesis testing (Batterham & Hopkins, 2006;
Hopkins et al., 2009; Page, 2014). However, the underpinning theory has been
questioned (Sainani, 2018; Welsh & Knight, 2015) and subsequently defended
(Hopkins & Batterham, 2016; Hopkins & Batterham, 2018). Given the
fundamental limitations to null-hypothesis testing (Page, 2014), and the
difficulties in choosing an appropriate prior distribution to inform a full Bayesian
analysis, magnitude-based inferences, by choosing a dispersed uniform prior,
provides the most appropriate approach in this instance. Key to the application of
magnitude-based inference, is identifying a value equivalent to the smallest-
important difference. A strength of this study was that we provided analysis based
upon two different choices of smallest-important difference, to help the reader
understand the practical value of these data. Finally, ‘possibly’ small differences
are often also ‘possibly’ trivial. Such instances have been clearly stated in our
results section.
Conclusion
This is the first paper to evaluate the factor structure of the FMSTM in a youth
athlete population. Our findings support previous research in demonstrating the
FMS TM has poor internal consistency, and FMSTM total score should not be used
as a gestalt measure of fundamental movement in this cohort. Principle
14
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
component analysis revealed two factors which represented stability or mobility.
Stability is likely influenced by maturation, but maybe an important construct to
develop in young athletes. Whilst the FMSTM may have limited redundancy in its
current form, modifications to the seven tasks, and improvements to the 4-point
scoring scale could be warranted to ensure the FMSTM is fit for purpose to assess
movement competence in young athletes.
15
382
383
384
385
386
387
Table 1. Participant characteristics presented as mean ± standard deviations,
median and range for boys and girls before-, at and after- peak-height velocity.
Table 2. Pearson’s correlation matrix for all participants. Boys and girls with the
uncertainty in the estimate represented with ±90% confidence limits. The
likelihood the correlation was either; trivial (<0.1), small (0.1 – 0.3), moderate
(0.3 to 0.5), large (0.5 – 0.7) of very large (0.7 – 0.9) or near perfect (>0.9) was
represented qualitatively (Hopkins et al., 2009). Possible, likely or very likely sex
differences are highlighted in the table.
Table 3. Raw data (mean ± standard deviation) in each of the seven FMSTM tasks.
Data is presented for all participants and split by both sex and maturation group.
Figure 1: Factor weighting for FMSTM tasks after extraction and promax rotation.
Circles represent tasks within component 1 (stability) and triangles represent tasks
within component 2 (mobility).
Figure 2. Differences in FMSTM task score between children before, at, or after
peak-height velocity. Data are presented as effect sizes for the mean difference
with 90% confidence limits. Circles represent tasks within component 1 (stability)
and triangles represent tasks within component 2 (mobility). Magnitude-based
inferences include the percentage likelihood that the task score was lower, trivial
or higher. Graphs A to C represent a smallest-important difference (SID) of 0.2
SD, and graph D to F a SID of 1 arbitrary unit (AU). Trunk-stability push-up is
represented as TUSP, and active straight-leg raise as ASLR.
Figure 3. Sex differences in FMSTM task score. Data are presented as effect sizes
for the mean difference with 90% confidence limits. Circles represent tasks within
component 1 (stability) and triangles represent tasks within component 2
16
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
(mobility). Magnitude-based inferences include the percentage likelihood that the
task score was higher in girls, trivial, or higher in boys respectively. Graph A
represents a smallest-important difference (SID) of 0.2 SD and graph B a SID of 1
arbitrary unit (AU). Trunk-stability push-up is represented as TUSP, and active
straight-leg raise as ASLR.
References
Abraham A, Sannasi R, Nair R. (2015). Normative values for the Functional Movement ScreenTM in adolescent school aged children. International Journal of Sports Physical Therapy, 10(1): 29-36.
Atkinson, G., & Nevill, A. M. (2001). Selected issues in the design and analysis of sport performance research. Journal of Sports Sciences, 19(10), 811–827. http://doi.org/10.1080/026404101317015447
Batterham, A.M., & Hopkins, W.G. (2006). Making meaningful inferences about magnitudes. International Journal of Sports Physiology and Performance, 1, 50–57. Retrieved from http://journals.humankinetics.com/ijspp
Bahr, R. (2016). Why screening tests to predict injury do not work—and probably never will…: a critical review. British Journal of Sports Medicine, 50(13), 776–780. http://doi.org/10.1136/bjsports-2016-096256
Bonazza, N. A., Smuin, D., Onks, C. A., Silvis, M. L., & Dhawan, A. (2017). Reliability, validity, and injury predictive value of the Functional Movement Screen: A systematic review and meta-analysis. American Journal of Sports Medicine, 45(3), 725–732. http://doi.org/10.1177/0363546516641937
Butler, R. J., Plisky, P. J., & Kiesel, K. B. (2011). Interrater reliability of videotaped performance on the Functional Movement Screen using the 100-Point scoring scale. Athletic Training & Sports Health Care, 4(3), 103–109. http://doi.org/10.3928/19425864-20110715-01
Catley, M. J., & Tomkinson, G. R. (2013). Normative health-related fitness values for children: analysis of 85347 test results on 9-17-year-old Australians since 1985. British Journal of Sports Medicine, 47(2), 98–108. http://doi.org/10.1136/bjsports-2011-090218
Cook, G., Burton, L., & Hoogenboom, B. (2006a). Pre-participation screening: the use of fundamental movements as an assessment of function - part 2. North American Journal of Sports Physical Therapy, 1(3), 132–139.
Cook, G., Burton, L., & Hoogenboom, B. (2006b). Pre-participation screening: The use of fundamental movements as an assessment of function–Part 1. North American Journal of Sports Physical Therapy, 1(2), 62–72.
Cook, G., Burton, L., Kiesel, K., Bryant, M., & Torine, J. (2010). Movement: functional movement systems: screening, assessment, and corrective strategies.CA: On Target Publications.
Costello, A. B., & Osborne, J. W. (2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical Assessment, Research & Evaluation, 10(7), Available online:
17
414
415
416
417
418
419
420
421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456
http://pareonline.net/getvn.asp?v=10&n=7. Cumming, S. P., Lloyd, R. S., Oliver, J. L., Eisenmann, J. C., & Malina, R. M.
(2017). Bio-banding in sport: applications to competition, talent identification, and strength and conditioning of youth athletes. Strength and Conditioning Journal, 39(2), 34–47. http://doi.org/10.1519/SSC.0000000000000281
Ford, P., Lloyd, R., Meyers, R., Moosavi, M., Oliver, J., Till, K., & Williams, C. (2011). The long-term athlete development model: physiological evidence and application. Journal of Sports Sciences, 29(4), 389–402. http://doi.org/10.1080/02640414.2010.536849
Gorsuch, R. L. (1983). Factor Analysis. Second Edition. Hillsdale, NJ: Laurence Erlbaum Associates.
Hopkins, W.G., & Batterham, A. M. (2016). Error rates, decisive outcomes and publication bias with several inferential methods. Sports Medicine, 46(10), 1563–1573. http://doi.org/10.1007/s40279-016-0517-x
Hopkins, W.G., & Batterham, A. M. (2018). The vindication of magnitude-based inference (draft 2). Sportsscience, 22, 19–27. Retrieved from sportsci.org/2018/mbivind.htm
Hopkins, W.G., Marshall, S. W., Batterham, A. M., & Hanin, J. (2009). Progressive statistics for studies in sports medicine and exercise science. Medicine & Science in Sports & Exercise, 41(1), 3–13. http://doi.org/10.1249/MSS.0b013e31818cb278
Hopkins, W.G. (2006). A spreadsheet for combining outcomes from several subject groups. Sportscience, 10, 51-53. sportsci.org/2006/wghcom.htm
Hulteen, R. M., Morgan, P. J., Barnett, L. M., Stodden, D. F., & Lubans, D. R. (2018). Development of foundational movement skills: A conceptual model for physical activity across the lifespan. Sports Medicine, 1–8. http://doi.org/10.1007/s40279-018-0892-6
Kazman, J. B., Galecki, J. M., Lisman, P., Deuster, P. A., & OʼConnor, F. G. (2014). Factor structure of the Functional Movement Screen in marine officer candidates. Journal of Strength and Conditioning Research, 28(3), 672–678. http://doi.org/10.1519/JSC.0b013e3182a6dd83
Kelleher, L. K., Beach, T. A. C., Frost, D. M., Johnson, A. M., & Dickey, J. P. (2018). Factor structure, stability, and congruence in the functional movement screen. Measurement in Physical Education and Exercise Science, 22(2), 109–115. http://doi.org/10.1080/1091367X.2017.1381608
Koehle, M. S., Saffer, B. Y., Sinnen, N. M., & MacInnis, M. J. (2016). Factor structure and internal validity of the Functional Movement Screen in adults. Journal of Strength and Conditioning Research / National Strength & Conditioning Association, 30(2), 540–546. http://doi.org/10.1519/JSC.0000000000001092
Kushner, A. M., Brent, J. L., Schoenfeld, B. J., Hugentobler, J., Lloyd, R. S., Vermeil, A., et al. (2015). The back squat: targeted training techniques to correct functional deficits and technical factors that limit performance. Strength and Conditioning Journal, 37(2), 13–60. http://doi.org/10.1519/SSC.0000000000000130
Laurent, C. W., Masteller, B., & Sirard, J. (2018). Effect of a suspension-trainer-based movement program on measures of fitness and functional movement in Children: A pilot study. Pediatric Exercise Science, 1–12. http://doi.org/10.1123/pes.2016-0278
18
457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506
Li, Y., Wang, X., Chen, X., & Dai, B. (2015). Exploratory factor analysis of the functional movement screen in elite athletes. Journal of Sports Sciences, 33(11), 1166–1172. http://doi.org/10.1080/02640414.2014.986505
Lloyd, R. S., & Oliver, J. L. (2012). The Youth Physical Development Model: A new approach to Long-Term Athletic Development. Strength and Conditioning Journal, 34(3), 61–72. http://doi.org/10.1519/SSC.0b013e31825760ea
Lloyd, R. S., Oliver, J. L., Radnor, J. M., Rhodes, B. C., Faigenbaum, A. D., & Myer, G. D. (2015). Relationships between functional movement screen scores, maturation and physical performance in young soccer players. Journal of Sports Sciences, 33(1), 11–19. http://doi.org/10.1080/02640414.2014.918642
Smith, C.A., Chimera, N.J., Wright, N.J., Warren, M. (2013) Interrater and intrarater reliability of the functional movement screen. Journal of Strength and Conditioning Research, 27(4), 982-987. doi: 10.1519/JSC.0b013e3182606df2
Malina, R., Sławinska, T., Ignasiak, Z., & Rożek, K. (2010). Sex differences in growth and performance of track and field athletes 11-15 years. Journal of Human Kinetics, 24, 79–85. http://doi.org/10.2478/v10078-010-0023-4
McCall, A., Carling, C., Nedelec, M., Davison, M., Le Gall, F., Berthoin, S., & Dupont, G. (2014). Risk factors, testing and preventative strategies for non-contact injuries in professional football: current perceptions and practices of 44 teams from various premier leagues. British Journal of Sports Medicine, 48(18), 1352–1357. http://doi.org/10.1136/bjsports-2014-093439
Mills, K., Baker, D., Pacey, V., Wollin, M., & Drew, M. K. (2016). What is the most accurate and reliable methodological approach for predicting peak height velocity in adolescents? A systematic review. Journal of Science and Medicine in Sport, 20(6), 1–6. http://doi.org/10.1016/j.jsams.2016.10.012
Mirwald, R. L., Baxter-Jones, A. D. G., Bailey, D. A., & Beunen, G. P. (2002). An assessment of maturity from anthropometric measurements. Medicine & Science in Sports & Exercise, 34(4), 689–694. http://doi.org/10.1097/00005768-200204000-00020
Newton, F., McCall, A., Ryan, D., Blackburne, C., Fünten, der, K. A., Meyer, T., et al. (2017). Functional Movement Screen (FMSTM) score does not predict injury in English Premier League youth academy football players. Science and Medicine in Football, 1(2), 102–106. http://doi.org/10.1080/24733938.2017.1283436
Okada, T., Huxel, K. C., & Nesser, T. W. (2011). Relationship Between Core Stability, Functional Movement, and Performance. Journal of Strength and Conditioning Research, 25(1), 252–261. http://doi.org/10.1519/JSC.0b013e3181b22b3e
Page, P. (2014). Beyond statistical significance: clinical interpretation of rehabilitation research literature. The International Journal of Sports Physical Therapy, 9(5), 726–736.
Philp, F., Blana, D., Chadwick, E. K., Stewart, C., Stapleton, C., Major, K., & Pandyan, A. D. (2018). Study of the measurement and predictive validity of the Functional Movement Screen. BMJ Open Sport & Exercise Medicine, 4(1), e000357. http://doi.org/10.1136/bmjsem-2018-000357
Portas, M. D., Parkin, G., Roberts, J., & Batterham, A. M. (2016). Maturational effect on Functional Movement Screen™ score in adolescent soccer players.
19
507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556
Journal of Science and Medicine in Sport, 19(10), 854–858. http://doi.org/10.1016/j.jsams.2015.12.001
Reider, B. (2015). Good, or just better? American Journal of Sports Medicine, 43(8), 1841–1843. http://doi.org/10.1177/0363546515595612
Sainani, K. L. (2018). The Problem with “Magnitude-Based Inference.” Medicine & Science in Sports & Exercise, 1–55. http://doi.org/10.1249/MSS.0000000000001645
Tomkinson, G. R., Tomkinson, G. R., Carver, K. D., Atkinson, F., Daniell, N. D., Lewis, L. K., et al. (2017). European normative values for physical fitness in children and adolescents aged 9-17 years: results from 2 779 165 Eurofit performances representing 30 countries. British Journal of Sports Medicine. http://doi.org/10.1136/bjsports-2017-098253
Viru, A., Loko, J., Harro, M., & Volver, A. (1999). Critical periods in the development of performance capacity during childhood and adolescence. European Journal of Physical Education, 4, 75-119. http://doi.org/10.1080/1740898990040106
Weaving, D., Marshall, P., Earle, K., Nevill, A., & Abt, G. (2014). Combining internal- and external-training-load measures in professional rugby league. International Journal of Sports Physiology and Performance, 9(6), 905–912. http://doi.org/10.1123/ijspp.2013-0444
Welsh, A. H., & Knight, E. J. (2015). “Magnitude-based inference.” Medicine & Science in Sports & Exercise, 47(4), 874–884. http://doi.org/10.1249/MSS.0000000000000451
Williams, S., Trewartha, G., Cross, M. J., Kemp, S. P. T., & Stokes, K. A. (2017). Monitoring What Matters: A systematic process for selecting training-load measures. International Journal of Sports Physiology and Performance, 12(Suppl 2), S2–101–S2–106. http://doi.org/10.1123/ijspp.2016-0337
Wright, M. D., Portas, M. D., Evans, V. J., & Weston, M. (2015). The effectiveness of 4 weeks of fundamental movement training on functional movement screen and physiological performance in physically active children. Journal of Strength and Conditioning Research, 29(1), 254–261. http://doi.org/10.1519/JSC.0000000000000602
20
557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588
Table 1. Participant characteristics presented as mean ± standard deviations, median and range for boys and girls before-, at and after- peak-
height velocity.
Age (years) Stature (cm) Body mass (kg) Age from PHV (years)Maturation Group (n) Mean ± SD Median (Range) Mean ± SD Median (Range) Mean ± SD Median (Range)
Mean ± SD Median (Range)
Before-PHV (26) 11.9 ± 1.8 12 (8 - 14) 146 ± 10.0 149 (125 - 162) 39.9 ± 15.1 39.0 (33.5 – 66.0) -1.7 ± 0.9 -1.5 (-4.0 – -0.6)Boys (11) 13.6 ± 0.7 14 (13 – 14) 152 ± 7.8 154 (138 – 162) 46.1 ± 10.1 45.0 (33.5 – 66.0) -1.8 ± 1.1 -1.4 (-4.0 – -0.6) Girls (15) 10.6 ± 1.2 10 (8 – 13) 142 ± 9.3 139 (125 – 159) 36.8 ± 8.9 32.0 (26.0 – 52.0) -1.7 ± 0.8 -1.7 (-2.9 – -0.6)
At-PHV (21) 13.9 ± 1.2 13 (10 - 15) 160 ± 12 161 (132 - 180) 56.5 ± 16.3 54.5 (32.0 – 86.0) 0.1 ± 0.3 0.1 (-0.5 – 0.4)Boys (12) 14.6 ± 0.9 15 (13 – 15) 168 ± 6.9 168 (155 – 180) 65.0 ± 14.8 62.0 (46.0 – 86.0) 0.1 ± 0.3 0.3 (-0.4 – 0.4)Girls (9) 13.0 ± 1.1 13 (10 – 14) 150 ± 9.4 150 (132 – 161) 45.9 ± 11.4 44.0 (32.0 – 64.0) 0.0 ± 0.3 0.1 (-0.5 – 0.4)
After-PHV (97) 14.6 ± 1.8 14 (10 - 18) 166 ± 12 165 (13 - 263) 59.9 ± 14.5 59.0 (47.0 - 100) 2.3 ± 1.3 2.1 (0.6 – 5.4)Boys (25) 15.4 ± 1.3 16 (12 – 18) 177 ± 12 168 (155 – 180) 69.7 ± 13.2 63.0 (52.0 – 100) 1.8 ± 0.8 1.9 (0.5 – 2.9)Girls (72) 14.5 ± 1.9 14 (10 – 18) 161 ± 8.8 160 (132 – 161) 56.5 ± 13.6 55.0 (30.0 – 87.0) 2.5 ± 1.5 2.1 (-0.6 – 6.1)
Key: PHV = Peak Height Velocity
1
589
590
591
Table 2. Pearson’s correlation matrix for all participants. Boys and girls with the uncertainty in the estimate represented with ±90% confidence limits. The likelihood the correlation was either; trivial (<0.1), small (0.1 – 0.3), moderate (0.3 to 0.5), large (0.5 – 0.7) of very large (0.7 – 0.9) or near perfect (>0.9) was represented qualitatively (Hopkins et al., 2009). Possible, likely or very likely sex differences are highlighted in the table.
All (n = 144)
Deep squat In-line lunge Hurdle step Rotational stability Shoulder mobility TSPU
In-line lunge Possibly moderate(0.32, ±0.12) -
Hurdle step Very likely small(0.26, ±0.13)
Very likely small (0.28, ±0.13) -
Rotational stability
Trivial(0.09, ±0.14)
Very likely small (0.26, ±0.13)
Very likely small (0.28, ±0.13) -
Shoulder mobility Trivial(0.08, ±0.14)
Trivial(0.08, ±0.14)
Trivial(0.09, ±0.14)
Likely small (-0.16, ±0.13) -
TSPU Trivial(0.09, ±0.14)
Very likely small (0.29, ±0.13)
Trivial(0.09, ±0.14)
Very likely small(0.22, ±0.13)
Trivial(-0.03, ±0.14) -
ASLR Possibly small(0.14, ±0.13)
Likely small(0.21, ±0.13)
Possibly small (0.12, ±0.14)
Trivial(-0.07, ±0.14)
Likely small(0.19, ±0.13)
Trivial(0.05, ±0.14)
Girls (n = 96)
In-line lunge Possibly moderate(0.30, ±0.15)
Hurdle step Possibly moderate(0.31, ±0.15)*
Possibly moderate(0.32, ±0.15)
Rotational stability
Trivial(0.07, ±0.17)
Possibly moderate(0.30, ±0.15)
Very likely small (0.28, ±0.16)
Shoulder mobility Trivial (-0.02, ±0.17)**
Trivial(0.01, ±0.17)
Trivial(0.03, ±0.17)
Likely small(-0.22, ±0.16)
TSPU Trivial (0.06, ±0.17)
Likely small (0.29, ±0.16)
Possibly small(0.10, ±0.17)
Possibly small (0.10, ±0.17)**
Trivial (0.09, ±0.17)
ASLR Trivial (0.08, ±0.17)**
Likely small(0.25, ±0.17)
Possibly small0.11, ±0.17)
Trivial(-0.04, ±0.17)
Possibly small (0.10, ±0.17)*
Likely small(0.25, ±0.16)***
Boys (n = 48)
In-line lunge Possibly moderate(0.36, ±0.21)
Hurdle step Possibly small(0.16, ±0.23)*
Likely small (0.21, ±0.23)
Rotational stability
Unclear(0.12, ±0.24)
Likely small(0.20, ±0.23)
Possibly moderate(0.32, ±0.22)
Shoulder mobility Likely small(0.24, ±0.23)**
Likely small(0.20, ±0.23)
Possibly small(0.17, ±0.23)
Trivial(-0.05, ±0.24)
TSPU Trivial (-0.01, ±0.24)
Possibly moderate(0.32, ±0.22)
Possibly small (0.15, ±0.24)
Likely moderate(0.42, ±0.20)**
Trivial(-0.01, ±0.24)
ASLR Possibly moderate(0.3, ±0.22)**
Possibly small(0.15, ±0.24)
Likely small(0.13, ±0.24)
Unclear(-0.1, ±0.24)
Possibly moderate(0.30, ±0.22)*
Likely small (-0.22, ±0.23)***
1
592593594
Key: Possibly (*), likely (**) or very likely (***) sex differences. TSPU = Trunk-stability push up. ASLR = Active straight-leg raise
2
Table 3. Raw data (mean ± standard deviation) for in each of the seven FMSTM tasks. Data is presented in all participants and split by both sex and maturation group.
All(n = 144)
Male(n = 48)
Female(n = 96)
Before-PHV(n = 26)
At-PHV(n = 21)
After-PHV(n = 97)
Mean ± standard deviationDeep squat 1.8 ± 0.6 1.8 ± 0.7 1.8 ± 0.6 1.8 ± 0.6 1.8 ± 0.8 1.8 ± 0.6
In-line lunge 1.8 ± 0.6 1.9 ± 0.7 1.8 ± 0.6 1.7 ± 0.6 2.0 ± 0.6 1.9 ± 0.6Hurdle step 1.8 ± 0.5 1.8 ± 0.6 1.8 ± 0.5 1.7 ± 0.6 1.9 ± 0.5 1.8 ± 0.5
Rotator stability 1.5 ± 0.6 1.6 ± 0.6 1.5 ± 0.5 1.3 ± 0.6 1.5 ± 0.5 1.6 ± 0.6Shoulder mobility 2.4 ± 0.7 2.2 ± 0.8 2.5 ± 0.6 2.5 ± 0.7 2.3 ± 0.7 2.4 ± 0.7
Trunk-stability push-up 1.8 ± 0.8 2.1 ± 0.7 1.6 ± 0.7 1.6 ± 0.6 2.0 ± 0.7 1.8 ± 0.8Active straight-leg raise 1.9 ± 0.7 1.8 ± 0.8 2.0 ± 0.7 2.0 ± 0.7 2.1 ± 0.6 1.9 ± 0.8Key: PHV = Peak-height velocity and n = number of participants
1
595
596597
Figure 1
1
598
599600601
Figure 2
2
602603
604
Figure 3
1
605606
607