FUNCTIONAL MOVEMENT SCREEN AS A PREDICTOR OF INJURY IN HIGH
SCHOOL BASKETBALL ATHLETES
by
ERIC A. SORENSON
A DISSERTATION
Presented to the Department of Human Physiology
and the Graduate School of the University of Oregon
in partial fulfillment of the requirements
for the degree of
Doctor of Philosophy
December 2009
11
University of Oregon Graduate School
Confirmation of Approval and Acceptance of Dissertation prepared by:
Eric Sorenson
Title:
"Functional Movement Screen as a Predictor of Injury in High School Basketball Athletes"
This dissertation has been accepted and approved in partial fulfillment of the requirements for
the degree in the Department of Human Physiology by:
Gary Klug, Chairperson, Human Physiology
Paul van Donkelaar, Member, Human Physiology
Andrew Lovering, Member, Human Physiology
Roland Good, Outside Member, Special Education and Clinical Sciences
and Richard Linton, Vice President for Research and Graduate Studies/Dean of the Graduate
School for the University of Oregon.
December 12,2009
Original approval signatures are on file with the Graduate School and the University of Oregon
Libraries.
© 2009 Eric A. Sorenson
iii
An Abstract of the Dissertation of
Eric A. Sorenson for the degree of
in the Department ofHuman Physiology to be taken
IV
Doctor of Philosophy
December 2009
Title: FUNCTIONAL MOVEMENT SCREEN AS A PREDICTOR OF INJURY IN
HIGH SCHOOL BASKETBALL ATHLETES
Approved: _
Gary A. Klug, PhD
Participation in athletics includes an inherent risk of becoming injured that is
related to the nature of the games and activities of the players. Current literature reports
that approximately seven million high school students participate in sports yearly in the
United States and, during the 2005-2006 sport seasons, 1.4 million injuries were reported.
Considering this high number of injuries, there is little doubt that definitive research into
the determination of factors that might help predict the degree of injury risk associated
with sport participation is warranted.
Despite common association of variables such as joint laxity, range of motion,
strength and balance with injury, these traditional measures have not proven to be reliable
predictors of vulnerability. Consequently, attempts have been made to identify practical
methods that may better permit identification of individuals who show a high likelihood
of injury during athletic competition.
vThis study examined one such system, the Functional Movement Screen (FMS),
which utilizes measures of mobility and stability to permit its developers to assert that it
can be used to practically and accurately identify vulnerable atWetes. Critical data on
inter-rater and intra-rater performance were first obtained on a team of atWetic trainers to
ensure that they could reliably execute the testing methods. Following confirmation of
this fact, 112 high school basketball athletes were screened with the FMS and their
injuries (non-contact neuromusculoskeletal tissue damage in school-sanctioned
basketball) were tracked throughout an entire season. Data analysis to determine if a
commonly-used FMS cutoff score of less than 14 out of21 could identify vulnerable
athletes revealed that this value was not significantly related to the likelihood of
sustaining an injury. Furthermore, logistic regression revealed that none ofthe individual
predictors (gender, FMS movements, and movement asymmetries) were significant
predictors of injury susceptibility. The results indicate that, despite the fact that multiple
evaluators and trials can be practically used to evaluate FMS scores in a large group of
high school basketball athletes, the test does not appear to be a valid tool in assessing
injury risk in this population during an entire season.
VI
CURRICULUM VITAE
NAME OF AUTHOR: Eric A. Sorenson
PLACE OF BIRTH: Anaheim, California
DATE OF BIRTH: December 13,1980
GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED:
University of Oregon
Point Lorna Nazarene University
DEGREES AWARDED:
Doctor of Philosophy, 2009, University of Oregon
Master of Science, 2006, University of Oregon
Bachelor of Arts, 2003, Point Lorna Nazarene University
AREAS OF SPECIAL INTEREST:
Sports Medicine and Health Sciences
PROFESSIONAL EXPERIENCE:
Adjunct Faculty, Exercise and Movement Science Program,
Lane Community College, Eugene, OR, 2009
Athletic Trainer, Tensegrity Physical Therapy, Eugene, OR, 2007-2009
Graduate Teaching Fellow, Department of Human Physiology,
University of Oregon, Eugene, OR, 2004 - 2009
National Athletic Trainer's Association, Certified Athletic Trainer, 2003
GRANTS, AWARDS AND HONORS:
Core Training Systems, Research Grant In-kind, 2008
National Athletic Trainer's Association, Osternig Master's Research Grant, 2006
Northwest Health Foundation, Research Grant, 2006
Professional Baseball Athletic Trainer's Association, Scholarship, 2006
Vll
ACKNOWLEDGMENTS
I would like to express my sincere appreciation for the guidance of several
individuals that have greatly impacted my education at the University of Oregon. I am
thankful for Dr. Susan Verscheure, my friend and mentor, from whom I have gained
many insights about teaching, scholarship, the outdoors, and living life in the diverse
culture of Eugene. Rick Troxel has been a supportive mentor that guided my exploration
of educational technology and opened my eyes to learning that goes far beyond Human
Physiology or Sports Medicine. Dr. Louis Ostemig opened doors for me to gain
admission to the university, and was a great example of conducting high caliber research
and critical thinking.
I am also thankful to Dr. Leon "Kug" Kugler for being a good friend and mentor.
He has guided me spiritually, academically and supported my life's journey. Kug is a
phenomenal example of how to live life with God's love for others. His humility and
mission to support students has shaped my future in academia and beyond.
I would like to thank my dissertation committee for all the insight they gave me in
formulating and conducting my research. Dr. Klug was instrumental in providing
constructive feedback to strengthen this clinical study, and he thoroughly inspected and
shaped my final manuscript. Dr. van Donkelaar and Dr. Lovering were helpful in making
sure my protocol was well thought out and provided insights for how to better support the
clinical practices that I refer to in this paper. I would like to thank Dr. Good for sharing
his expertise in statistics and for being an educator in the process of helping me with my
analyses. I enjoyed the meetings I had with each one of you.
Vlll
I would like to thank my great group of friends for their impact on my life. To my
best friends in California - thanks for being supportive from so far away! I always look
forward to seeing you. Brent, Molly, Susan, Dustin, Cindie and Brian - I am so thankful
for the many adventures we have shared. Thank you for bringing me in as a friend and
making my Eugene experience so great. I would also like to thank all of my Athletic
Training classmates. We shared many fun times at the University of Oregon since 2003,
which I will always remember. To Sean and Rachel Roach - thank you for bringing me
\
in as an employee at Tensegrity Physical Therapy and for investing in my education.
Also, to those of you that assisted me with my data collection, both from the U of 0
Graduate Athletic Training Program and Eugene/Springfield high schools, many thanks
for your support. This research would not be possible without your help.
Most importantly, I would like to thank my family and Nicole for all their
support. While the majority of days on this adventure have been stimulating and
enjoyable, I am thankful for your support during the challenging moments. To my
parents, I am so grateful for your love, encouragement and financial support. To Jeff and
his family, moving to Oregon and being closer to you has been a blessing. Thanks for
always welcoming me to your house and for many fun adventures in the Northwest. And
to Nicole, my future wife, the best thing about moving to Eugene was finding you! I am
stoked to share my life with you. Thank you for your support and for being such an
amazing companion with whom to create so many memories. I can't wait to get married
in summer 20l0!
This research was supported by Core Training Systems, Research Grant In-kind.
IX
DEDICATION
To my parents, for their endless support and inspiration to achieve great things.
To my brother Jeff, for his support and inspiration to live an adventurous life each day.
And to Nicole, my future wife and companion.
I love each of you.
xTABLE OF CONTENTS
Chapter Page
I. INTRODUCTION 1
Operational Definitions 1
Risk Factors of Injury : 2
Previous Injury 2
Ligamentous Laxity 2
Range of Motion 5
Muscle Strength 8
Balance 11
Summary 12
Functional Testing 14
Deep Squat 16
Hurdle Step 17
In-line Lunge 18
Shoulder Mobility 19
Active Straight Leg Raise 20
Trunk Stability Push-Up 21
Rotary Stability 22
Study Description 23
Specific Aim #1 23
Specific Aim #2 24
Hypotheses 26
II. METHODS 27
The Functional Movement Screen 27
Deep Squat. 27
Hurdle Step 28
In-line Lunge 29
Shoulder Mobility 29
Active Straight Leg Raise 30
Trunk Stability Push-Up 30
Rotary Stability 31
Research Studies 32
Reliability Studies 32
Subjects 32
Data Analysis 33
Inter-rater Reliability Study 33
Intra-rater Reliability Study 34
Interpretation of Data 35
Injury Prediction Study 36
Subjects 36
Functional Movement Screen Rating 36
Chapter
Xl
Page
Injury Data 37
Data Analysis , 38
III. RESULTS 41
Reliability Studies 41
Inter-rater Reliability Study , 42
Intra-rater Reliability Study 44
Injury Prediction Study 45
Demographics 45
Injuries 50
Injury Prediction with Cutoff Score of 14 52
Injury Prediction with Logistic Regression 55
Previous Injury 56
IV. DISCUSSION 57
Statement of the Problem 57
Reliability 60
FMS Performance 62
Injuries 63
FMS Injury Prediction 65
APPENDICES 69
A. FUNCTIONAL MOVE~·'IENT SCREEN SCORING RUBRIC 69
B. INFORMED CONSENT FORM RELIABILITY STUDy 71
C. AGREEMENT FOR VIDEOTAPING 74
D. FUNCTIONAL MOVEMENT SCREEN SCORING SHEET 76
E. INFORMED CONSENT FORM INJURY PREDICTION STUDy 78
F. DEMOGRAPHICS FORM 81
G. INJURY REPORT FORM 83
REFERENCES 85
XlI
LIST OF FIGURES
Figure Page
1.1. Grade I Sprain of Lateral Ankle Ligaments - Firm End-FeeL 3
1.2. Grade II Sprain of Lateral Ankle Ligaments- Soft End-FeeL 4
1.3. Grade III Sprain of Lateral Ankle Ligaments - Empty End-Feel 4
1.4. Goniometric Measurement of Joint ROM 5
1.5. Manual Muscle Testing of Knee Extension 8
1.6. Action of the Peroneal Muscles in the Prevention of an Inversion Ankle
Sprain 10
1.7. Balance Testing 12
1.8. Deep Squat. 16
1.9. Hurdle Step 17
1.10. In-line Lunge : 18
1.11. Shoulder Mobility 19
1.12. Active Straight Leg Raise 20
1.13. Trunk Stability Push-up 21
1.14. Rotary Stability 22
3.1. Comparison of FMS Movement Scores for Boys and Girls 48
3.2. Comparison of FMS Scores for Boys and Girls 48
X111
LIST OF TABLES
~k P~
1.1. Nonnal Range of Motion End-Feels 6
1.2. Abnonnal Range ofMotion End-Feels 7
1.3. Grading System for Manual Muscle Tests 9
2.1. Example Calculation for Median Reliability Coefficient... 34
3.1. Values for Functional Movement Screen of Reliability Study Participants 41
3.2. Inter-rater Reliability Results 43
3.3. Intra-rater Reliability Results 44
3.4. Values for Functional Movement Screen Tests and Final Score for Boys 46
3.5. Values for Functional Movement Screen Tests and Final Score for Girls 47
3.6. One-Way Analysis of Variance Summary Table for the Effects of Gender on
FMS Score 47
3.7. Functional Movement Screen Asymmetries for Boys 49
3.8. Functional Movement Screen Asymmetries for Girls 49
3.9. Number of Injuries by Region 51
3.10. One-Way Analysis of Variance Summary Table for the Effects ofInjury Type
on FMS Score 52
3.11. Chi-square Results with Cutoff of 14 , 53
3.12. Chi-square Results with Cutoffof 12 53
3.13. Chi-square Results with Cutoff of 13 53
3.14. Chi-square Results with Cutoff of 15 54
3.15. Chi-square Results with Cutoff of 16 54
3.16. Logistic Regression for Predictors in Isolation '" 55
3.17. One-Way Analysis of Variance Summary Table for the Effects of Previous
Injury on FMS Score 56
1CHAPTER I
INTRODUCTION
Operational Definitions
Injury - any tissue damage to skeletal, muscular, connective or neural tissues
Sprain - stretch or tear of a ligament
Strain - stretch or tear of a muscle or tendon
Mobility - the ability to produce a desired and efficacious motion
Stability - the capacity to resist a motion
Participation in athletics includes an inherent risk of becoming injured based upon
the nature of the games and activities of the players. Current literature reports that
approximately seven million high school students are participating in sports in the United
States (Rechel, Yard, & Comstock, 2008). Of these athletes, 1.4 million injuries were
sustained during the 2005-2006 sport seasons (Rechel et aI., 2008). A report of the 1995-
1997 high school sport seasons indicates that more than two million injuries were
sustained, requiring 500,000 doctor visits and 30,000 hospitalizations (Powell & Barber-
Foss, 1999). The volume of injuries reported in this setting, along with the fact that many
of the more significant sports-related injuries may lead to long-term physical impairment
(Powell & Barber-Foss, 1999), warrants research into the possibility of utilizing pre-
2participation screening methods that are able to identify athletes that are at a high risk of
becoming injured. If such determinations could be made, sports medicine professionals
could intervene to correct biomechanical deficits in an effort to promote safe participation
and reduce the incidence of injury.
The following describes several risk factors for injury and the typical methods for
evaluating an individual for those factors. Furthermore, potential use of a Functional
Movement Screen for assessing basic movement patterns that could possibly be useful in
pre-participation screening methods is described.
Risk Factors of Injury
Previous Injury
One variable that has been identified as a significant risk factor for injury is a
history of previous injury (Emery, Meeuwisse, & Hartmann, 2005). Research reports an
increase of four to five fold in the likelihood of reinjury at the site ofprevious injury for
high school cross country, football, soccer and cheer (Murphy, Connolly, & Beynnon,
2003; Emery et aI., 2005; Caine, Maffulli, & Caine, 2008). This may be related to
deficiencies resulting from the initial injury including increased ligamentous laxity,
altered range of motion, or decreased muscle strength or balance (Knowles et aI., 2006;
Caine et aI., 2008).
Ligamentous Laxity
The primary function of ligaments is to guide joint motion and, in the context of
injury, they also serve to control excessive joint motion. Ligamentous laxity describes the
3stiffness qualities within a joint's connective tissues, which can be evaluated through
specific ligamentous and capsular tests that are administered by health care practitioners
relying on the end-feel (the quality ofligamentous resistance at the end range ofjoint
motion) to grade the tests. Results quantifying end-feel are based on a four-point scale
with zero indicating normal, one, a firm, two, a soft, and three, an empty end-feel. Grades
one through three may also be accompanied by pain. The actual quality of this parameter
is determined by testing bilaterally to compare the injured to the uninjured joint. A firm
end-feel indicates slight stretching of the ligament with an end-feel close to that of the
healthy side. A soft end-feel indicates partial tearing of the ligamentous fibers with an
increased glide of the joint surfaces upon one another or the joint line gapping
significantly when compared to the contralateral side. An empty end-feel is consistent
with complete tearing of the ligament with excessive joint motion during the testing.
Figure 1.1: Grade I Sprain of Lateral Ankle Ligaments - Firm End-Feel
4Figure 1.2: Grade II Sprain of Lateral Ankle Ligaments - Soft End-Feel
Figure 1.3: Grade III Sprain of Lateral Ankle Ligaments - Empty End-Feel
When evaluating the knee, more sophisticated instrumentation (KT 1000/2000)
can be used to assess the anterior cruciate ligament. With these devices, anterior tibial
displacement is assessed in millimeters as an indirect measure of anterior cruciate
ligament laxity. Unfortunately, this type of measurement is not readily available for other
joints of the body and, therefore, most joints require manual assessments of the
parameter. Increased joint laxity following injury may also give a patient the feeling of
5instability, as the ligament no longer restrains excessive joint motion (McKay, Goldie,
Payne, & Oakes, 2001). Furthermore, the lack of stabilization provided by the injured
ligament may predispose an athlete to further injury (Ekstrand & Gillquist, 1983).
Range ofMotion
Range of motion (ROM) testing is another common assessment during a patient
evaluation. These measurements should be compared bilaterally and to normative data for
the joint (Starkey & Ryan, 2002). ROM can be determined via gross observation by the
practitioner or by using a goniometer. Goniometric ROM testing requires the
identification of the approximate axis ofjoint rotation so the fulcrum of the goniometer
can be placed at this location. Next, the stationary and movement arms are placed on the
proximal and distal segments, parallel to the respective bones. Once the joint is moved
through its full ROM, the amount of motion can be easily measured in degrees on the
goniometer.
Figure 1.4: Goniometric Measurement of Joint ROM
6Active ROM describes movement of a limb produced by the patient without
assistance and tests the ability of the muscles to produce full-range movement while
evaluating the health of the joint surfaces. Conversely, passive ROM is performed by the
evaluator moving the limb for the patient and tests the length of the connective tissue and
muscles resisting motion around the joint. ROM end-feels describe the quality of motion
felt by the practitioner and represent nonnal and pathological conditions as described in
Tables 1.1 and 1.2.
Table 1.1: Normal Range of Motion End-Feels (Starkey & Ryan, 2002)
End-Feel
Soft
Firm
Hard
Structure
Soft tissue approximation
Muscular stretch
Capsular/ligamentous stretch
Bone contacting bone
Example
Knee flexion - contact between
soft tissue of the posterior leg
and thigh
Hip flexion with the knee
extended - passive tension of
the hamstring muscles
Extension of the
metacarpophalangeal joints of
the fingers - tension in the
palmar capsule
Elbow extension - contact
between the olecranon process
of the ulna and the olecranon
fossa of the humerus
7Table 1.2: Abnormal Range of Motion End-Feels (Starkey & Ryan, 2002)
End-Feel
Soft
Firm
Hard
Empty
Description
Occurs sooner or later in the
ROM than is usual or occurs in
a joint that normally has a firm
or hard end-feel; feels boggy
Occurs sooner or later in the
ROM than is usual or occurs in
a joint that normally has a soft
or hard end-feel
Occurs sooner or later in the
ROM than is usual or occurs in
a joint that normally has a firm
or firm end-feel; feels like bony
block
Pain limits ability to test end-
feel; no resistance felt except
for patient's protective muscle
guarding
Example
Soft tissue edema
Increased muscular tone
Capsular, muscular or
ligamentous shortening
Osteophyte
Loose bodies in joint
Fracture
Joint dislocation
Joint inflammation
Joint dislocation
While bilateral comparisons have always been a necessary part of an injury
evaluation, asymmetries between two healthy limbs have more recently gained support in
the literature as a risk factor for injury (Knapik, Bauman, Jones, Harris, & Vaughan,
1991; Baumhauer, Alosa, Renstrom, Trevino, & Beynnon, 1995; Nadler et ai., 2001).
Specifically, Knapik, et al. (1991) showed a positive relationship between ROM
asymmetries and musculoskeletal injuries, where females with an asymmetry of greater
than fourteen percent knee flexion or hip extension ROM are reported to be 2.6 times
more likely to suffer an injury than those with more symmetrical strength.
8Muscle Strength
Muscle strength, defined as the external force that a muscle or group of muscles
can produce, is measured clinically using manually-resisted ROM testing through the
joint's full range. These tests can be used to assess a specific joint ROM, multiplanar
joint motion, or a muscle group. During manual resistance, the limb is stabilized proximal
to the joint to prevent compensatory motions while resistance is provided against the
distal joint segment. Strength is manually graded on a five-point scale as detailed in
Table 1.3.
Figure 1.5: Manual Muscle Testing of Knee Extension
Table 1.3: Grading System for Manual Muscle Tests (Starkey & Ryan, 2002)
9
Description
Normal
Good
Fair
Poor
Trace
None
5/5
4/5
3/5
2/5
1/5
0/5
Numerical Grade Clinical Finding
The patients can move the body part
through the full range of motion
against maximal pressure.
The patients can move the body part
through the full range of motion
against moderate pressure.
The patient can move the body part
through the full range of motion
against gravity, but not against
resistance
The patient can move the body part
through the full range of motion in a
gravity-eliminated position.
The patient cannot produce motion, but
a muscle contraction is palpable.
No contraction is palpable.
While manual tests are practical and convenient for a clinical setting, the
resistance that can be generated during these tests is significantly less when compared to
the forces that exist during sports participation and those generated during vigorous
activity in healthy populations. Traditionally, the one-repetition maximum (or multiple-
repetition maximum), defined as the greatest resistance that can be moved through the
full ROM in a controlled manner, has been the standard for strength assessment
(American College of Sports Medicine, 2010). Typical measures of upper body strength
10
include the bench press and overhead press, whereas measures for the lower body include
the leg press and squat (American College of Sports Medicine, 2010).
In a study measuring the isokinetic strength of 145 college-aged athletes,
participants with asymmetrical ankle strength between antagonist muscle groups had
higher rates of inversion ankle sprains (Baumhauer et aL, 1995). More specifically,
athletes with a greater amount of plantarflexion strength relative to dorsiflexion strength
were more likely to be affected. Similarly, those with greater inversion strength compared
to eversion strength were more likely to sustain inversion ankle sprain.
Following injury, a loss of strength would limit the dynamic stabilization of a
body segment. For example, since the peroneal muscle group acts as a stabilizer of the
lateral ankle during forced inversion, a lack of strength or delayed muscle contraction
during ankle inversion would make the ankle will be more susceptible to damage of the
passive support structures of the ankle, such as the ligaments and joint capsule
(Baurnhauer et a1., 1995).
t
'll' '\'~ '! III', I
'Ill/I,
\11 I
~
Figure 1.6: Action of the Peroneal Muscles in the Prevention of an Inversion Ankle
Sprain
11
Balance
Balance is the ability to maintain the center of gravity of an object within the base
of support with minimal postural sway. Success requires the concurrent sensory functions
ofthe vestibular system (inner ear), vision, and perception of pressure and proprioception
acting in concert with the motor system that controls muscle actions based upon this input
(Guskiewicz & Perrin, 1996). Evaluation of balance is typically measured using timed,
single leg stance with variations of sensory input disabled (Starkey & Ryan, 2002). For
example, the clinician may progress balance testing from single leg stance with eyes
open, to eyes closed (visual feedback disabled), to eyes closed with the head tilted back
(visual and vestibular feedback altered) (Guskiewicz & Perrin, 1996). Each of these steps
alters the sensory input for controlling balance and increases the demand on the other
systems of feedback. A deficit in balance originating from any of these systems would
quite naturally limit the ability to maintain stability during motion or static positioning.
Following injury, neurological feedback about joint position from the mechanoreceptors
in the region may become disrupted, resulting in reduced proprioception and balance, and
increased injury susceptibility (Guskiewicz & Perrin, 1996).
12
Figure 1.7: Balance Testing
Summary
Although the clinical measurements and tests listed above are commonly used
during a pre-participation physical examinations, the evidence is, at best, mixed as to
whether joint laxity, ROM, strength or balance are indeed significant risk factors for
injury occurrence. Joint laxity has been shown by several studies to be significant in both
males and females (Krivickas & Feinberg, 1996; Ostenberg & Roos, 2000; Soderman,
Alfredson, Pietila, & Werner, 2001), whereas other studies found no association
(Godshall, 1975; Baurnhauer et aI., 1995; Hopper, Hopper, & Elliott, 1995; Beynnon,
Renstrom, A10sa, Baumhauer, & Vacek, 2001). The same variability has been reported
for a lack ofjoint ROM (Milgram et aI., 1991; Barrett & Bilisko, 1995; Amason,
Gudmundsson, Dahl, & Johannsson, 1996; Krivickas & Feinberg, 1996; Wies1er, Hunter,
Martin, Curl, & Hoen, 1996; Twellaar, Verstappen, Huson, & van Meche1en, 1997;
13
Kaufinan, Brodine, Shaffer, Johnson, & Cullison, 1999; McKay et aI., 2001; Knapik et
aI., 2001; Beynnon et aI., 2001; Soderman et aI., 2001).
With regard to muscle strength, two studies (Baumhauer et aI., 1995; Soderman et
aI., 2001) found an association between strength differences between antagonist muscles
in the leg and thigh and injury, and one (Ekstrand & Gillquist, 1983) reported an
association between decreased strength of the quadriceps muscles and injury, findings
questioned by three other reports (Milgrom et aI., 1991; Ostenberg & Roos, 2000;
Beynnon et aI., 2001).
Finally, research regarding the association between balance and injury has also
displayed conflicting evidence with findings showing a positive relationship between
balance and injury (Tropp, Ekstrand, & Gillquist, 1984; McGuine, Greene, Best, &
Leverson, 2000), and others (Beynnon et aI., 2001; Soderman et aI., 2001) demonstrating
no relationship. It must be noted that such comparisons are difficult between each of
these areas of study as varying research designs associated with injury inclusion criteria,
populations tested, planes of motion and confounding variables such as gender (Zelisko,
Noble, & Porter, 1982; Gray et aI., 1985; Hickey, Fricker, & McDonald, 1997;
Myklebust, Maehlum, Holm, & Bahr, 1998; Gwinn, Wilckens, McDevitt, Ross, & Kao,
2000; Hewett, 2000; Powell & Barber-Foss, 2000; Ostenberg & Roos, 2000; Knapik et
aI., 2001), age (Chomiak, Junge, Peterson, & Dvorak, 2000) and skill level (Hopper et aI.,
1995; Chomiak et aI., 2000; Stevenson, Hamer, Finch, Elliot, & Kresnow, 2000; Hosea,
Carey, & Harrer, 2000) vary markedly across experiments. Collectively, all this work
plus a fundamental understanding of human movement strongly suggest that the design of
14
an effective pre-season screening tool will require much more research to carefully find
and evaluate variables that may predispose an athlete to injury.
Functional Testing
Whereas gross instability or significant functional impairments can be identified
with orthopedic testing, and these variables may be well associated with injury risk, such
tests still fail to adequately replicate the complex, integrated demands of mobility and
stability between multiple joints that are necessary to perform the basic movements
associated with sports. Specifically, the mobility that is required in sporting events is
defined as the ability to produce desired and efficacious motion and is impacted by the
architecture of the joints, soft tissue length, and the strength and neural control of the
surrounding muscle groups (Hickmans, 2007). Conversely, stability is the capacity to
resist motion through the passive and active restraints around the joints. The complexity
of this factor is further impacted as the passive elements include the joint capsule,
ligaments and the joint surface's architecture, whereas muscle strength, neural control of
the muscles, and balance are all determinants of the adequacy of the response of the
active elements. Stability can also be static as a function ofposture and balance, or
dynamic through the production of controlled movement (Hickmans, 2007). For example,
static stability can be described as maintaining an upright posture while sitting or
balancing on one leg, whereas dynamic stability can be thought of as something like
maintaining alignment of the hips, knees and feet while performing a squat.
In an attempt to create a pre-participation functional evaluation, Gray Cook and
Lee Burton (http://functionalmovement.com) developed the Functional Movement Screen
15
(FMS). This battery of tests was designed to simultaneously evaluate joint mobility and
stability through a series of seven movements. Although none are sport-specific
maneuvers, they do challenge the upper and lower extremities and the trunk in functional
tasks not unlike those that occur during athletic performance. As designed, the evaluation
is practical, as the desired movements can be tested within five to ten minutes, allowing
the clinician to quickly screen for deficiencies that may require more in-depth evaluation
and possible rehabilitation to reduce the risk of injury. The movements of the FMS
include the deep squat, hurdle step, in-line lunge, shoulder mobility, active straight leg
raise, trunk stability push up, and rotary stability. (Figures 1.8-1.14)
Deep Squat
The purpose of the deep squat is to assess:
Bilateral, dorsiflexion of the ankles, and flexion of the knees and hips while the
feet are in contact with the ground (closed-chain)
Extension of the thoracic spine
Stability of the lumbar spine
Flexion and abduction of the shoulders.
The deep squat mimics the "athletic" or "ready" position that is a part of many sporting
activities. It is also designed to assess symmetrical movement of the lower extremities
with concurrent stability of the trunk and upper extremities.
Figure 1.8: Deep Squat
16
17
Hurdle Step
The purpose of the hurdle step is to assess:
Mobility of the hips, knees and ankles of the step-limb
Stability of the hips, knees and ankles of the stance-limb
Hip extension of the stance-limb
Stability of the trunk
It evaluates stability in single leg stance along with mobility of the stepping limb, which
is incorporated in jumping and running motions.
Figure 1.9: Hurdle Step
In-line Lunge
The purpose of the in-line lunge is to assess:
Bilateral mobility and stability of the hips, knees and ankles
Stability of the trunk
The in-line lunge mimics a split stance position that is utilized for running motions. It
also challenges pelvic stability with concurrent hip flexion and extension of the
contralateral limbs.
Figure 1.10: In-line Lunge
18
19
Shoulder Mobility
The purpose of the shoulder mobility test is to assess:
Bilateral shoulder ROM, combining internal rotation, adduction and extension
with one shoulder, and external rotation, abduction and flexion of the contralateral
shoulder
Scapular mobility
Thoracic spine extension
These combined motions of the shoulder in the transverse, frontal and sagittal planes
allow the clinician to quickly evaluate shoulder mobility. Pain during this test may
indicate impingement syndrome of the rotator cuff.
Figure 1.11: Shoulder Mobility
Active Straight Leg Raise
The purpose of the active straight leg raise is to assess:
Mobility of the hamstring, gastrocnemius and soleus muscles
Active extension of the contralateral limb
Stability of the pelvis
This movement tests mobility of the moving limb with concurrent stability and active
extension of the stationary limb. It also challenges pelvic stability with concurrent hip
flexion and extension of the contralateral limbs.
Figure 1.12: Active Straight Leg Raise
20
21
Trunk Stability Push-Up
The purpose of the trunk stability push-up is to assess:
Trunk stability while a symmetrical, closed-chain upper extremity motion is
performed
Upper extremity strength
The trunk stability push-up challenges core strength, specifically at the lumbar spine and
hips, and upper extremity strength during a one-repetition push-up.
Figure 1.13: Trunk Stability Push-up
22
Rotary Stability
The purpose of the rotary stability test is to assess:
Multiplanar trunk stability in a quadruped position with combined upper and
lower extremity motion
The rotary stability test stresses the core musculature in maintaining alignment of the
trunk while undergoing changes to support contacts with the ground and moving the
upper and lower extremities. This test is designed to mimic complex motions associated
with sports that demand stability during movement.
Figure 1.14: Rotary Stability
23
Study Description
There is little doubt that the amount of injuries that are sustained as the result of
high school athletics, as well as the associated health care costs, and the effects on the
lives of young athletes, warrants definitive research into the risk factors associated with
participation. Despite the fact that biomechanical measures such as joint laxity, ROM,
strength and balance have been identified as variables possibly associated with injury,
these traditional measures alone may be inadequate to be valuable in mitigating these
issues. Therefore, in this study, the potential for the FMS to use those variables in
combination may permit evaluation of functional mobility and stability and ultimately
yield an improved capacity to efficiently and correctly identify athletes that are at risk of
becoming injured.
Specific Aim #1
Since the FMS requires subjective evaluation by trained observers, it was
necessary to establish inter-rater reliability and intra-rater reliability for the team of
evaluators. Inter-rater reliability is defined as the degree of agreement between raters
while intra-rater reliability is the degree of agreement within a single rater across multiple
trials. Strong agreement between raters and within raters is obligatory to confidently
establish that the tests can be administered with reproducible results, an important factor
if the screen is to be effectively used clinically. In this study, each rater was required to
score athletes on the FMS individually, so it was essential to first verify that variability
between raters was minimal. To-date, no such research has been published regarding the
24
inter-rater and intra-rater reliability of scoring the FMS, but in the absence of these
assumptions, the practicality of the FMS for general use is severely compromised.
Specific Aim #2
Once adequate inter-rater and intra-rater reliability are achieved, it becomes
possible to evaluate the most critical question posed in the study which is, can a low
score threshold be established on the FMS that could serve as a reliable predictor of
injury for high school basketball athletes throughout a season? Because the FMS is
intended to evaluate and address deficiencies in the mobility and stability of an athlete
that might be linked to injury, the only injuries that were included in this study were non-
contact injuries. They included any tissue damage that resulted from participation in a
school-sanctioned practice or game. Chronic, overuse injuries that presumably resulted
from the accumulation of repeated forces to a joint or body segment that led to the
insidious onset of symptoms, and acute injuries that were the result of a single traumatic
force to a body region were both included. Those disqualified occurred from contact with
the ground (other than the foot contacting the ground), the ball, another player or any
other object in the gymnasium as they would have introduced elements that were
irrelevant to that which would have been expected to be associated with what could be
evaluated by the FMS.
Kiesel, Plisky and Voight (2007) have shown in professional football athletes that
a score less than fifteen out of twenty-one on the FMS predisposes an athlete to an
eleven-fold increased chance of injury compared to those individuals that score fifteen or
higher on the tests. This study included contact injuries and was successful at identifying
25
athletes that were at risk of becoming injured. However, their inclusion of contact injuries
may have skewed the results and potentially weakens the impact of their results on the
question asked here. For example, an individual could score perfectly on the FMS but get
blind-side tackled and sustain an impact injury. On the other hand, it could be argued that
a high level of total-body stability and mobility could allow an individual to better avoid
contact or tolerate a greater level of contact without sustaining an injury. However, for
the purposes of this research, we chose to exclude contact injuries in attempt to minimize
the potential role of unpreventable contact injuries on our results. Furthermore, Kiesel's
study (2007) with professional football players only included those injuries that resulted
in three or more weeks of missed participation due to injury. Thus, minor injuries that
could indeed alter individual performance without any absence from games or practices
are, by default, excluded. Lastly, although the large numbers and high incidence of injury
that occur in football make it a reasonable sport to evaluate injuries, accurately
differentiating contact versus non-contact injuries in this sport can be quite difficult due
to the nature of the sport. Also, results from studying professional athletes may also limit
the external validity for applying the outcomes to lower skilled or younger athletes, of
which the numbers are much greater nationally.
Although the first published paper on the application of the FMS to injury
prediction shows promise, there still remains little data to support the widely adopted,
clinical-use of a cutoff score of 14 out of 21 to predict injury. This study was designed to
further investigate the relationship between performance on the FMS and risk for non-
contact injuries in high school basketball athletes.
26
Hypotheses
It was hypothesized that the inter-rater and intra-rater reliability would exceed
0.80, which would support the use of multiple raters and multiple sessions for screening a
large group of athletes with the FMS. Secondly, it was hypothesized that high school
basketball athletes that scored below a 14 out of 21 on the FMS would be more likely to
sustain a non-contact injury during a basketball season. The potential that the players'
gender, playing level, individual movement scores, net FMS score, and movement
performance asymmetries was also evaluated as variables that might be useful in injury
prediction.
27
CHAPTER II
METHODS
The Functional Movement Screen
The Functional Movement Screen (FMS) consists of seven tests. Each test is
scored on a four-point scale, with three indicating perfect performance, two, minor
deficits or perfect performance with modifications, and one, the inability to perform the
movement. In all tests, the prevalence of pain during movement corresponds to a zero.
Three attempts are performed for each test with the highest score recorded. If a score of
three is achieved at any time, no further repetitions are required. For the tests that are
divided for right and left, the composite score (lower of the two net scores) is used for the
final score. The maximum score is 21.
Deep Squat
The deep squat begins by having the subject place his/her feet shoulder width
apart, aligned in the sagittal plane while holding a dowel with both hands. Contact is then
made between the dowel and the top of the head and the width of their hands on the
dowel is adjusted so that the elbows are flexed to 90 degrees. The subject then presses the
dowel overhead by extending the elbows, and a full squat is performed. A score of three
is recorded if the individual is able to squat to a depth that achieves 90 degrees of knee
flexion while maintaining:
28
heel contact with the floor
knees aligned over feet
torso and arms parallel with tibia or toward vertical
dowel aligned over feet and symmetrical in the frontal plane
If the subject is not able to perform the deep squat to satisfy these criteria within three
attempts, a 2-inch base is placed under their heels and the task is repeated. If the fore-
mentioned criteria are subsequently achieved with a heel lift, a score of two is awarded.
When one or more of the criteria are not achieved despite the modification, a score of one
is earned.
Hurdle Step
The hurdle step is tested by positioning the subject with feet together and toes
touching the base of a hurdle. The hurdle should be adjusted to the height of the tibial
tuberosity, defined as the distance from the ground to the anatomical marker. The dowel
is held with both hands resting on the shoulders behind the neck. The individual then
executes a step over the hurdle, touching their heel to the ground, and then returning to
the starting position. Eye contact must be maintained at the horizon and not down to the
hurdle. After a maximum of three attempts, a score of three is recorded if the individual is
able to clear the hurdle while maintaining the hips, knees, and feet aligned in the sagittal
plane, dowel aligned in the frontal plane, and minimal flexion of the spine. If any of the
above-mentioned criteria are not achieved, but the hurdle is cleared and balance is
maintained, a score of two is awarded, whereas contact with the hurdle or loss of balance
earns a one.
29
In-line Lunge
The in-line lunge is tested by first measuring the height of the tibial tuberosity
(defined above) and then placing one foot at the end ofa 2-inch by 6-inch board. The heel
of the contralateral foot is placed on the board at the distance corresponding to the height
of the tibial tuberosity (distance measured from the toe of the back foot to the heel of the
front foot). The subject then holds a dowel along the spine, with the hand ipsilateral to the
back foot securing the rod at the cervical spine and the contralateral hand securing the rod
at the lumbar spine. It must be in contact with the head, thoracic spine, and sacrum. The
subject then lowers the back knee until it touches the board and immediately returns to
the starting position. After a maximum of three attempts, a score of three is recorded if
the individual is able to contact the board with their knee while maintaining sagittal
alignment of both feet, upright and sagittal alignment of the trunk, and three points of
contact with the dowel throughout the movement. If any of the above-mentioned criteria
are not achieved, but the lunge is attempted and balance is maintained, a score of two is
recorded. A one is earned if balance is lost or the starting position cannot be obtained.
Shoulder Mobility
Measurements of shoulder mobility are evaluated by first measuring the length of
the subject's hand from the distal palmar wrist crease to the tip of the third digit. While
standing in an upright posture, both hands are clinched in a fist and in one smooth
motion, they are placed and maintained on the back as close as possible. This is achieved
by maximally internally rotating, adducting and extending one shoulder while maximally
externally rotating, adducting and flexing the contralateral shoulder. The evaluator
30
measures the distance between the two closest bony prominences of the fists. If after a
maximum of three attempts, the distance measured is within one hand-length, a score of
three is recorded, a two if the measurement is within one and one-half hand-lengths, and
a one is if the measurement is greater than one and one-halfhand-Iengths.
Active Straight Leg Raise
Evaluation of the active straight leg raise begins in the supine position with head
on the floor, arms by the side, palms up, both feet together, and ankles dorsiflexed to 90
degrees. A 2-inch by 6-inch board is placed under both knees, and the mid-point between
the subject's anterior superior iliac spine and mid-point of the patella is identified. A
dowel is placed at this location, perpendicular to the ground. The test leg is then lifted in
the sagittal plane while maintaining an extended knee and dorsiflexed foot position. A
valid test requires that the contralateral limb does not rotate and the knee must remain in
contact with the board. After maximal hip flexion is achieved, the position of the medial
malleolus of the test limb is measured in relationship to the contralateral knee and dowel.
Movement of the medial malleolus past the dowel towards the subject's head earns a
three, and a two corresponds with the medial malleolus passing the contralateral knee but
not the dowel. If the medial malleolus fails to pass the contralateral knee, the score is
reduced to a one.
Trunk Stability Push-Up
The trunk stability push-up commences in the prone position with feet together
and the hands placed shoulder width apart. Male subjects start with the hands in-line with
the top of the head and continue by lowering them to the chin if necessary, whereas
31
female subjects commence with the hands at chin level and lower them to the clavicle if
necessary. One push-up is executed with the body elevating as a unit and no lag in the
lumbar spine, followed by returning to the starting position while maintaining the same
movement criteria and hand positioning. Males earn a three if the criteria are maintained
with hands at head-level, and females with hands at chin-level. A score of two
corresponds with successful completion of the criteria while hands are kept at chin-level
(males), and clavicle-level (females), and a score of one is consistent with failure to
maintain the second position.
Rotary Stability
The test begins in the quadruped position with shoulders and hips flexed to 90
degrees, and a board is placed between the hands and knees of the subject. The hands
(palmar aspect) and knees are in contact with the ground, directly below the shoulders
and hips, respectively, with the ankles dorsiflexed. One hip and knee is extended while
flexing the shoulder and extending the elbow of the ipsilateral limb. While maintaining
the elevated appendages in-line with the board, and the trunk in-plane with the board, an
attempt is made to contact the ipsilateral knee and elbow over it, followed by a return to
the extended positions. Completion of the task within the required criteria earns a score
of three. In case of failure, the task is repeated in the diagonal with opposite arm and leg
raised. A score of two corresponds to completion of the diagonal task within the required
criteria, whereas the subject earns a one ifhe/she is not able to complete the diagonal task
within the required criteria.
32
Research Studies
Reliability Studies
Subjects
The subjects for this study (N = 8) were Certified Athletic Trainers who
volunteered their participation as evaluators in the injury prediction study. They were
trained to execute and rate the FMS by the primary investigator who had previously been
instructed at an FMS seminar (Perform Better; Los Angeles, CA). The process began
with an introduction DVD (http://functionalmovement.com) and grading rubric
(Appendix 1) given to the evaluators to gain exposure to the principles of the FMS.
Subsequently, a group training session of two hours that included a review of the
individual tests in the FMS (i.e. proper set up, execution, and scoring criteria of the tests)
was used. After completion of this session, each evaluator carried out a complete FMS
while the primary investigator provided corrective feedback on the execution of the tests.
A sample of fifteen college-aged subjects (eleven males, four females; age: 24.7
years; height: 165.87cm; weight: 73.07kg) volunteered to perform the movements of the
FMS while being video taped from a frontal and sagittal view. Prior to participation, each
subject signed a consent form and agreement for videotaping approved by the Human
Subjects Compliance Committee at the University of Oregon. (Appendices 2 & 3) Each
participant was instructed in the performance of all test movements by the primary
investigator to ensure a standardized explanation. The videos were compiled by the
primary investigator and one repetition of each movement was presented to the
evaluators. The evaluators scored the video performances on a written score sheet
33
(Appendix 4) two times with a two-week interval between sessions. After the sessions,
their scores were given to the primary investigator with whom they were kept
confidential and ultimately transcribed into Excel and transferred to SPSS for data
analysis.
Data Analysis
Inter-rater Reliability Study
Inter-rater reliability for the group of eight subjects was tested using an inter-item
correlation matrix, in which the scores of each evaluator are compared to the scores of
each other evaluator. The dependent variables (8) included each movement of the FMS
(deep squat, hurdle step, in-line lunge, shoulder mobility, active straight leg raise, trunk
stability push up, and rotary stability) and the FMS score. For movements divided for
right and left, the composite score (lower of the two net scores, as used in the FMS total)
was used in determining inter-rater reliability for the movement. Scores for each test were
analyzed from the first video scoring session with the evaluator serving as the
independent variable (8). Statistical analyses were run using SPSS 13 Grad Pack for Mac
OS X. With this type of analysis, the scores of the evaluator are compared individually to
each of the others (Ex. 1-2, 1-3, 1-4,2-3,2-4, etc.). In order to evaluate the reliability for
each test, the median reliability coefficient was calculated for each Athletic Trainer and
the median of the medians was calculated for the entire group. The median reliability
coefficient was defined as the score that separated the higher and lower halves for the
inter-rater reliability pairings. (Table 2.1) This value was used instead of the mean value
34
in order to protect the outcomes from skewed results or outliers. Furthermore, the median
of the medians (middle score of all evaluators' inter-rater reliability) was used to report
the performance for the group of evaluators. This allowed the primary investigator to
evaluate the degree of agreement across individuals and the group. Median reliability
coefficients were calculated using Microsoft Excel 2008 for Mac.
Table 2.1: Example Calculation for Median Reliability Coefficient
Pairings
Score
1-2
0.78
1-3
0.83
1-4
0.85
1-5
0.88
1-6
0.90
1-7
0.90
1-8
0.94
Note. The reliability coefficient of 0.88 is the median score because it splits the inter-rater
reliability scores into high and low halves
Intra-rater Reliability Study
Intra-rater reliability for each subject was tested using an intrac1ass reliability
coefficient, which evaluates the degree of consistency for each rater scoring the FMS
over time. The dependent variables (8) were each movement of the FMS (deep squat,
hurdle step, in-line lunge, shoulder mobility, active straight leg raise, trunk stability push
up, and rotary stability) and the FMS score. For movements divided for right and left, the
composite score (lower of the two net scores, as used in the FMS total) was used in
determining intra-rater reliability for the movement. The independent variable (1) was
time (session one and two). Intra-rater reliability for each evaluator was run using SPSS
13 Grad Pack for Mac OS X. The median reliability coefficient (middle score for the
35
group) was then calculated for each movement and the FMS scores (Microsoft Excel
2008 for Mac). This value permitted the primary investigator to evaluate the level of
agreement across individuals for each test without the scores being influenced by skewed
results or outliers.
Interpretation ofData
Generally, there are no universally applicable standards as to how high scores
must be to constitute acceptable reliability, and the use of the scores typically dictates the
minimum level that must be met (Shrout & Fleiss, 1979; Eliasziw, Young, Woodbury, &
Fryday-Field, 1994; Hayen, Dennis, & Finch, 2007). For example, reliability scores for
screening purposes are commonly considered appropriate at much lower levels than for
outcomes that will directly influence the choice of treatment for a patient (Downing,
2004; Hayen et aI., 2007). However, it is important to carefully pre-determine the
acceptable level in order to have confidence that a group is consistent in scoring the
required tests. In this study, a cutoff of 0.80 on the FMS score was used to exclude
evaluators from the injury prediction study. Scoring above 0.80 was considered
acceptable reliability, whereas scores below 0.80 resulted in exclusion from the study
because the evaluator's performance may not be adequate for screening purposes relative
to the needs of the study. As stated above, these values correspond to the level of
consistency for the scores that were recorded, as compared between and within each
evaluator.
36
Injury Prediction Study
Subjects
The subjects for this study (N = 112) included 52 male and 60 female basketball
players from participating high schools in the 4J (Eugene, OR) and Springfield
(Springfield, OR) school districts. Prior to participation, each subject or, for minors, their
legal guardian signed a consent form approved by the Human Subjects Compliance
Committee at the University of Oregon. (Appendix 5) Fifteen of the subjects who
participated were at the freshman level, 37 at the junior varsity level, and 60 at the varsity
level. None of the subjects had internal or external appliances in the extremities (i.e.:
rods, joint replacement, prosthetic) or an injury status that prevented full participation in·
school-sanctioned basketball practices or games.
Each subject completed a demographic information form that included their
name, age, gender, year-in-school, previous musculoskeletal injuries, school name, team
level, years-of-participation in basketball, projected player status (starter, second-string),
and projected position. (Appendix 6) All information was self-reported and collected
prior to the screening process.
Functional Movement Screen Rating
The FMS was administered by a team of Certified Athletic Trainers that had
completed the inter-rater and intra-rater reliability study and scored 0.80 or greater on the
FMS score (see Results section). Each subject was screened prior to a school-sanctioned
basketball practice or at a mutually agreeable time outside of practice. No testing was
administered immediately following a basketball practice, competition, or strength and
37
conditioning session to limit the impact of potentially confounding variables such as
muscular fatigue, lack of flexibility or soreness on the testing. All scoring was conducted
live by individual evaluators and transcribed on a written score sheet (Appendix 4).
Asymmetries were recorded as the difference between the scores recorded for right and
left movements (Ex: Hurdle step right = 1, hurdle step left = 3. Asymmetry recorded),
whereas weighted asymmetries were designated as the total difference between those
measurements (Ex: Hurdle step right = 1, hurdle step left = 3. Weighted asymmetry of2
recorded). The latter were calculated to determine if the degree to which a movement was
asymmetrical was a better predictor than unweighted asymmetries. FMS and asymmetry
scores were transcribed into Excel and transferred into SPSS for data analysis.
Injury Data
Injury and time-loss data were collected throughout the duration of the 2008-2009
high school basketball season by Certified Athletic Trainers employed by the
participating high schools using a Daily Injury Report Form. (Appendix 7) Classification
was made by:
Date of injury
Date of retum-to-full-participation
Injured body part
Injured side
Type of injury
Diagnosis
Occurrence of previous injury
Cause of injury (overuse or acute)
Time of injury (practice or competition)
Non-contact vs. contact
Type of contact
38
Injuries utilized in this study were limited to those classified as neuromusculoskeletal
impairments reported to and/or recognized by the school's coaching staff or Certified
Athletic Trainer. Injuries were excluded if they did not occur during a school-sanctioned
practice or game, were unrelated to training or competition, or were caused by contact
with a ball, another player, the floor, or any combination of the above. The latter were
recorded nonetheless so that the primary investigator could evaluate the type of contact
and determine proper use of the exclusion criteria. Injury data collection began on the day
of the FMS testing and was concluded on the last day of competition for each team.
Data Analysis
A one-way analysis of variance was performed to verify that no significant
differences existed between boys' and girls' FMS scores so that data from both groups
could be pooled for injury prediction analyses. The dependent variable (1) was the FMS
score with gender serving as the independent variable (2). Furthemlore, to rule out type
of injury (acute and chronic) as a potentially confounding variable, the FMS total scores
for those sustaining acute versus chronic injuries were compared using a one-way
analysis of variance to indentify ifthere were significant differences between the groups.
No differences would indicate that type of injury was not a confounding variable in this
study and the data could be pooled.
Previous research (Kiesel, Plisky, & Voight, 2007) utilized a receiver-operator
characteristic (ROC) curve to determine a cutoff score on the FMS that maximized
sensitivity (the percentage of injured athletes that were predicted by the FMS) and
39
specificity (the percentage of uninjured athletes that were predicted by the FMS). This
cutoff was then used with the same data set in order to evaluate the efficacy of the cutoff,
despite the fact that this practice often overestimates the findings as compared to using
the cutoff with a unique, prospective sample of subjects (Kiesel et aI., 2007). Therefore,
the previously identified score of fourteen was utilized for our study.
Chi-square determined the predictive capacity of the FMS score cutoff of 14 out
of2l to identify basketball athletes that became injured. The dependent variable (1) for
this analysis was injury (injured, not injured), whereas the independent variable (1) was
FMS score (14 and below, 15 and above). The subjects were divided into a 2x2
contingency table for the chi-square test with column one assigned to subjects that scored
14 or above on the FMS and column two containing those subjects with lower scores.
Row one included subjects that were not injured during the season and row two, those
that were. If the FMS was able to perfectly identify at-risk subjects, all participants would
fall into two cells: the cell for row one/column one would include all subjects that scored
above 14 or greater and, therefore, did not become injured, whereas row two/column two
would contain the remainder of the subjects that scored lower and were injured.
Logistic regression isolated demographic variables and components of the FMS in
order to detemline their predictive capacity for identifying injury-prone athletes. This tool
identifies a combination of independent variables that best predict membership in a
particular group, as measured by a categorical dependent variable (Mertler & Vannatta,
2005). The categorical dependent variable was injury (injured, not injured) measured
against the independent variables (11) gender, level (Freshman, Junior Varsity, Varsity)
40
each movement of the FMS (deep squat, hurdle step, in-line lunge, shoulder mobility,
active straight leg raise, trunk stability push up, and rotary stability), the number of
asymmetries, and weighted asymmetries. The goal was to determine if the independent
variables listed above could predict if an athlete was vulnerable to injury, either
individually or grouped.
Although research indicates that previous injury is a significant risk factor for
repeat injmy (Emery et aI., 2005), the methods for collecting this type of data relied upon
the memory of the subjects and/or their parents to accurately report the athlete's history
of injury. After reviewing the data, the replies were incomplete and lacking details. For
this reason, previous injury data was not included in the logistic regression model.
However, we were able to obtain complete data regarding the history of those athletes
that became injured during the season, as reported by the school's Athletic Trainer. This
data was analyzed using a one-way analysis of variance to determine if there were
significant differences between the FMS total scores of the subjects that sustained new
versus recurring injuries.
41
CHAPTER III
RESULTS
Reliability Studies
Table 3.1 displays the minimum and maximum scores for the group of 15 subjects
performing the FMS, as rated by the Athletic Trainers participating in the reliability
studies.
Table 3.1: Values for Functional Movement Screen of Reliability Study Participants
Source
Deep Squat
Hurdle Step
In-line Lunge
Shoulder Mobility
Active Straight
Leg Raise
Trunk Stability
Push Up
Rotary Stability
Score
Note. n = 15
mm
o
1
o
o
1
1
o
10
max
3
3
3
3
3
3
2
18
42
Inter-rater Reliability Study
The median inter-rater reliability coefficient was considered acceptable (greater
than 0.80) for all of the individual tests except for the rotary stability test (0.73).
However, our evaluation of FMS total scores, which included the latter test, yielded
acceptable reliability, indicating that, for the screening as a whole, evaluators are able
produce FMS scores similarly to their peers. (Table 3.2) Since Kiesel et al. (2007)
proposed that the composite FMS score is the primary indicator of injury risk, its
reliability score was used to determine the cutoff for exclusion as a rater in this study. All
evaluators scored above 0.80 and were included. However, the results for the rotary
stability test should be considered with caution when interpreting results derived from
that test individually.
43
Table 3.2: Inter-rater Reliability Results
Source median ofmIll max
medians
Deep Squat 0.84 0.91 0.90
Hurdle Step 0.75 0.90 0.84
In-line Lunge 0.86 0.95 0.88
Shoulder Mobility 0.97 0.97 0.97
Active Straight
Leg Raise 0.85 0.92 0.90
Trunk Stability
Push Up 0.77 0.87 0.84
Rotary Stability 0.66 0.81 0.73
FMS Total Score 0.89 0.92 0.90
Note. n = 8
--------------------------
44
Intra-rater Reliability Study
The median intra-rater reliability coefficient was considered acceptable for all of
the tests and FMS score, indicating that evaluators are able to score the FMS consistently
over time. (Table 3.3)
Table 3.3: Intra-rater Reliability Results
Source
Deep Squat
Hurdle Step
In-line Lunge
Shoulder Mobility
Active Straight
Leg Raise
Trunk Stability
Push Up
Rotary Stability
FMS Total Score
Note. n = 8
min
0.75
0.64
0.86
0.93
0.69
0.71
0.50
0.86
max
1.00
1.00
1.00
1.00
1.00
1.00
1.00
0.95
median
0.92
0.85
0.92
0.97
0.88
0.88
0.82
0.88
45
Injury Prediction Study
Demographics
The subjects for this study (N = 112) included 52 male and 60 female basketball
players from participating high schools in the 4J (Eugene, OR) and Springfield
(Springfield, OR) school districts. Fifteen of the subjects participated at the freshman
level, 37 at the junior varsity level, and 60 at the varsity level. The minimum, maximum,
mean and standard deviation for each of the FMS movements and final scores can be
found in Tables 3.4 and 3.5, which are sub-divided for boys and girls. A comparison of
their performances can be found in Table 3.6 and Figures 3.1 and 3.2, which show that
the effect of gender on FMS score was not significant, F(1, 110) = 0.29, p > .05. This
indicates that there is no significant difference in the FMS score between boys and girls
and consequently, the injury results for this study can be pooled in a single group.
46
Table 3.4: Values for Functional Movement Screen Tests and Final Score for Boys
Source mm max M SD
Deep Squat 1 3 1.62 0.56
Hurdle St~p 1 3 1.83 0.43
In-line Lunge 1 3 2.21 0.61
Shoulder Mobility 1 3 2.62 0.72
Active Straight
Leg Raise 1 3 2.48 0.64
Trunk Stability
Push Up 1 3 1.83 0.79
Rotary Stability 1 3 1.88 0.65
Total Score 10 20 14.46 2.13
Note. n = 52
47
Table 3.5: Values for Functional Movement Screen Tests and Final Score for Girls
Source mm max M SD
Deep Squat 1 3 1.68 0.57
Hurdle Step 1 3 1.88 0.45
In-line Lunge 1 3 2.23 0.59
Shoulder Mobility 2 3 2.90 0.30
Active Straight
Leg Raise 1 3 2.67 0.57
Trunk Stability
Push Up 1 3 1.43 0.65
Rotary Stability 1 3 1.87 0.57
Total Score 10 20 14.67 1.91
Note. n = 60
Table 3.6: One-Way Analysis of Variance Summary Table for the Effects of Gender on
FMS Score
Source
Gender
Error
Total
df
1
110
111
SS
1.17
446.26
447.43
MS
1.17
4.06
F
0.29
48
a Boy;;
GirlsII
Deep Squat Hurdle Slep In-Ime Lunge Shoulder AClive Straight Trunk Stahility RNary Stahility
\lnhility Leg Raise I'ush-up
Figure 3.1: Comparison ofFMS Movement Scores for Boys and Girls
21
18
15
!.I 12~
c)!f ,,; q
6
3
0
Boys GIrls
Figure 3.2: Comparison ofFMS Scores for Boys and Girls
49
Five of the seven FMS tests are divided for right and left, and provide potential
areas for movement asymmetries (defined as a difference in test scores between the right
and left). The average number of asymmetries for boys was one, whereas girls had less
than one. (Tables 3.7 and 3.8) There was little difference between the means for
asymmetries and weighted asymmetries (see Methods for definition), and therefore, these
calculations may be redundant for future studies.
Table 3.7: Functional Movement Screen Asymmetries for Boys
Source
Asymmetry
Weighted
Asymmetry
Note. n = 52
min
o
o
max
3
3
M
1.00
1.10
SD
0.77
0.85
Table 3.8: Functional Movement Screen Asymmetries for Girls
Source
Asymmetry
Weighted
Asymmetry
Note. n = 60
mIll
o
o
max
3
3
M
0.83
0.82
SD
0.83
0.83
50
Injuries
Thirty-two non-contact neuromusculoskeletal injuries were reported during the
season. The majority of injuries affected the lower extremities (27), whereas only two
affected the upper extremity and three impacted the trunk. (Table 3.9) The volume of
injuries to the lower extremity compared to the upper extremities and trunk was greater
than has been reported by Rechel et aI. (2008), studying a subset of high school
basketball players. However, that study included contact injuries, which would have
likely increased the number of reported injuries to the upper extremity and trunk.
Twenty-six of the injuries occurred during a practice session (81 %) and six (19%) during
games. These findings are significantly skewed towards practice compared to previous
research (Rechel et aI., 2008), which showed that there is only a difference of2.6 to 10.6
percent between practice and competition, with more injuries occurring during games for
girls. Thirteen of these injuries were sustained by boys, nine of which were chronic in
nature and four could be attributed to acute mechanisms. Nineteen injuries were sustained
by girls, thirteen of which were chronic and six acute. The total amount of time lost from
these injuries was 61 days for both boys and girls, totaling 122 days of time-loss for the
study.
51
Table 3.9: Number ofInjuries by Region
Region
Foot & Ankle
Leg
Knee
Thigh & Groin
Hip
Low back
Shoulder
Forearm
Number
6
8
5
6
2
3
1
1
The mean FMS total scores for those sustaining acute or chronic injuries were
13.90 (SD = 2.03) and 14.64 (SD = 1.50), respectively. Results for the comparison of the
FMS total scores between athletes sustaining acute versus chronic injuries can be found
in Table 3.10, which shows that the effect of injury type on FMS score was not
significant, F(l, 30) = 1.33,p > .05. This indicates that there is no significant difference
between the FMS scores of those suffering injuries as the result of acute versus chronic
mechanisms, and therefore, this data was pooled into a single group.
52
Table 3.10: One-Way Analysis of Variance Summary Table for the Effects ofInjury
Type on FMS Score
Source
Injury Type
Error
Total
eft
1
30
31
SS
3.73
83.99
87.72
MS
3.73
2.80
F
1.33
Injury Prediction with CutoffScore of14
Utilizing the cutoff score below 14 to categorize athletes, 76 subjects scored
above the benchmark and 36 scored below. Data analysis revealed that the commonly-
used FMS cutoff score of less than 14 out of 21 was not significantly related to the
likelihood of sustaining an injury, X2 (1, n = 112) = O.03,p > .50. (Table 3.11) In fact, the
FMS displayed very poor ability to predict at-risk athletes as a greater percentage of those
scoring fourteen or higher became injured (24%) compared to those below the cutoff
(22%). This contradicts the assertation of a cutoff identifying vulnerable athletes because
a greater percentage of athletes scoring above 14 became injured compared with the
group that was expected to have a greater risk of injury. Furthermore, analyses of higher
and lower cutoff scores indicated that no other cutoff could accurately predict an
increased risk of injury. (Tables 3.12-3.15)
Table 3.11: Chi-square Results with Cutoff of 14
FMS Score
Injury
< 14 14+
No Injury 28 58
One or more injuries 8 18
Note. n = 112
Table 3.12: Chi-square Results with Cutoff of 12
FMS Score
Injury
< 12 12+
No Injury 4 82
One or more injuries 1 25
Note. X2 (1, n = 112) = 0.30,p > .50
Table 3.13: Chi-square Results with Cutoff of 13
FMS Score
Injury
< 13 13+
No Injury 12 74
One or more injuries 3 23
Note. X2 (1, n = 112) = 0.10, p > .50
53
Table 3.14: Chi-square Results with Cutoff of 15
FMS Score
Injury
< 15 15+
No Injury 41 45
One or more injuries 14 12
Note. X2 (1, n = 112) = 0.30, p > .50
Table 3.15: Chi-square Results with Cutoff of 16
FMS Score
Injury
< 16 16+
No Injury 56 30
One or more injuries 20 6
Note. X2 (1, n = 112) = 1.28,p > .10
54
55
Injury Prediction witlt Logistic Regression
In order to detennine if a subset of demographic variables, FMS movements and
asymmetrical movements could more accurately predict injury risk, logistic regression
was used (see Methods, pg. 39). Data analysis revealed that none of the predictors were
significant, either in isolation (Table 3.16) or in combination (df= 11,p>.10).
Table 3.16: Logistic Regression for Predictors in Isolation
Predictors ~ p
Gender 0.55 1.74
Level
-0.91 0.40
Deep Squat 0.39 1.48
Hurdle Step 0.56 1.75
In-line Lunge
-0.81 0.45
Shoulder Mobility
-0.72 0.49
Active Straight Leg Raise
-0.14 0.87
Trunk Stability Push Up 0.10 1.11
Rotary Stability
-0.29 0.75
Asymmetries 0.11 1.12
Weighted Asymmetries
-0.51 0.60
Note. df= 11
56
Previous Injury
The prevalence of injury was equally divided between new injuries and
recurrences with mean FMS total scores of 14.21 (SD = 1.81) and 14.42 (SD = 1.62) for
each group, respectively. Results for the comparison of the FMS total scores between
athletes sustaining a new or recurring injury can be found in Table 3.17, which shows
that the effect of previous injury on FMS score was not significant, F(l, 24) = 0.89,p >
.05. This indicates that there is no significant difference for the FMS scores of newly
versus previously injured athletes.
Table 3.17: One-Way Analysis of Variance Summary Table for the Effects of Previous
Injury on FMS Score
Source
Previous Injury
Error
Total
df
24
25
SS
0.27
71.27
71.54
MS
0.27
2.97
F
0.89
57
CHAPTER IV
DISCUSSION
Statement of the Problem
It has long been the responsibility of the sports medicine team of physicians,
athletic trainers, physical therapists, and strength and conditioning coaches to ensure the
safety of those participating in sport. Their involvement in the well-being of athletes
varies based upon the setting, but each of these individuals has an integral part in the
process. The current model of health care and advances in medical science are strongest
in their response to injury. That is, once an injury occurs, there are many theories and
techniques available that contribute to its accurate diagnosis and management. With
respect to musculoskeletal pathology, there are often efforts to reduce the likelihood of
reinjury by correcting biomechanical deficiencies in range of motion, strength or
movement patterns. However, there is scant information available as to how to intervene
prior to the occurrence of an injury so that the event is prevented before it ever happens.
The number of athletes that participate in sport, ranging from recreational
involvement to high-level competition, and the wide spectrum of injuries that occur,
suggests that any attempt to develop prevention protocols would be daunting. In addition,
practicality would require a streamlined and standardized approach for their
implementation. Within organized athletics, this effort could be simplified because many
58
schools, clubs and teams employ sports medicine professionals who work specifically
with the athletes to monitor the incidence of injury and manage their rehabilitation.
Consequently, they are well placed to develop, implement and test promising injury
prevention models. At the elite level, preventing injuries is economically desirable as it
helps to ensure that highly paid athletes remain active for competition. Despite only
anecdotal evidence and limited research supporting various methods of identifying
athletes that are at risk of becoming injured, some institutions and teams have already
hired personnel that are specifically charged with the duty of executing injury prevention
programs. Although this forward thinking attitude towards injury prevention warrants
praise, the methods such groups employ to design prevention programs are rarely
grounded on strong, evidence-based practice. This gap between evidence and practice is
what prompted the research described here.
The focus of the study was evaluation of the reliability and validity of the
Functional Movement Screen (FMS) (http://functionalmovement.com), a battery oftests
developed in 1995 that relies upon common, basic movements to identify athletes that are
at an elevated risk of injury. The FMS was constructed around seven basic movement
patterns that were deemed to represent foundational actions of many sport maneuvers.
The tests devised by its creators measure mobility and stability for the upper and lower
extremities, as well as the trunk. Three stance positions form the core of the FMS and are
designed to challenge the whole body while standing upon various bases of support. They
include a symmetrical foot position during the deep squat, a split stance for the in-line
lunge, and a single leg stance in the hurdle step. Core stability, defined as the capacity to
59
resist motion about the trunk, is required during the deep squat, in-line lunge, trunk
stability push-up and rotary stability movements, whereas mobility of the upper
extremities is tested in the deep squat and shoulder mobility maneuvers. Although the
justification for the choice of these tests is logical, they were nonetheless chosen based
solely upon the clinical experience of sports medicine practitioners in the absence of any
research to validate the inclusion of each movement. Furthermore, no evidence has yet
been published to establish the screen's reliability and only one study to-date supports the
developer's assertation that a cutoff score of 14 out of 21 on the FMS is a reliable
predictor of injury. Despite the paucity of information and support for its design, the FMS
has been widely accepted in clinical practice, likely due to its simple, practical and
systematic protocol for executing the screen, as well as its delineated, ordinal scoring
system that enables practitioners to easily translate evaluation of movement into
numerical scores.
A subset of high school basketball athletes from the Eugene and Springfield, OR
region were evaluated with the FMS and followed throughout a single season to
determine if the test, and its widely-used numerical cutoff standard, demonstrated the
same predictive ability as that previously observed in professional football players
(Kiesel et aI., 2007). This subset of athletes was chosen because the successful
application of the FMS to this population could have a profound impact on sports in the
United States. That is, seven million high school students participate yearly in various
sports (Rechel et aI., 2008), whereas professional football players number less than 1,700
nationwide (http://www.nft.com). Furthermore, a high volume of injuries occurs in this
60
age group (Rechel et aI., 2008) and basketball is a sport in which upper and lower
extremities movements are common, which should fit well into the testing paradigm of
the FMS.
Reliability
Prior to the implementation of the FMS in a clinical or research setting, it is
incumbent upon the users to establish adequate reliability, across and within a subset of
evaluations, in scoring the tests. In essence, this means that the scoring performance on
the FMS movements should be able to be executed rapidly and efficiently, over time and
between evaluators, with consistent results. Without adequate reliability, attempts to
establish the validity of the FMS, with respect to injury prevention, cannot be performed
with any confidence and the outcomes of any attempts to do so in the absence of such
data must be viewed with caution. In addition, because the intent of the FMS is to screen
a large population of athletes, questions about practicality surface. This study was
designed using eight evaluators to efficiently screen subjects and minimize the impact of
such data collection on the teams' practice time. However, ifuse of multiple evaluators
was deemed untenable, reliance on a smaller number of personnel or, worse yet, a single
rater, would most certainly diminish efficiency and practicality.
The reliability study for this research tested eight Athletic Trainers' ability to
score the FMS tests consistently across multiple trials and evaluators. Fifteen subjects
(see Methods for demographics) were instructed on how to perform the FMS movements
by the primary investigator and subsequently video taped from the front and side while
61
they executed each test. The videos were then compiled and randomly presented to the
team of evaluators for scoring. Because the composite score is the primary predictor of
injury and the focus of this investigation, it was used as the reliability standard for
including or excluding raters for this study. Results indicate that, in all subjects, an
acceptable reliability coefficient of greater than or equal to 0.80 was achieved across and
within evaluators in obtaining FMS scores. (Tables 3.2 and 3.3) Thus, each rater was
included in the study of validating injury prediction. The data also show that all of the
tests could be scored individually with adequate reliability over time, whereas the rotary
stability test was found to be the only test that was not reliable between raters.
The results from the rotary stability test are noteworthy and should generate
caution. The lack of consistency in evaluating the movement might well be related to the
anatomical position of the subject and its complexity as the test is performed low to the
ground with many variables to assess. Scoring is further complicated by the fact that it is
not possible to evaluate all aspects of the test in a single trial or from a single viewing
location, requiring the subject to repeat the test several times. Due to the fact that subjects
were required to perform repeated trials, inconsistent performances are a possibility as
they attempt to execute the test while maintaining stability. Furthermore, this test
presented the greatest instructional challenge on proper execution since the quadruped,
unilateral movement required to earn a score of three is a complex movement.
In general, the well-defined movement criteria and scoring rubric are a strength of
the FMS; however, the rotary stability test may need further development to improve the
consistency of the movement scoring or it may be best to drop it and seek an alternative
62
method to assess the parameters it targets. It must be noted that these data are the first
published reliability study for the FMS. In light of the results, further research should be
conducted to specifically address the use of the rotary stability portion of the FMS to
determine if its inclusion compromises the overall screening procedure. Furthermore,
future research should attempt to design a real-time scoring scenario to ensure that the
reliability of the tests are evaluated as they would be scored in a real-life setting. The 2-D
nature of the video scoring design, although practical, may have over-simplified the
evaluations and enhanced the screen's reliability.
FMS Performance
Once inter-rater and intra-rater reliability of scoring the FMS were established,
the primary goal was to answer the question: can the FMS accurately predict athletes that
are most vulnerable to becoming injured? Thus, FMS and injury data were collected
throughout the duration of a single basketball season. For this subset of high school
basketball athletes, there were no significant differences in the FMS score for boys and
girls (Table 3.6), which allowed the researchers to pool the injury data. It should be noted
that the group means for two of the tests, the shoulder mobility and active straight leg
raise, were greater than the rest of the movements. These tests are designed to evaluate
mobility of the upper and lower extremities, and, in this young population, the subjects
routinely displayed sufficient range of motion to score high on these tests. (Tables 3.4
and 3.5, and Figure 3.1) In contrast, the high school athletes displayed a trend for low
scores on the deep squat (mean score of 1.62 and 1.68 for boys and girls respectively),
63
which evaluates the greatest number of joints moving simultaneously and requires
mobility in the ankles, knees, hips, thoracic spine and shoulders while stabilizing
extraneous motion in the knees and trunk. The complexity of this motion may have
caused both genders to score low. Furthermore, as noted in clinical practice, the deep
squat is typically one of the most difficult movements across all ages. (Unpublished
observation) Finally, girls scored the lowest for the group in any test with the trunk
stability push-up (mean score of 1.43), a test that demands upper extremity and core
strength.
Injuries
The frequency of injury in this study exceeded the findings from the general high
school athletic population (Rechel et aI., 2008), in that twenty-nine percent of the
basketball athletes became injured during the basketball season compared to twenty
percent in the previous study. In addition, the volume of injuries that were sustained
during practice compared to games was greater than the previous basketball population
studied by Rachel, et aI. (2008). This may well be simply related to the frequency of
practices versus games during a basketball season, although this explanation does not
account for the difference noted in the previous research. Anecdotally, there were several
early-season, overuse injuries sustained by athletes as they transitioned from other sports
like football, cross-country and volleyball, which could have influenced the volume of
injuries noted in the pre-season basketball practices. However, a comparison of FMS
scores for those suffering acute versus chronic injuries indicates that there is no
64
difference between the groups (Table 3.10) and suggests that type of injury is not a
confounding variable in this study. In any case, for boys and girls alike, the majority of
injuries affected the lower extremity, a finding that is consistent across a wide range of
sporting activities (Murphy et aI., 2003; Rechel et aI., 2008).
It should be noted that, although it is recommended to report injury using
exposure data (Fuller et aI., 2006), the logistics of this study prevented participation data
from being accurately collected. The researchers originally planned for each team's head
coach to record daily participation numbers, but compliance was lacking and the data set
was incomplete. Future studies should design an instrument for tracking this data through
the attending medical staffor post-season analysis of athletes' medical charts in order to
determine time-loss, and thus, participation.
Injuries included in this study were limited to those evaluated by the institution's
athletic trainer, sustained during a school-sanctioned basketball event, and void of any
contact with the ball, another player, the floor (with the exception of the feet) or any other
object in the gymnasium. In contrast with the professional football study, these criteria
resulted in inclusion of less severe injuries that occurred during basketball and did not
require any time-loss. This would result in more liberal inclusion of certain types of
injuries that might alter the FMS's sensitivity compared to the previous study. A self-
reporting system for the athletes to volunteer additional injuries that were not reported to
the athletic trainer was not used, although it may have captured an even greater number
of injuries due to the common underreporting of injuries that characterizes high school
athletes. This decision was made, as inclusion of this population would have
65
compromised the quality of data by relying on the memory and maturity of the athletes in
their responses.
The professional football study (Kiesel et aI., 2007) identified athletes that
became injured from both contact and non-contact mechanisms, whereas our study was
limited to non-contact injuries. The justification for that course of action was that, we
believed that although a good performance on the FMS tests is consistent with a high
level of mobility and stability, it is unclear how those qualities could positively assist an
individual in avoiding a contact injury. It may be theorized that increased mobility and
stability could help an athlete avoid contact or tolerate greater impact forces against the
body. However, we chose to exclude such injuries so as to eliminate any contribution
from those that occurred from random encounters on the court. Confirming this decision,
one athlete practiced for only two days during the season and sustained a finger fracture
on each of those days, causing her to miss the entire season. The relationship between
these types of injuries and the FMS could not be rationalized, and therefore, they were
excluded in their entirety.
FMS Injury Prediction
The previous research with professional football players (Kiesel et aI., 2007)
shows that athletes who score below 14 out of21 are eleven times more likely to sustain
an injury that results in greater than three weeks of missed participation. In the current
study, utilizing the cutoff score below fourteen to categorize athletes, seventy-six subjects
scored above and thirty-six scored below the accepted value. Twenty-four percent of the
66
athletes who scored fourteen or above became injured, whereas twenty-two percent
testing below fourteen sustained an injury. (Table 3.11) These data clearly contradict the
previous assertation (Kiesel et al., 2007) that a cutoff of fourteen identifies vulnerable
athletes. In an attempt to determine if a different cutoff score could more accurately
discriminate within this population, the value was altered to higher and lower scores and
the chi-square analysis was repeated. (Tables 3.12-3.15) It can be seen that, in all
comparisons, no alternative cutoff could be found that was able to identify at-risk
athletes.
While the FMS score was not successful in identifying vulnerable high school
basketball players, it might be possible that combinations of demographic information
(gender and level), the scores from FMS movements, and asymmetrical movements
(different score for right and left limbs) could actually prove more useful in injury
prediction than the FMS score alone. To this end, a logistic regression model was used,
which employs a combination of independent variables that best predict membership in a
particular group, as measured by a categorical dependent variable, in this case, injury
(Mertler & Vannatta, 2005). For example, anatomical and movement asymmetries have
previously been identified by several authors as potential indicators of injury risk (Knapik
et al., 1991; Baumhauer et al., 1995; Nadler et al., 2001; Plisky, Rauh, Kaminski, &
Underwood, 2006). The theory is that uneven stresses applied to the body during
asymmetrical movement lead to chronic injuries and/or predispose an individual to an
acute injury during landing or pivoting maneuvers (Plisky et al., 2006).
67
Despite the historical connection between asymmetries and injury, our results
failed to identify any of the variables included in the model as significant predictors of
injury, either alone or in combination with any other parameter. Similarly, Burton (2006),
a developer of the FMS, found that FMS movement asymmetries were not significant
predictors of injury in a population of firefighters. Furthermore, while past authors
(Murphy et aI., 2003; Emery et aI., 2005; Caine et aI., 2008) have indicated that a history
of previous injury is a significant predictor of future injury, a comparison of the FMS
scores for those sustaining a new versus recurring injury displayed no significant
differences between the groups (Table 3.17), which suggests that previous injury was not
a significant predictor of injury.
Collectively, our results provide the first evidence that, despite the fact that
anecdotal, clinical findings have led to widespread adoption of functional screening, the
FMS is not in any way a viable evaluation tool for identifying high school athletes
throughout a basketball season as being at high risk for injury. Our data do not rule out
the potential value of the FMS in other sport or recreational settings, as our protocols
were quite specific to the studied population. As described previously (see Introduction,
pg. 12), past efforts to identify tests that are predictive of injury have yielded conflicting
results and this study introduces similar disparity with the professional football study
(Kiesel et aI., 2007). Furthermore, the definition of injury was more inclusive than the
previous FMS study (Kiesel et aI., 2007), and may have introduced too much variability
and/or captured injuries that were unrelated to performance on the FMS, thus limiting the
screen's ability to accurately identify at-risk athletes. However, activities in each test
68
seem to fail to replicate the high forces and velocities that are produced during actual
practice or competition, which may contribute to our inability to find any positive
correlations. That is, the physical demands of various sports are extremely variable and a
single battery of tests may simply be insufficient to accurately predict the level of risk
across a broad spectrum of athletes. For example, football and basketball have great
dissimilarity in levels of contact, movements, playing surfaces, equipment, goals for the
sport, etc. that may easily account for the disparity in findings between the two studies.
Over several decades, many researchers have attempted to identify risk factors for
injury (Godshall, 1975; Knapik et ai., 1991; Twellaar et ai., 1997; Murphy et ai., 2003);
however, meaningful success in this area has been elusive. The complexity of injury
(mechanisms, predisposing conditions, fatigue, psychosocial factors, etc) may simply be
too great to allow anyone test, screen or other collection of predictive variables to
consistently predict injury. The answer may lie in creating a battery of tests that is sport-
specific so that the uniqueness of each activity can be properly evaluated. However,
despite the difficulties facing individuals who seek progress in this area, the effects on the
lives of those injured, the associated health care costs, long-term disability, and the loss
of physical vitality call for more extensive investigation into the utility of screening
methods, and/or the identification of other variables, or combination thereof, that may
predict injury more accurately. As players get larger, faster and stronger, the stress put on
them will likely raise the prospect of injury. Advances in treatment alone will not be
sufficient to meet the needs of athletes and society in the decades to come.
APPENDIX A
FUNCTIONAL MOVEMENT SCREEN SCORING RUBRIC
69
70
Movement III II I
• Upper torso parallel * Add heel raise • Unable to
with tibia or toward perform
vertical • Same criteria as movement
• Femur below score of III properly with
Deep Squat horizontal heel raise
• Knees aligned over
feet
• Dowel aligned over
feet
• Hips, knees & ankles • One or more of • Contact
aligned in sagittal scoring criteria between foot &
Hurdle Step plane for III is not hurdle
• Erect posture performed • Loss of balance
maintained
• Dowel contacts • One or more of • Loss of balance
remain with head, T- scoring criteria
In-line Lunge spine & L-spine for III is not
• Dowel & feet aligned performed
in sagittal plane
• Knee touches boardI
• Fists are within one • Fists are within • Fists are notShoulder hand length 1 112 hand within 1 112Mobility lengths hand lengths
Active • Ankle passes mid- • Ankle between • Ankle does not
Straight Leg thigh point knee & mid- pass knee
Raise thigh
• Males: 1 rep; thumbs • Males: 1 rep; • Males: unable
aligned with top of thumbs aligned to achieve
Trunk forehead with chin score of II
Stability • Females: 1 rep; • Females: 1 rep; • Females:
Push-up thumbs aligned with thumbs aligned unable to
chin with clavicle achieve score
of II
• Performs unilateral • Performs • Unable to
repetition diagonal perfoffil
Rotary • Spine parallel to repetition diagonal
Stability board • Same criteria as repetition
• Knee & elbow touch III
over board
* FMS created by: Gray Cook, PT, OCS, CSCS & Lee Burton, PhD, ATC, CSCS
** Note: Pain = 0
APPENDIXB
INFORMED CONSENT FORM RELIABILITY STUDY
71
72
INFORMED CONSENT
Reliability of Scoring the Functional Movement Screen and Y-balance Test
You are invited to participate in a graduate research study conducted by Eric Sorenson,
from the University of Oregon's Department of Human Physiology. I hope to learn about
the reliability of sports medicine professionals scoring the functional movement screen and
Y-balance test. The results from this study will contribute to the completion of my doctoral
dissertation.
If you decide to participate, you will be asked to attend the testing session at (testing
location) on (date) at (time). You will be measured for height and weight, and your age
will be recorded. You will then be tested for your ability to perform basic movements that
are a part of a functional movement screen. Movements included in the screen are a:
1. Deep squat without weight
2. Lunge
3. Hurdle step - stand on one leg; step over an obstacle
4. Straight leg raise -laying on back; raise leg towards sky
5. Shoulder flexibility test - touch both hands as close as possible on back
6. Push-up
7. Core stability test - positioned on hands and knees; touch elbow and knee
together
8. Y-balance test - stand on one leg; reach out other foot towards three targets
a. Following the Y-balance test, you will be asked to lie on a table in order
to have your leg length measured. The measurement will be taken from
your hip to your ankle using a tape measure.
You will be asked to bring shorts, a short-sleeved shirt, and athletic shoes to the day of
testing. All testing will take approximately thirty minutes. Evaluation of your screen will be
performed by 6-10 Certified Athletic Trainers and Physical Therapists. Movements 1-7, as
listed above, will be scored simultaneously by all of the evaluators. Movement 8, the Y-
balance test and leg measurements, will be conducted by each evaluator individually.
You will also be video taped from the front and side during your functional movement
screen (movements 1-7 as listed above). This information will be used by the evaluators to
score your movements several times following the live scoring session. We will use this
method of scoring to test the evaluators' consistency with scoring your movements.
Furthermore, the videos may be used in professional presentations and conferences. Your
identity will remain confidential, but your face will be visible on the video.
There are minimal risks involved in this study but precautions will be taken to minimize the
possible risk. All the functional movement tests will be within a normal range of motion and
73
speed. You will be under complete control over the amount of movement produced during
each of the tests. The investigator will monitor you at all times for any discomfort associated
with the activities. Data collection will stop if an activity is causing undesired discomfort.
This study hopes to add to the body of knowledge available regarding the consistency of
sports medicine professional with scoring functional movements. However, I cannot
guarantee that you personally will receive any benefits from this research.
Your participation is voluntary. Your decision whether or not to participate will not affect
your relationship with the University of Oregon or the Department of Human Physiology. If
you decide to participate, you are free to withdraw your consent and discontinue
participation at any time without penalty.
You are advised to go the University's Student Health Center if you become injured. You
will be held responsible for any treatment costs in the case that you are injured.
If you have any questions, please feel free to contact Eric Sorenson at (541) 206-2586,
esorensl@uoregon.edu, or his faculty advisor Dr. Gary Klug at (541) 346-4181. If you have
questions regarding your rights as a research subject, contact Human Subjects Compliance,
University of Oregon, Eugene, OR 97403, (541) 346-2510.
Your signature indicates that you have read and understand the information provided above,
that you willingly agree to participate, that you may withdraw your consent at any time,
discontinue participation without penalty, and that you are not waiving any legal claims,
rights or remedies. It also indicates that I have received an adequate description of the
purpose and procedures for videotaping during the course of the proposed research study.
I give my consent to be videotaped during participation in the study, and for those
videotapes to be viewed by persons involved in the study, as well as for use in
professional presentations and conferences as described to me. I understand that my face
may be visible on the tape, but that all information will be kept confidential and will be
reported in an anonymous fashion, and that the videotapes will be kept indefinitely. I
further understand that I may withdraw my consent at any time.
Subject's Name
Subject's Signature Date
APPENDIX C
AGREEMENT FOR VIDEOTAPING
74
75
AGREEMENT FOR VIDEOTAPING
I have received an adequate description of the purpose and procedures for videotaping
sessions during the course of the proposed research study. I give my consent to be
videotaped during participation in the study, and for those videotapes to be viewed by
persons involved in the study, as well as for use in professional presentations and
conferences as described to me. I understand that my face may be visible on the tape, but
that all information will be kept confidential and will be reported in an anonymous
fashion, and that the videotapes will be kept indefinitely. I further understand that I may
withdraw my consent at any time.
Print Name
---------------------
Signature of participant Date
------------ ----
APPENDIXD
FUNCTIONAL MOVEMENT SCREEN SCORING SHEET
76
77
Subject
Test Raw Score Final
lDeep
Squat
2Hurdle St
R
2Hurdle St L
3Lunge R
3Lunge L
4Sh Mob R
4Sh Mob L
Impinge R
Impinge L
SASLR R
SASLR L
6Push up
Extension
7Quad R
7Quad L
Flexion
* Note: This table had to be modified for publication. Three columns of the table above
were used on the actual scoring sheet.
APPENDIXE
INFORMED CONSENT FORM INJURY PREDICTION STUDY
78
79
INFORMED CONSENT
Functional Movement Screen and Y-balance as Predictors of Injury in High
School Athletes
You are invited to participate in a graduate research study conducted by Eric Sorenson, from
the University of Oregon's Department of Human Physiology. I hope to learn about the
ability of a functional movement screen to identify athletes that are at risk of injury. The
results from this study will contribute to the completion of my doctoral dissertation.
If you decide to participate, you will be asked to complete a demographic information
form indicating your age, gender, year-in-school, previous musculoskeletal injuries (self-
reported), school name, sport, team level (varsity, junior varsity, etc), years-of-
participation, projected player status (starter, second-string, etc), and position. (See
attached) Next, you will be measured for your height and weight, and your ability to
perform basic movements that are a part of a functional movement screen. Movements
included in the screen are a:
1. Deep squat without weight
2. Lunge
3. Hurdle step - stand on one leg; step over an obstacle
4. Straight leg raise - laying on back; raise leg towards sky
5. Shoulder flexibility test - touch both hands as close as possible on back
6. Push-up
7. Core stability test - positioned on hands and knees; touch elbow and knee
together
8. Y-balance test - stand on one leg; reach out other foot towards three targets
a. Following the Y-balance test, you will be asked to lie on a table in order
to have your leg length measured. The measurement will be taken from
your hip to your ankle using a tape measure ..
You will be asked to bring shorts, a short-sleeved shirt, and athletic shoes to the day of
testing. All testing will take approximately fifteen minutes and will be conducted as a part of
your typical training session with your team.
During the course of your athletic season, your coach will keep track of the number of
athletes participating in a competition or practice for each day . Your name will not be
included in these records. In the circumstance that you may become injured as a part of your
participation with your sport, your coach and athletic trainer will fill out an injury report
fonn. This fonn will include infonnation about your injury, including the date of injury, date
that your return to full participation, injured body part, type of injury (sprain, strain, fracture,
etc), diagnosis, whether you have had a similar injury to the same area in the past, when the
injury occurred (practice or competition), and the cause of injury. The injury report fonn
will contain your name, but this information will remain in a locked filing cabinet. With
your pennission, the injury report fonn will be made available to me for analysis. Any
80
information that is obtained in connection with this study and that can be identified with you
will remain confidential and will be disclosed only with your permission. Subject identities
will be kept confidential by a coding system known only by the investigator.
There are minimal risks involved in this study but precautions will be taken to minimize the
possible risk. All the functional movement tests will be within a normal range of motion and
speed. You will be under complete control over the amount of movement produced during
each of the tests. The investigator will monitor you at all times for any discomfort associated
with the activities. Data collection will stop if an activity is causing undesired discomfort.
This study hopes to add to the body of knowledge available regarding the ability of
functional movement screens to identify athletes that are at-risk of injury. However, I cannot
guarantee that you personally will receive any benefits from this research.
Your participation is voluntary. Your decision whether or not to participate will not affect
your relationship with the (specified school district) or (specified school). If you decide to
participate, you are free to withdraw your consent and discontinue participation at any time
without penalty.
You are advised to notify your team's coach, athletic trainer or school nurse if you become
injured during the testing or over the course of the season. You will be held responsible for
any treatment costs in the case that you are injured.
If you have any questions, please feel free to contact Eric Sorenson at (541) 206-2586,
esorensl@uoregon.edu, or his faculty advisor Dr. Gary Klug at (541) 346-4181. If you have
questions regarding your rights as a research subject, contact Human Subjects Compliance,
University of Oregon, Eugene, OR 97403, (541) 346-2510.
Your signature indicates that you have read and understand the information provided above,
that you willingly agree to participate, that you may withdraw your consent at any time,
discontinue participation without penalty, and that you are not waiving any legal claims,
rights or remedies. It also indicates that you are willing to release injury information to Eric
Sorenson during the 2008-2009 school year. This information will not include any past
medical information, except for musculoskeletal injury information, which you volunteer
during on the demographic information form.
Subject's Name
Subject's Signature
Parent's Name (if subject is under 18 years old)
Parent's Signature (if subject is under 18 years old)
Date
Date
APPENDIXF
DEMOGRAPHICS FORM
81
Name:
Birth Date (Month/DaylYear): _
82
Gender: Male Female
Year in school: FR SO JR SR
Team Level: FR JV Varsity
Projected status: Starter Second-team
Position(s):
Number of years playing this sport: _
Previous Injuries:
Third-team
Month/Year RlL Body part Type of Injury Surgery? YIN
APPENDIXG
INmRY REPORT FORM
83
Date of Injury: _
84
Date of return to full participation: _
Injured body part
o head/face
11 k/ . I .
'--' nec cervlca spme
o sternum/ribs/upper back
o abdomen
o low back/sacrum/pelvis
Injured body part
o right o left
[J shoulder /clavicle
n upper arm
n~ elbow
n forearm
o wrist
o hand/finger/thumb
o not applicable
o hip/groin
o thigh
o knee
o leg/Achilles tendon
n
<..-.-' ankle
o foot/toe
Type of Injury
o concussion
o fracture
o other bony injury
o dislocation
o sprain
Diagnosis
o lesion of meniscus or cartilage
o muscle strain/cramp
o tendon injury/bursitis
o dental injury
o other (please specify)
o hematoma/contusion
o abrasion
o laceration
o nerve injury
Has the player had a previous injury of the same type at the same location (i.e. this injury is a
reoccurrence)? 0 yes 0 no
If Yes, specify the date of player's return to full participation from the previous injury:
Was the injury caused by overuse or trauma?
When did the injury occur? o training
o overuse
o game
o trauma
Was the injury caused by contact?
o yes, with another player
Ono
o yes, with the ball o yes, with other object:
85
REFERENCES
American College of Sports Medicine. (2010). ACSM's Guidelines for Exercise Testing
and Prescription (8th ed.). Lippincott Williams and Wilkins.
Amason, A., Gudmundsson, A., Dahl, H. A., & Johannsson, E. (1996). Soccer injuries in
Iceland. Scandinavian Journal ofMedicine & Science in Sports, 6(1),40-45.
Barrett, J., & Bilisko, T. (1995). The role of shoes in the prevention of ankle sprains.
Sports Medicine, 20(4), 277-280.
Baurnhauer, J. F., Alosa, D. M., Renstrom, P. A. F. H., Trevino, S., & Beynnon, B.
(1995). A prospective study of ankle injury risk factors. The American Journal of
Sports Medicine, 23(5), 564-570.
Beynnon, B. D., Renstrom, P. A., Alosa, D. M., Baurnhauer, J. F., & Vacek, P. M.
(2001). Ankle ligament injury risk factors: a prospective study of college athletes.
Journal ofOrthopaedic Research, 19(2),213-220.
Burton, S. L. (2006). Performance and Injury Predictability during Firefighter Candidate
Training. Virginia Polytechnic Institute and State University, Blacksburg, Virginia.
Caine, D., Maffulli, N., & Caine, C. (2008). Epidemiology of injury in child and
adolescent sports: injury rates, risk factors, and prevention. Clinical Journal of
Sports Medicine, 27(1),19-50, vii.
Chomiak, J., Junge, A., Peterson, L., & Dvorak, J. (2000). Severe injuries in football
players. Influencing factors. American Journal ofSports Medicine, 28(5 Suppl),
S58-68.
Downing, S. M. (2004). Reliability: on the reproducibility of assessment data. Medical
Education, 38(9), 1006-1012.
Ekstrand, J., & Gillquist, J. (1983). Soccer injuries and their mechanisms: a prospective
study. Medicine & Science in Sports & Exercise, 15(3),267-270.
Eliasziw, M., Young, S. L., Woodbury, M. G., & Fryday-Fie1d, K. (1994). Statistical
methodology for the concurrent assessment of interrater and intrarater reliability:
using goniometric measurements as an example. Journal ofPhysical Therapy,
74(8),777-788.
86
Emery, C. A., Meeuwisse, W. H., & Hartmann, S. E. (2005). Evaluation of risk factors
for injury in adolescent soccer: implementation and validation of an injury
surveillance system. American Journal ofSports Medicine, 33(12), 1882-1891.
Fuller, C. W., Ekstrand, J., Junge, A., Andersen, T. E., Bahr, R., Dvorak, J. et al. (2006).
Consensus statement on injury definitions and data collection procedures in studies
of football (soccer) injuries. British Journal ofSports Medicine, 40(3),193-201.
Godshall, R. W. (1975). The predictability of athletic injuries: an eight-year study.
Journal ofSports Medicine, 3(1), 50-54.
Gray, J., Taunton, J. E., McKenzie, D. C., Clement, D. B., McConkey, J. P., & Davidson,
R. G. (1985). A survey of injuries to the anterior cruciate ligament of the knee in
female basketball players. International Journal ofSports Medicine, 6(6), 314-316.
Guskiewicz, K. M., & Perrin, D. H. (1996). Research and clinical applications of
assessing balance. Journal ofSports Rehabilitation, 5, 45-63.
Gwinn, D. E., Wilckens, J. H., McDevitt, E. R., Ross, G., & Kao, T. C. (2000). The
relative incidence of anterior cruciate ligament injury in men and women at the
United States Naval Academy. American Journal ofSports Medicine, 28(1), 98-
102.
Hayen, A., Dennis, R. J., & Finch, C. F. (2007). Determining the intra-and inter-observer
reliability of screening tools used in sports injury research. Journal ofScience and
Medicine in Sport, 10(4),201-210.
Hewett, T. E. (2000). Neuromuscular and hormonal factors associated with knee injuries
in female athletes. Strategies for intervention. Sports Medicine, 29(5),313-327.
Hickey, G. J., Fricker, P. A., & McDonald, W. A. (1997). Injuries of young elite female
basketball players over a six-year period. Clinical Journal ofSport Medicine, 7(4),
252-256.
Hickmans, J. (2007). Addressing Functionality: The Development of the Functional
Movement Screen. Strength and Conditioning Coach, 15(4), 3-6.
Hopper, D. M., Hopper, J. L., & Elliott, B. C. (1995). Do selected kinanthropometric and
performance variables predict injuries in female netball players? Journal ofSports
Sciences, 13(3),213-222.
Hosea, T. M., Carey, C. C., & Harrer, M. F. (2000). The gender issue: epidemiology of
ankle injuries in athletes who participate in basketball. Clinical Orthopaedics and
Related Research, 372),45-49.
87
Kaufman, K. R., Brodine, S. K., Shaffer, R. A., Johnson, C. W., & Cullison, T. R. (1999).
The effect of foot structure and range of motion on musculoskeletal overuse
injuries. American Journal ofSports Medicine, 27(5), 585-593.
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 ofSports Physical Therapy, 2(3),147-158.
Knapik, J. J., Bauman, C. L., Jones, B. H., HarTis, J. M. A., & Vaughan, L. (1991).
Preseason strength and flexibility imbalances associated with athletic injuries in
female collegiate athletes. American Journal ofSports Medicine, 19(1), 76-81.
Knapik, J. J., Sharp, M. A., Canham-Chervak, M., Hauret, K., Patton, J. F., & Jones, B.
H. (2001). Risk factors for training-related injuries among men and women in basic
combat training. Medicine & Science in Sports & Exercise, 33(6), 946-954.
Knowles, S. B., Marshall, S. W., Bowling, J. M., Loomis, D., Millikan, R., Yang, J. et al.
(2006). A prospective study of injury incidence among North Carolina high school
athletes. American Journal ofEpidemiology, 164(12),1209-1221.
Krivickas, L. S., & Feinberg, J. H. (1996). Lower extremity injuries in college athletes:
relation between ligamentous laxity and lower extremity muscle tightness. Archives
ofPhysical Medicine & Rehabilitation, 77(11), 1139-1143.
McGuine, T. A., Greene, J. J., Best, T., & Leverson, G. (2000). Balance As a Predictor of
Ankle Injuries in High School Basketball Players. Clinical Journal ofSport
Medicine, 10(4),239-244.
McKay, G. D., Goldie, P. A., Payne, W. R., & Oakes, B. W. (2001). Ankle injuries in
basketball: injury rate and risk factors. British Journal ofSports Medicine, 35(2),
103-108.
Mertler, C. A., & Vannatta, R. A. (2005). Advanced and Multivariate Statistical
Methods: Practical Application and Interpretation; 3rd edition. Los Angeles, CA:
Pyrczak Publishing.
Milgrom, C., Shlamkovitch, N., Finestone, A., Eldad, A., Laor, A., Danon, Y. L. et al.
(1991). Risk factors for lateral ankle sprain: a prospective study among military
recruits. Foot & Ankle, 12(1),26-30.
Murphy, D. F., Connolly, D. A., & Beynnon, B. D. (2003). Risk factors for lower
extremity injury: a review of the literature. British Journal ofSports Medicine,
37(1),13-29.
88
Myklebust, G., Maehlum, S., Holm, I., & Bahr, R. (1998). A prospective cohort study of
anterior cruciate ligament injuries in elite Norwegian team handball. Scandinavian
Journal ojMedicine & Science in Sports, 8(3),149-153.
Nadler, S. F., Malanga, G. A., Feinberg, J. H., Prybicien, M., Stitik, T. P., & DePrince,
M. (2001). Relationship Between Hip Muscle Imbalance and Occurrence of Low
Back Pain in Collegiate Athletes: A Prospective Study. American Journal oj
Physical Medicine & ~ehabilitation, 80(8), 572-577.
Ostenberg, A., & Roos, H. (2000). Injury risk factors in female European football. A
prospective study of 123 players during one season. Scandinavian Journal oj
Medicine & Science in Sports, 10(5),279-285.
Plisky, P. J., Rauh, M. J., Kaminski, T. W., & Underwood, F. B. (2006). Star Excursion
Balance Test as a predictor of lower extremity injury in high school basketball
players. The Journal oj Orthopaedic & Sports Physical Therapy, 36(12), 911-919.
Powell, J. W., & Barber-Foss, K. D. (2000). Sex-related injury patterns among selected
high school sports. American Journal ojSports Medicine, 28(3),385-391.
Powell, J. W., & Barber-Foss, M. S. (1999). Injury Patterns in Selected High School
Sports: A Review of the 1995 - 1997 Seasons. Journal ojAthletic Training, 34(3),
277-284.
Rechel, J. A., Yard, E. E., & Comstock, R. D. (2008). An epidemiologic comparison of
high school sports injuries sustained in practice and competition. Journal ojAthletic
Training, 43(2), 197-204.
Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: uses in assessing rater
reliability. Psychological Bulletin, 86(2), 420-428.
Soderman, K., Alfredson, H., Pietila, T., & Werner, S. (2001). Risk factors for leg
injuries in female soccer players: a prospective investigation during one out-door
season. Knee Surgery, Sports Traumatology, Arthroscopy, 9(5), 313-321.
Starkey, C., & Ryan, J. (2002). Evaluation oj Orthopedic and Athletic Injuries, 2nd Ed.
Philadelphia: FA Davis Company.
Stevenson, M. R., Hamer, P., Finch, C. F., Elliot, B., & Kresnow, M. (2000). Sport, age,
and sex specific incidence of sports injuries in Western Australia. British Journal oj
Sports Medicine, 34(3), 188-194.
89
Tropp, H., Ekstrand, J., & Gillquist, J. (1984). Stabilometry in functional instability of
the ankle and its value in predicting injury. Medicine & Science in Sports &
Exercise, 16(1),64-66.
Twellaar, M., Verstappen, F. T., Huson, A., & van Mechelen, W. (1997). Physical
characteristics as risk factors for sports injuries: a four year prospective study.
International Journal ojSports Medicine, 18(1), 66-71.
Wiesler, E. R., Hunter, D. M., Martin, D. F., Curl, W. W., & Hoen, H. (1996). Ankle
flexibility and injury patterns in dancers. American Journal ojSports Medicine,
24(6),754-757.
Zelisko, J. A., Noble, H. B., & Porter, M. (1982). A comparison of men's and women's
professional basketball injuries. American Journal ojSports Medicine, 1O(5), 297-
299.