[ research report ]

DEYDRE S. TEYHEN, PT, PhD1,2 • SCOTT W. SHAFFER, PT, PhD2 • CHELSEA L. LORENSON, PT3 • JOSHUA P. HALFPAP, PT3
DUSTIN F. DONOFRY, PT3 • MICHAEL J. WALKER, PT, DSc4 • JESSICA L. DUGAN, PT5 • JOHN D. CHILDS, PT, PhD2,6

The Functional Movement

Screen: A Reliability Study

M
ore than 10 000 Amer-
TTSTUDY DESIGN: Reliability study. TTRESULTS: The average  SD score on the FMS icans seek medical treat-
TTOBJECTIVES: To determine intrarater test- was 15.7  0.2 points, with 15.6% (n = 10) of the
ment for sports, re-
retest and interrater reliability of the Functional participants scoring less than or equal to 14 points,
Movement Screen (FMS) among novice raters. the recommended cutoff for predicting time-loss creational activity, and
injuries. The intrarater test-retest and interrater
TTBACKGROUND: The FMS is used by various reliability of the FMS composite score resulted in
exercise-related injuries on a
examiners to assess movement and predict time- an ICC3,1 of 0.76 (95% CI: 0.63, 0.85) and an ICC2,1 daily basis.20 Researchers have
loss injuries in diverse populations (eg, youth to
professional athletes, firefighters, military service
of 0.74 (95% CI: 0.60, 0.83), respectively. The stan- estimated that 50% to 80% of
dard error of the measurement of the composite
members) of active participants. Unfortunately, test was within 1 point, and the MDC95 values were
these injuries are overuse in
critical analysis of the reliability of the FMS is 2.1 and 2.5 points on the 21-point scale for interra- nature and involve the lower extrem-
currently limited to 1 sample of active college-age ter and intrarater reliability, respectively. The inter- ity.1,11,25 In the military, physical training
participants. rater agreement of the component scores ranged and exercise-related injuries account for
TTMETHODS: Sixty-four active-duty service from moderate to excellent (κw = 0.45-0.82).
30% of hospitalizations and 40% to 60%
members (mean  SD age, 25.2  3.8 years; TTCONCLUSION: Among novice raters, the FMS of all outpatient visits, with 10 to 12 inju-
body mass index, 25.1  3.1 kg/m2) without composite score demonstrated moderate to good
a history of injury were enrolled. Participants ries per 100 soldier-months.12 Although
interrater and intrarater reliability, with accept-
completed the 7 component tests of the FMS in a
able levels of measurement error. The measures the risk of musculoskeletal conditions
counterbalanced order. Each component test was and injuries is multifactorial,7,9,10,15,17-19
of reliability and measurement error were similar
scored on an ordinal scale (0 to 3 points), resulting
for both intrarater reliability that repeated the preliminary evidence suggests that neuro-
in a composite score ranging from 0 to 21 points.
assessment of the movement patterns over a 48- muscular and strength training programs
Intrarater test-retest reliability was assessed
to-72–hour period and interrater reliability that had may be beneficial for preventing the oc-
between baseline scores and those obtained with
2 raters assess the same movement pattern si-
repeated testing performed 48 to 72 hours later. currence of these conditions.7,9,10,15,17-19
multaneously. The interrater agreement of the FMS
Interrater reliability was based on the assessment However, tools that assess movement
component scores was good to excellent for the
from 2 raters, selected from a pool of 8 novice
push-up, quadruped, shoulder mobility, straight leg to help predict those at highest risk for
raters, who assessed the same movements on day
raise, squat, hurdle, and lunge. Only 15.6% (n = 10) musculoskeletal conditions and injuries
2 simultaneously. Descriptive statistics, weighted
of the participants were identified to be at risk for have been lacking for both athletic and
kappa (κw), and percent agreement were calcu-
injury based on previously published cutoff values.
lated on component scores. Intraclass correlation military populations. The Functional
J Orthop Sports Phys Ther 2012;42(6):530-540,
coefficients (ICCs), standard error of the measure- Movement Screen (FMS) is a relatively
Epub 14 May 2012. doi:10.2519/jospt.2012.3838
ment, minimal detectable change (MDC95), and
TTKEY WORDS: injury prediction, injury preven-
new tool that attempts to address mul-
associated 95% confidence intervals (CIs) were
calculated on composite scores. tion, injury risk, interrater, intrarater tiple movement factors, with the goal of
predicting general risk of musculoskeletal

1
Commander, US Army Public Health Command Region-South, Fort Sam Houston, TX. 2Associate Professor, US Army-Baylor University, Fort Sam Houston, TX. 3Physical Therapy
Intern, US Army-Baylor University, Fort Sam Houston, TX. 4Assistant Professor, US Army-Baylor University, Fort Sam Houston, TX. 5Researcher, TRUE Research Foundation, San
Antonio, TX. 6Director of Musculoskeletal Research, Department of Physical Therapy (MSGS/SGCUY), 81st Medical Group, Keesler Air Force Base, Biloxi, MS. This research
study was approved by the Brooke Army Medical Center Institutional Review Board. The views expressed herein are those of the authors and do not reflect the official policy or
position of Brooke Army Medical Center, the US Army Medical Department, the US Army Office of the Surgeon General, the Department of the Army, Department of the Air Force,
Department of Defense, or the US Government. Address correspondence to Dr Deydre S. Teyhen, US Army-Baylor University, 3151 Scott Road, Room 1303 (ATTN: MCCS-HGE-
PT), Fort Sam Houston, TX 78234. E-mail: [email protected] or [email protected]

530  |  june 2012  |  volume 42  |  number 6  |  journal of orthopaedic & sports physical therapy

42-06 Teyhen.indd 530 5/22/2012 5:38:56 PM

 conditions and injuries.3-5,13,14,16
The FMS was designed to identify
functional movement deficits and asym-
metries that may be predictive of general
musculoskeletal conditions and injuries,
with an ultimate goal of being able to
modify the identified movement deficits
through individualized exercise prescrip-
tion.3,4 The FMS consists of 7 fundamen-
tal movement component tests (FIGURE 1)
that are scored on a scale of 0 to 3, with
the sum creating a composite score rang-
ing from 0 to 21 points.3,4 The 7 move-
ment patterns that are assessed include
the deep squat, in-line lunge, hurdle
step, shoulder mobility, active straight leg
raise, trunk stability push-up, and quad-
ruped rotary stability.
Preliminary research by Kiesel et al14
suggests that National Football League
(NFL) players (n = 46) who had a com-
FIGURE 1. Functional Movement Screen tests. (A) In-line lunge, (B) hurdle step, (C) deep squat, (D) quadruped
posite score less than or equal to 14 on
rotary stability, (E) active straight leg raise, (F) shoulder mobility, and (G) trunk stability push-up.
the FMS had an odds ratio of 11.7 (95%
confidence interval [CI]: 2.5, 54.5) and a etal injuries are predicted by low FMS novice and expert raters in a sample of
positive likelihood ratio of 5.8 (95% CI: composite scores, and whether the origi- active college-age participants (to in-
2.0, 18.4) to sustain a time-loss injury. Al- nal cutoff score of less than or equal to 14 clude college varsity athletes). However,
though the specificity was relatively high points on the FMS is valid in the different this study had several limitations: (1) it
(0.9; 95% CI: 0.8, 1.0), the sensitivity was populations. did not assess test-retest reliability, (2)
low (0.5; 95% CI: 0.3, 0.7), indicating Additionally, researchers have found all raters assessed the same movement
that FMS composite scores less than or that FMS composite scores increased pattern via videotaped analysis, and (3)
equal to 14 may suggest higher injury risk in football players,13 firefighters,6 and it only assessed agreement of individual
but FMS composite scores greater than service members8 following corrective FMS component scores and did not as-
14 do not rule out future injury risk. In a exercises that addressed possible impair- sess the overall FMS composite score,
separate study on a group of Marines, a ments associated with altered movement which is typically used as the primary
composite score less than or equal to 14 patterns noted on the FMS component indicator of injury risk. Traditionally,
on the FMS demonstrated limited abil- tests. In a group of Marines, 80% of those the FMS is assessed in real time, with-
ity to predict all future musculoskeletal with a score less than or equal to 14 also out the benefit of video playback. Vari-
injuries (traumatic or overuse), with a demonstrated lower fitness scores on a ability of human movement across trials
sensitivity of 0.45 and specificity of 0.71, standardized fitness test compared to theoretically should exist; therefore, test-
while the same cutoff value was able to those who had an FMS composite score retest analysis could lower the reported
predict a serious injury (any injury that greater than 14.21 However, Okada et al22 agreement values. Additionally, the FMS
was severe enough to remove the par- found that FMS composite scores were is often assessed in a group setting (eg,
ticipant from the training program) with not related to performance or core stabil- preseason physical or preparticipation
a sensitivity of 0.12 and a specificity of ity measures among healthy participants. screening), requiring the use of multiple
0.94.21 The FMS was also able to predict Interpretation of FMS scores is lim- raters, who may or may not be the same
injury risk in female collegiate athletes.2 ited by the scant evidence16 regarding raters to assess the movement at follow-
Finally, in another study, firefighters with the FMS's psychometric properties and, up testing. Therefore, a more robust reli-
a previous history of injury demonstrated in particular, the reliability of both com- ability study is required to enhance the
lower FMS composite scores.23 However, posite and individual component scores. understanding of the psychometric prop-
it is not clear for which sports or profes- An initial study by Minick et al16 found erties of the FMS.
sions the FMS is optimal in predicting acceptable levels of interrater agreement Although these initial FMS studies,
injury risk, what types of musculoskel- on the FMS component scores among which established the validity of the FMS

journal of orthopaedic & sports physical therapy | volume 42 | number 6 | june 2012 | 531

42-06 Teyhen.indd 531 5/22/2012 5:38:57 PM

 [ research report ]
for predicting musculoskeletal conditions
and injuries and the response to training, TABLE 1 Demographics
are encouraging, their data are prelimi-
nary and not published in widely accessi-
ble journals. Exploring the psychometric Type Mean  SD 95% CI
properties of the FMS in a large active Age, y 25.2  3.8 24.3, 26.2
population would enhance the general- Height, cm 175.5  9.6 173.1, 177.9
izability of the previous findings beyond Weight, kg 77.5  12.5 74.4, 80.7
a limited subgroup of professional and Body mass index, kg/m2 25.1  3.1 24.3, 25.9
collegiate athletes and students. The Abbreviation: CI, confidence interval.
primary purpose of this study was to de-
termine the intrarater (test-retest) and
interrater reliability of the FMS com-
ponent and composite scores in young, TABLE 2 FMS Descriptive Analysis*
healthy service members, when tested by
a counterbalance group of novice raters
FMS Component Score
in real time. Specifically, agreement was
Test 0 1 2 3 Mean  SD
assessed on the FMS component scores,
Trunk stability push-up 0 7 29 27 2.3  0.7
whereas reliability, response stability, and
Quadruped rotary stability 0 3 56 5 2.0  0.3
error threshold measurements were ob-
Shoulder mobility 0 2 19 43 2.6  0.6
tained for the FMS composite scores. A
Active straight leg raise 0 1 36 27 2.4  0.5
secondary purpose of this study was to
Deep squat 0 3 42 19 2.3  0.5
describe the FMS component and com-
Hurdle step 0 1 51 12 2.2  0.4
posite scores in this population.
In-line lunge 0 1 29 33 2.5  0.5

METHODS Abbreviation: FMS, Functional Movement Screen.

*The data displayed represent the first analysis of rater 1 on the first day of data collection (n = 64).

Participants

T
he convenience sample included due to other musculoskeletal injuries; potential perception of coercion. All par-
participants who were recruited had a history of fracture (stress or trau- ticipants signed consent forms approved
over an 8-week period from service matic) in the femur, pelvis, tibia, fibula, by the Brooke Army Medical Center In-
members in training at Fort Sam Hous- talus, or calcaneus; or were known to be stitutional Review Board.
ton, TX. Potential participants were pregnant.
provided a briefing about the study and Potential participants were provided Examiners
were given the opportunity to volunteer. an overview of the research study and The novice examiners participating in
Participants were eligible for inclusion if specific details of the entrance criteria. this study consisted of 8 physical therapy
they were between the ages of 18 and 35 After the presentation was completed, students enrolled in their second and
years or emancipated minors (17-year- those who met the entrance criteria were third semesters of a doctor of physical
olds who are considered adults and al- asked to squat and then hop unilaterally therapy training program prior to their
lowed to join the armed services), fluent on each leg in the group setting. Individ- 1-year clinical internship. Before testing,
in English, and had no current or previ- uals who met the entrance criteria and all examiners underwent 20 hours of
ous complaint of lower extremity pain, did not have pain on the squat and hop FMS training led by 4 physical therapists
spine pain, or medical or neuromuscu- tests were informed about upcoming data and 1 research assistant. Four physical
loskeletal disorders that limited partici- collection dates. Those individuals who therapy students were randomly as-
pation in work or exercise in the last 6 opted to volunteer returned the follow- signed to the participants to assess intra-
months. Participants were excluded if ing week to sign informed consent forms rater test-retest reliability by assessing
they were currently seeking medical care and were enrolled in the study. Within the FMS on day 1 and day 2. The goal
for lower extremity injuries or had pre- the military training environment, these of randomly selecting a rater to perform
vious medical history that included any procedures allowed potential partici- the intrarater test-retest reliability was
surgery for lower extremity injuries. Par- pants the option to not return if they to increase the variability in the study
ticipants were also excluded if they were were not interested in volunteering in the design. Each rater used for the intrarater
unable to participate in physical training study, and were designed to minimize any test-retest reliability measured between

532  |  june 2012  |  volume 42  |  number 6  |  journal of orthopaedic & sports physical therapy

42-06 Teyhen.indd 532 5/22/2012 5:38:58 PM

 14 and 18 participants. There were no on the quality of movement, with 3 be- analyses were conducted using SPSS Ver-
differences in outcomes across raters, so ing the maximum score.3-5 A score of 2 sion 17.0 (SPSS Inc, Chicago, IL).
aggregate data were analyzed. indicated that the participant required
A second set of 4 physical therapy stu- some type of compensation or was un- RESULTS
dents were randomly assigned to view the able to complete the entire movement. A

S
participants’ movement simultaneously score of 1 was given if the individual was ixty-four participants (53
with the first set of raters for the inter- unable to remain in the movement posi- males, 11 females) met the inclusion
rater reliability assessment on day 2. To tion throughout the movement, lost bal- and exclusion criteria and complet-
minimize bias, raters were randomly as- ance during the test, or did not meet the ed the study (TABLE 1). The mean  SD
signed, raters for day 2 were blinded to minimum criteria to score a 2. Pain dur- age of the participants was 25.2  3.8
day 1 raters’ measurements, pairs of rat- ing any of the FMS component tests or years and their body mass index was 25.1
ers on day 2 were blinded to each other's during any of the clearing tests indicated  3.1 kg/m2. Overall, the participants in-
analysis and scoring, and 48 to 72 hours a score of 0. All participants were allowed cluded routine exercisers who endorsed
of time elapsed between intrarater test- to perform each component test up to 3 a statement that they exercised a mini-
retest reliability measurements. The goal times, and the maximal score achieved mum of 4 days per week (n = 54, 78.2%).
of having a random set of 2 raters assess was recorded. The scores of the compo- Although the participants were attending
each participant was to increase the vari- nent tests were summed, resulting in a training for their military occupation, the
ability in the study design to more closely composite score from 0 to 21 points, with majority of the participants were routine
mimic field conditions, which often in- 21 being the maximum composite score. exercisers for more than 3 years. Specifi-
clude mass screenings that utilize mul- Additional details on scoring of each of cally, 29 (45.3%) participants reported
tiple raters. the component tests and the composite performing routine exercise for more
score are provided elsewhere3-5 and in the than 5 years, 21 (32.8%) for 3 to 5 years,
Procedures APPENDIX. 9 (14.1%) for 1 to 3 years, and 5 (7.8%) for
The FMS is composed of 7 component less than 1 year. Descriptive statistics on
tests used to assess different fundamental Statistical Analysis FMS performance are provided in TABLE 2.
movement patterns.3-5 Participants com- Descriptive statistics and frequency None of the participants had pain on the
pleted the component tests in a coun- counts were calculated. Agreement of 3 FMS clearing tests. Interrater reliability
terbalanced order, including the deep the component tests was analyzed with was calculated on 63 participants, based
squat, hurdle step, in-line lunge, shoulder a weighted kappa statistic. The weighted on an illness of 1 of the raters on day 2 of
mobility, active straight leg raise, trunk kappa scores were as follows: 80% and testing. Only 15.6% (n = 10) of the par-
stability push-up, and quadruped rotary higher, excellent agreement; from 60% ticipants were identified to be at risk for
stability tests (FIGURE 1). Five of 7 compo- to 79.9%, substantial levels of agreement; injury, based on an FMS composite score
nent tests assess asymmetry by measuring from 40% to 59.9%, moderate agree- of less than or equal to 14 points.
the test bilaterally. If discrepancies exist ment; and below 40%, poor to fair agree- Agreement of the 7 component tests
between the left and right sides, asymme- ment.24 Reliability of the composite test of the FMS (scored 0 to 3) demonstrated
try is noted for that component test and scores was analyzed using intraclass cor- moderate to excellent interrater agree-
the lower of the 2 scores is included in the relation coefficients (ICCs). ICC values of ment (TABLE 3). Specifically, the novice
FMS composite score. In addition to the 0.75 and above represent good reliability, raters demonstrated excellent interrater
7 component tests, the FMS includes 3 those between 0.50 and 0.74 represent agreement on the trunk stability push-
clearing tests that assess for pain: shoul- moderate reliability, and those below up; substantial interrater agreement
der internal rotation and abduction with 0.50 indicate poor reliability.24 Intrarater on the quadruped rotary stability, deep
the hand placed on the opposite shoulder, test-retest reliability was assessed using squat, active straight leg raise, hurdle
lumbar extension performed in the prone an ICC3,1 model, while interrater reliabil- step, and shoulder mobility component
press-up position, and end-range lumbar ity was assessed using an ICC2,1 model. tests; and moderate interrater agreement
flexion in quadruped. Pain on a clearing Response stability of the intrarater and on the in-line lunge. Intrarater (test-re-
test resulted in a score of 0 for the shoul- interrater reliability of the composite test) agreement scores at 48 to 72 hours
der mobility, trunk stability push-up, or scores was calculated using the standard demonstrated substantial agreement on
rotary stability test, respectively. Partici- error of the measurement (SEM) at the the trunk stability push-up, shoulder mo-
pants performed all tests without a pre- 95% level of confidence. The minimal bility, active straight leg raise, deep squat,
participation warm-up. detectable change (MDC95) values at the and in-line lunge component tests; mod-
Each component test was scored on 95% level of confidence were calculated erate agreement on the hurdle step; and
an ordinal scale (0 to 3 points), based to determine error thresholds. Statistical poor agreement on the quadruped rotary

journal of orthopaedic & sports physical therapy | volume 42 | number 6 | june 2012 | 533

42-06 Teyhen.indd 533 5/22/2012 5:38:59 PM

 [ research report ]
less than 1 point, while the MDC95 ranged
Agreement of FMS Component    from 2.1 to 2.5 points on the 21-point
TABLE 3
Scores (0-3 points) scale. The SEM and MDC values were
similar for both intrarater reliability that
Type/Test Percent Agreement κw 95% CI repeated the assessment of the movement
Interrater patterns over a 48-to-72–hour period and
Trunk stability push-up 78 0.82 0.73, 0.90 interrater reliability that had 2 raters as-
Quadruped rotary stability 92 0.77 0.57, 0.96 sess the same movement pattern simul-
Shoulder mobility 86 0.73 0.57, 0.89 taneously. Therefore, one can expect
Active straight leg raise 84 0.69 0.51, 0.87 the error of measurement to be within 1
Deep squat 83 0.68 0.51, 0.85 point across raters and across time, while
Hurdle step 88 0.67 0.45, 0.88 a minimum improvement between 2 and
In-line lunge 68 0.45 0.25, 0.65 3 points on the 21-point scale would be
Intrarater required to demonstrate a real change
Trunk stability push-up 68 0.68 0.55, 0.81 over time.
Quadruped rotary stability 83 0.29 0.05, 0.50 These results are consistent with a
Shoulder mobility 81 0.68 0.53, 0.80 prior publication on FMS reliability.
Active straight leg raise 80 0.60 0.42, 0.74 Minick et al16 reported substantial to ex-
Deep squat 88 0.76 0.63, 0.85 cellent interrater agreement on individ-
Hurdle step 86 0.59 0.42, 0.73 ual FMS component scores when using
In-line lunge 83 0.69 0.48, 0.77 2 novice and 2 expert raters assessing
Abbreviations: CI, confidence interval; FMS, Functional Movement Screen. videotape performance of active college-
age students and varsity athletes. Add-
ing to the literature, our study provides
Reliability of FMS Composite    detailed information on the intrarater
TABLE 4
Scores (0-21 points) and interrater reliability of both FMS
component and composite scores by
Type ICC 95% CI SEM MDC95 randomly assigned novice raters. Spe-
Interrater 0.76 0.63, 0.85 0.92 2.54 cifically, our study utilized 8 entry-level
Intrarater (test-retest) 0.74 0.60, 0.83 0.98 2.07 physical therapy students as raters to
Abbreviations: CI, confidence interval; FMS, Functional Movement Screen; ICC, intraclass correlation collect data prior to their clinical intern-
coefficient; MDC95, minimal detectable change at the 95% level of confidence; SEM, standard error of ship. Additionally, these raters measured
measurement. all movements in real time, without the
benefit of being able to replay a video-
stability component test. test-retest reliability was 0.98 points and tape (the methodology used by Minick et
The interrater reliability (same day) the MDC95 was 2.07 points. al16). The increased number of raters and
of the FMS composite score (scored real-time analysis of movement in mul-
0-21) resulted in an ICC2,1 of 0.76 (95% DISCUSSION tiple participants in our study mimic a
CI: 0.63, 0.85) and was considered good preparticipation screening environment,

T
(TABLE 4). The SEM for interrater reliabil- he FMS has an adequate level of thus enhance the generalizability of the
ity of the composite test was 0.92 points, reliability when assessed in healthy results. Further research is needed to as-
and the MDC95 was 2.54 points on the service members by novice raters. sess the stability of the FMS scores over
21-point scale. Visual representation of The interrater agreement of the FMS longer periods. Ultimately, the reliability
the FMS composite scores between rat- component scores ranged from moder- of this group of novice raters was com-
ers is provided in FIGURE 2. The intrarater ate to excellent, with 6 of the 7 tests cat- parable to previously published research
reliability (test-retest at 48 to 72 hours) egorized as having substantial agreement and provides further support for the FMS
of the FMS composite scores resulted in (κw60%). The intrarater and interrater as a reliable tool to screen in a relatively
an ICC3,1 of 0.74 (95% CI: 0.60, 0.83) and point estimates of the FMS composite diverse, noncollegiate but physically ac-
was considered to be moderate (TABLE 4). score reliability ranged from 0.74 to 0.76, tive population.16
Visual representation of the intrarater with the 95% CIs suggestive of moderate Only 15.6% (n = 10) of the partici-
test-retest FMS composite scores is pro- to good reliability. The SEMs for both pants in this study had an FMS compos-
vided in FIGURE 3. The SEM for intrarater interrater and intrarater reliability were ite score less than or equal to 14 points.

534  |  june 2012  |  volume 42  |  number 6  |  journal of orthopaedic & sports physical therapy

42-06 Teyhen.indd 534 5/22/2012 5:39:00 PM

 Although this may not seem surprising,
given that the participants were relative- Rater 1
ly healthy, it supports the suggestion by 12 13 14 15 16 17 18 19 Total
Cook et al3,4 that FMS scores can identify
altered movement patterns in generally 12 2 1 1 0 0 0 0 0 4
healthy and pain-free participants. Our 13 1 0 1 0 0 0 0 0 2
results are similar to those published by 14 1 2 1 0 0 0 0 0 4
O’Connor et al,21 who found that 10% of

Rater 2
the 874 Marine officer candidates scored
15 0 0 3 4 3 0 0 0 10
less than or equal to 14 points on the 16 0 1 4 1 6 1 0 0 13
FMS. If the initial research that identified 17 0 0 1 3 3 2 3 2 14
the cutoff value were validated, it would
suggest that the FMS would be capable
18 0 0 0 1 1 6 4 0 12
of identifying a subset of individuals at 19 0 0 0 0 0 1 1 1 3
increased risk for time-loss injury within 20 0 0 0 0 0 0 0 1 1
a population of young, healthy service
Total 4 4 11 9 13 10 8 4 63
members. Based on the use of the FMS
for mass screenings (eg, preseason or an-
nual physical examinations), an injury FIGURE 2. Comparison of Functional Movement Screen composite scores between rater 1 and rater 2. Green boxes
indicate agreement (n = 20), yellow boxes indicate a composite-score difference of only 1 point (n = 27), and
prediction screening that could identify
orange boxes indicate a composite-score difference of 2 to 3 points (n = 16).
only 15.6% of the population as having
a high injury risk would allow the asso-
ciated medical staff to prioritize the al-
FMS Day 2 Composite Score
location of limited resources toward the
development of individualized injury 12 13 14 15 16 17 18 19 Total
prevention interventions (eg, corrective 10 1 0 0 0 0 0 0 0 1
exercise prescriptions) for this group.
11 0 0 0 0 0 0 0 0 0
FMS Day 1 Composite Score

However, the validity of the 14-point cut-

off score for this sample cannot be verified 12 1 1 1 0 0 0 0 0 3
in this study, because longitudinal follow- 13 2 1 1 1 0 0 0 0 5
up was not performed to assess actual in-
14 0 1 1 1 2 1 0 0 6
jury rates. Based on the SEM of 1 point
and the MDC95 value between 2.1 and 2.5 15 0 1 2 5 3 1 0 0 12
points, it would be more conservative to 16 0 0 5 0 4 3 1 0 13
use a cutoff score of 15 (based on SEM) or 17 0 0 0 2 4 1 2 0 9
16 to 17 (based on MDC95) to determine
those who may benefit from corrective 18 0 0 1 0 1 2 4 4 12
exercise prescription to help mitigate in- 19 0 0 0 0 0 2 0 0 2
jury risk, until the validity of the 14-point 20 0 0 0 0 0 0 1 0 1
cutoff value can be determined.
One of the limitations noted in the Total 4 4 11 9 14 10 8 4 64
FMS component tests was a restriction
in the range of scores. Specifically, based FIGURE 3. Comparison of FMS composite scores for rater 1 (day 1 to day 2). Green boxes indicate agreement (n =
17), yellow boxes indicate a composite-score difference of only 1 point (n = 26), orange boxes indicate a composite-
on our inclusion/exclusion criteria, no
score difference of 2 to 3 points (n = 20), and the red box indicates a composite-score difference greater than 3
participants scored a 0 on any of the points (n = 1). Abbreviation: FMS, Functional Movement Screen.
FMS component tests, and only 18 of the
446 scored movement patterns resulted the in-line lunge was determined to have score of 3. Compared to the other FMS
in a score of 1; the remaining movement a weighted kappa of 0.45; for this test, no component scores, the in-line lunge and
patterns either received a score of 2 or movements were scored as a 0 or 1. Ad- the quadruped rotary stability had the
3. This restriction in range might have ditionally, only 11 of the 63 paired ratings biggest discrepancy between the percent
reduced the reliability estimates of the had a disagreement, with 25 agreements agreement (68% and 83%, respectively)
FMS component scores. For example, for a score of 2 and 27 agreements for a and weighted kappa (0.45 and 0.29, re-

journal of orthopaedic & sports physical therapy | volume 42 | number 6 | june 2012 | 535

42-06 Teyhen.indd 535 5/22/2012 7:42:05 PM

 [ research report ]
spectively). Interestingly, the lowest levels els of measurement error. The measures Warren, and First Lieutenant Sam Wood.
of agreement between novice raters for of reliability and measurement error were Illustrations for the APPENDIX were provided by
both our study and Minick et al16 involved similar for both intrarater reliability that Elizabeth Holder.
the in-line lunge and quadruped rotary repeated the assessment of the move-
stability tests. Difficulty in performing ment patterns over a 48-to-72–hour pe-
the quadruped rotary stability test (only riod and interrater reliability that had 2 REFERENCES
5 of the 64 participants obtained a score raters assess the same movement pattern
1. Almeida SA, Williams KM, Shaffer RA, Brodine
of 3 on day 1) also limited variability and simultaneously. The interrater agreement SK. Epidemiological patterns of musculoskeletal
potentially intrarater agreement of this of the FMS component scores was good injuries and physical training. Med Sci Sports
measure. Although the restricted range to excellent for the push-up, quadru- Exerc. 1999;31:1176-1182.
might have influenced statistical calcula- ped, shoulder mobility, straight leg raise, 2. Chorba RS, Chorba DJ, Bouillon LE, Overmyer
CA, Landis JA. Use of a functional movement
tions, it is important to point out that rat- squat, hurdle, and lunge. Only 15.6% (n screening tool to determine injury risk in female
er experience and lack of clearly defined = 10) of the participants were identified collegiate athletes. N Am J Sports Phys Ther.
scoring criteria, especially mid-range as being at risk for injury based on previ- 2010;5:47-54.
performance, may have also influenced ously published cutoff values. t 3. Cook G, Burton L, Hoogenboom B. Pre-partici-
pation screening: the use of fundamental move-
results for these select measures. Future ments as an assessment of function - part 1. N
research should determine whether bet- KEY POINTS Am J Sports Phys Ther. 2006;1:62-72.
ter criteria may help to differentiate levels FINDINGS: When using novice raters, the 4. Cook G, Burton L, Hoogenboom B. Pre-partici-
of performance on the quadruped rotary FMS composite scores had moderate to pation screening: the use of fundamental move-
ments as an assessment of function - part 2. N
stability test or to determine the influ- good reliability (ICC = 0.74 and 0.76; Am J Sports Phys Ther. 2006;1:132-139.
ence of removing the quadruped rotary SEM, 1.0 points; MDC95, 2.1 and 2.5 5. Cook G, Burton L, Kiesel K, Rose G, Bryant MF.
stability test on the predictive validity of points) and the FMS component scores Movement: Functional Movement Systems:
the FMS composite score. ranged from moderate to excellent Screening—Assessment—Corrective Strategies.
Aptos, CA: On Target Publications; 2010.
Future study designs should assess the agreement (κw = 0.29-0.82). 6. Cowen VS. Functional fitness improvements
reliability of the FMS using novice raters IMPLICATIONS: The FMS has adequate re- after a worksite-based yoga initiative. J Bodyw
and participants with varying activity liability when assessed in young, healthy Mov Ther. 2010;14:50-54. http://dx.doi.
levels and sport-specific requirements. service members by novice raters over a org/10.1016/j.jbmt.2009.02.006
7. Emery CA, Rose MS, McAllister JR, Meeuwisse
Additional longitudinal studies are also 48-to-72–hour period. WH. A prevention strategy to reduce the in-
required to establish the predictive va- CAUTION: Reliability data must be inter- cidence of injury in high school basketball: a
lidity and optimal cut score for various preted within the context of the sample cluster randomized controlled trial. Clin J Sport
populations. This level of critical investi- tested and the methods used (ie, time Med. 2007;17:17-24. http://dx.doi.org/10.1097/
JSM.0b013e31802e9c05
gation would help to enhance the external between testing for test-retest reliability
8. Goss DL, Christopher GE, Faulk RT, Moore J.
validity of the FMS and to substantiate its estimates). Functional training program bridges reha-
use in the general clinical population, as bilitation and return to duty. J Spec Oper Med.
well as in specific sports settings. Future ACKNOWLEDGEMENTS: This study was done in 2009;9:29-48.
9. Hale SA, Hertel J, Olmsted-Kramer LC. The ef-
research should also determine whether collaboration with research assistants from
fect of a 4-week comprehensive rehabilitation
there is a ceiling effect in the ability of the the University of Texas Health Science Center, program on postural control and lower extremity
FMS to detect change over time. Based Physical Therapy Department, San Antonio, function in individuals with chronic ankle insta-
on the MDC95 of 2.1 to 2.5 points, posi- TX: Mark Bauernfeind, Francis Bisagni, Jor- bility. J Orthop Sports Phys Ther. 2007;37:303-
311. http://dx.doi.org/10.2519/jospt.2007.2322
tive change may not be able to be noted dan Boldt, Cindy Boyer, Cara Dobbertin, Steve
10. Holm I, Fosdahl MA, Friis A, Risberg MA, Mykle-
for individuals who score greater than Elliot, Angela Gass, Germaine Herman, Lacey bust G, Steen H. Effect of neuromuscular train-
18 points at baseline testing. Different Jung, Jake Mitchess, Teddy Ortiz, Kelly Rabon, ing on proprioception, balance, muscle strength,
scoring criteria or cutoff values may be Jason Smith, Megan Swint, Joshua Trock, and and lower limb function in female team handball
players. Clin J Sport Med. 2004;14:88-94.
needed to better differentiate high-end Jerry Yeung. Additional research assistants
11. Jones BH, Cowan DN, Tomlinson JP, Robinson
performance on the FMS. from US Army-Baylor University, Depart- JR, Polly DW, Frykman PN. Epidemiology of
ment of Physical Therapy, US Army Medical injuries associated with physical training among
CONCLUSIONS Department Center and School, San Anto- young men in the army. Med Sci Sports Exerc.
1993;25:197-203.
nio, TX: First Lieutenant Moshe Greenberg,
12. Jones BH, Knapik JJ. Physical training and

A
mong novice raters, the FMS Captain Sarah Hill, First Lieutenant Crys- exercise-related injuries. Surveillance, research
composite score demonstrated tal Straseske, First Lieutenant Sarah Villena, and injury prevention in military populations.
moderate to good interrater and First Lieutenant Christina Yost, First Lieu- Sports Med. 1999;27:111-125.
13. Kiesel K, Plisky P, Butler R. Functional move-
intrarater reliability, and acceptable lev- tenant Kristen Zosel, First Lieutenant Rick

536  |  june 2012  |  volume 42  |  number 6  |  journal of orthopaedic & sports physical therapy

42-06 Teyhen.indd 536 5/22/2012 5:39:02 PM

 ment test scores improve following a stan- and balance training on power, balance, and org/10.1249/MSS.0b013e318223522d
dardized off-season intervention program in landing force in female athletes. J Strength 22. Okada T, Huxel KC, Nesser TW. Relationship
professional football players. Scand J Med Cond Res. 2006;20:345-353. http://dx.doi. between core stability, functional movement,
Sci Sports. 2011;21:287-292. http://dx.doi. org/10.1519/R-17955.1 and performance. J Strength Cond Res.
org/10.1111/j.1600-0838.2009.01038.x 18. Myer GD, Ford KR, McLean SG, Hewett TE. The 2011;25:252-261. http://dx.doi.org/10.1519/
14. Kiesel K, Plisky PJ, Voight ML. Can serious effects of plyometric versus dynamic stabiliza- JSC.0b013e3181b22b3e
injury in professional football be predicted by a tion and balance training on lower extremity bio- 23. Peate WF, Bates G, Lunda K, Francis S,
preseason functional movement screen? N Am J mechanics. Am J Sports Med. 2006;34:445-455.
Bellamy K. Core strength: a new model
Sports Phys Ther. 2007;2:147-158. http://dx.doi.org/10.1177/0363546505281241
for injury prediction and prevention. J Oc-
15. Mandelbaum BR, Silvers HJ, Watanabe DS, 19. Myer GD, Ford KR, Palumbo JP, Hewett TE.
cup Med Toxicol. 2007;2:3. http://dx.doi.
et al. Effectiveness of a neuromuscular and Neuromuscular training improves performance
org/10.1186/1745-6673-2-3
proprioceptive training program in prevent- and lower-extremity biomechanics in female
ing anterior cruciate ligament injuries in athletes. J Strength Cond Res. 2005;19:51-60. 24. Portney LG, Watkins MP. Foundations of Clinical
female athletes: 2-year follow-up. Am J Sports http://dx.doi.org/10.1519/13643.1 Research: Applications to Practice. 2nd ed. Up-
Med. 2005;33:1003-1010. http://dx.doi. 20. National Center for Injury Prevention and Con- per Saddle River, NJ: Prentice Hall; 2000.
org/10.1177/0363546504272261 trol, Centers for Disease Control and Prevention. 25. van Mechelen W. Running injuries. A review
16. Minick KI, Kiesel KB, Burton L, Taylor A, Plisky CDC Injury Research Agenda. Atlanta, GA: US of the epidemiological literature. Sports Med.
P, Butler RJ. Interrater reliability of the func- Department of Health and Human Services; 1992;14:320-335.
tional movement screen. J Strength Cond Res. 2002.
2010;24:479-486. http://dx.doi.org/10.1519/ 21. O’Connor FG, Deuster PA, Davis J, Pappas CG,

@ MORE INFORMATION
JSC.0b013e3181c09c04 Knapik JJ. Functional movement screening:
17. Myer GD, Ford KR, Brent JL, Hewett TE. The predicting injuries in officer candidates. Med Sci
effects of plyometric vs. dynamic stabilization Sports Exerc. 2011;43:2224-2230. http://dx.doi. WWW.JOSPT.ORG

APPENDIX

FUNCTIONAL MOVEMENT SCREEN

Score Criteria Illustration
Deep Squat
3 • Upper torso is parallel with tibia or toward vertical
• Femur below horizontal
• Knees are aligned over feet
• Dowel aligned over feet
2 Performed with heels on 2 × 6-in board
• Upper torso is parallel with tibia or toward vertical
• Femur below horizontal
• Knees are aligned over feet
• Dowel aligned over feet
1 Performed with heels on 2 × 6-in board
• If any of the 4 criteria are not met when the squat is performed with
heels on 2 × 6-in board, the score is 1
0 • Pain during test

journal of orthopaedic & sports physical therapy | volume 42 | number 6 | june 2012 | 537

42-06 Teyhen.indd 537 5/22/2012 5:39:03 PM

 [ research report ]
APPENDIX

Score Criteria Illustration

Hurdle Step (test both right and left sides)*
3 • Foot clears cord (does not touch) and remains dorsiflexed as leg is
lifted over hurdle
• Hips, knees, and ankles remain aligned in the sagittal plane
• Minimal to no movement is noted in lumbar spine
• Dowel and hurdle remain parallel
2 • Alignment is lost between hips, knees, and ankles
• Movement is noted in lumbar spine
• Dowel and hurdle do not remain parallel
1 • Contact between foot and hurdle
• Loss of balance is noted
0 • Pain during test

In-line Lunge (test both right and left sides)*

3 • Knee touches board behind heel
• Dowel and feet remain in sagittal plane
• Dowel contacts remain (head, thoracic spine, sacrum)
• Dowel remains vertical, no torso movement noted
2 • Knee does not touch behind heel
• Dowel and feet do not remain in sagittal plane
• Dowel contacts do not remain
• Dowel remains vertical
• Movement is noted in torso
1 • Loss of balance is noted
• Inability to achieve start position
• Inability to touch knee to board
0 • Pain during test

Active Straight Leg Raise (test both right and left sides)*
3 • Malleolus of tested lower extremity located in the region between
mid-thigh and anterior superior iliac spine of opposite lower extrem-
ity (green region)
• Opposite hip remains neutral (hip does not externally rotate), toes
remain pointing up
• Opposite knee remains in contact with board
2 • Malleolus of tested lower extremity located in the region between
mid-thigh and knee joint line of opposite lower extremity (yellow
region) while other criteria are met
1 • Malleolus of tested lower extremity located in the region below knee
joint line of opposite lower extremity (red region) while other criteria
are met
0 • Pain during test

538  |  june 2012  |  volume 42  |  number 6  |  journal of orthopaedic & sports physical therapy

42-06 Teyhen.indd 538 5/22/2012 5:39:05 PM

 APPENDIX

Score Criteria Illustration

Shoulder Mobility (test both right and left sides)*
3 • Fists are within 1 hand length
2 • Fists are within 1.5 hand lengths
1 • Fists are not within 1.5 hand lengths
0 • Pain during test
Shoulder mobility clearing test: if pain is noted as elbow is lifted,
shoulder mobility is scored as 0

Trunk Stability Push-up

3 • Perform 1 repetition; the thumbs are aligned with forehead for males
and chin for females
• Body is lifted as 1 unit (no sag in lumbar spine)
2 • Perform 1 repetition; the thumbs are aligned with chin for males and
clavicle for females
• Body is lifted as 1 unit (no sag in lumbar spine)
1 • Unable to perform 1 repetition with thumbs aligned with chin for
males and clavicle for females
0 • Pain during test
Extension clearing test: if pain is noted during a prone press-up, push-
up is scored as 0

journal of orthopaedic & sports physical therapy | volume 42 | number 6 | june 2012 | 539

42-06 Teyhen.indd 539 5/22/2012 5:39:06 PM

 [ research report ]
APPENDIX

Score Criteria Illustration

Quadruped Rotary Stability (test both right and left sides)*
3 • 1 unilateral repetition (lift arm and leg from same side of body)
• Keep spine parallel to board
• Knee and elbow touch in line over the board and then return to the
start position
2 • 1 diagonal repetition (lift arm and leg from opposite sides of body)
• Keep spine parallel to board
• Knee and elbow touch in line over the board and then return to the
start position
1 • Inability to perform diagonal repetition
0 • Pain during test
Flexion clearing test: if pain is noted during quadruped flexion, rotary
stability is scored as 0

*For component tests that are scored for both the right and left sides, the lower score is used when calculating the Functional Movement Screen
composite score.

GO GREEN By Opting Out of the Print Journal

JOSPT subscribers and APTA members of the Orthopaedic and Sports
Physical Therapy Sections can help the environment by “opting out” of
receiving the Journal in print each month as follows. If you are:

· A JOSPT subscriber: Email your request to [email protected] or call the

Journal office toll-free at 1-877-766-3450 and provide your name and
subscriber number.
· An APTA Orthopaedic or Sports Section member: Go to www.apta.org
and update your preferences in the My Profile area of myAPTA.
Select “myAPTA” from the horizontal navigation menu (you’ll be asked
to login, if you haven’t already done so), then proceed to “My Profile.”
Click on the “Email & Publications” tab, choose your “opt out”
preferences and save.

Subscribers and members alike will continue to have access to JOSPT

online and can retrieve current and archived issues anytime and anywhere
you have Internet access.

540  |  june 2012  |  volume 42  |  number 6  |  journal of orthopaedic & sports physical therapy

42-06 Teyhen.indd 540 5/22/2012 5:39:07 PM