Functional Movement Screening: Predicting

Injuries in Officer Candidates

FRANCIS G. OCONNOR
1
, PATRICIA A. DEUSTER
1
, JENNIFER DAVIS
1
, CHRIS G. PAPPAS
2
,
and JOSEPH J. KNAPIK
3
1
Uniformed Services University, Consortium for Health and Military Performance, Bethesda, MD;
2
Department of Family
Medicine, Womack Army Community Hospital, Fort Bragg, NC; and
3
Public Health Command (Prov), Aberdeen
Proving Grounds, MD
ABSTRACT
OCONNOR, F. G., P. A. DEUSTER, J. DAVIS, C. G. PAPPAS, and J. J. KNAPIK. Functional Movement Screening: Predicting
Injuries in Officer Candidates. Med. Sci. Sports Exerc., Vol. 43, No. 12, pp. 22242230, 2011. Purpose: Functional movement screening
(FMS) is a musculoskeletal assessment method that incorporates seven movements and yields an overall score based on movement
quality. The objectives of this study were to document the distribution of scores and to determine whether FMS scores could predict
injury in a large military cohort. Methods: A cohort of 874 Marine officer candidates were recruited, consented, completed demographic
questionnaires, and had FMS performed during medical in-processing. Candidates were enrolled in either long-cycle (LC: 68 d; n = 427)
or short-cycle (SC: 38 d; n = 447) training and followed up for injuries occurring in training. Results: The mean FMS score (score
range = 021) among all candidates was 16.6 T 1.7; approximately 10% of candidates had FMS scores e14. A score of e14 on the FMS
predicted any injury with a sensitivity of 0.45 and a specificity of 0.71 and serious injury with a sensitivity of 0.12 and a specificity of
0.94. Both LC and SC cohorts demonstrated higher injury risk among candidates who had scores e14 compared with those with scores
914 (LC: risk ratio (RR) = 1.65, 95% confidence interval = 1.052.59, P = 0.03; SC: RR = 1.91, 95% confidence interval = 1.213.01,
P G 0.01). Overall, 79.8% of persons with scores e14 were in the group with fitness scores G280 (/300), whereas only 6.6% of candidates
in the group with fitness scores Q280 had scores e14. Conclusions: This was the first large-scale study performed in an active-
duty military cohort to examine the utility of FMS during medical in-processing. Further work is warranted to evaluate FMS and
the potential for injury prediction and prevention. Key Words: INJURY PREVENTION, MILITARY PERSONNEL, PHYSICAL
FITNESS, PHYSICAL ACTIVITY
M
usculoskeletal injuries are among the leading
causes of morbidity and mortality in working
aged adults as well as service members (3,15,16).
For example, Cohen et al. (3) recently evaluated the etiology
of medical evacuations from Operation Iraqi Freedom and
Operation Enduring Freedom and found that, of 34,000
evacuations from January 2004 to December 2007, the most
common reason was musculoskeletal disorders (24%); com-
bat injuries were a distant second at 14%. Studies in basic
trainees across services report high injury rates (15,20). In-
juries among 22,000 male recruits undergoing 12 wk of
basic training at Marine Corps Recruit Depot, San Diego,
CA, resulted in more than 53,000 lost training days, which
was estimated to cost more than $16.5 million per year (1).
The morbidity of musculoskeletal injury directly translates
into lost duty days, missed training, early attrition from the
service, and diminished combat effectiveness in theaters of
operation. To reduce the high incidence of injury, multiple
studies have evaluated factors associated with injuries, with
low levels of previous occupational and leisure time physical
activity, previous injury history, high running mileage, low
physical fitness, cigarette smoking, older age, and biome-
chanical factors being major risk factors (2,15,16,20,29,30).
Numerous epidemiologic studies have identified low
physical fitness levels, tobacco use, sedentary lifestyle, and a
history of prior injury as some of the stronger predictors for
future risk of musculoskeletal injury (8,9,20,21,30). Re-
cently, the sports medicine community has become inter-
ested in functional movement and core stability programs
(10,26) because it is generally believed that these programs
may improve fitness and performance and also assist in in-
jury prevention (10,18). Despite this widely held viewpoint,
few large-scale prospective studies have validated the con-
cept that screening or correcting functional deficits will ei-
ther predict or minimize injury or improve performance.
Functional movement screening (FMS) is a series of
movements designed to assess the quality of fundamental
movement patterns and presumably identify an individuals
functional limitations or asymmetries. Previous small studies
Address for correspondence: Francis G. OConnor, M.D., M.P.H., Military
and Emergency Medicine, Uniformed Services University, 4301 Jones Bridge
Road, Bethesda, MD 20814; E-mail: foconnor@usuhs.mil.
Submitted for publication November 2010.
Accepted for Publication April 2011.
0195-9131/11/4312-2224/0
MEDICINE & SCIENCE IN SPORTS & EXERCISE

Copyright 2011 by the American College of Sports Medicine

DOI: 10.1249/MSS.0b013e318223522d
2224
C
L
I
N
I
C
A
L
S
C
I
E
N
C
E
S
Copyright 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
have demonstrated that low FMS scores (e14) are associated
with serious injury in American football players and that
FMS scores can be improved following a standardized in-
tervention (17,18). In addition, a large interventional study
in firefighters suggested that FMS assessment followed by
an 8-wk program to enhance functional movement reduced
time lost to injury by 62% when compared with historical
injury rates (28).
The purpose of the study reported here was to document
the distribution of FMS scores and assess the predictive
value of the FMS by comparing entry scores with subsequent
injury in Marine Corps Officer candidates during Officer
Candidate School (OCS) training. Specifically, we hypothe-
sized that FMS scores would predict injury rates, in partic-
ular, overuse injuries, and that FMS scores would predict
injury better than physical fitness scores. Thus, the study also
examined relationships among FMS, physical fitness, and
injury.
METHODS
Study Design
This project was a prospective cohort study approved by the
institutional review boards at the National Naval Medical
Center and the Uniformed Services University of the Health
Sciences, Bethesda, MD. Before enrollment, all subjects were
thoroughly briefed about the project including benefits and
risks, and those choosing to volunteer provided written in-
formed consent and signed Health Insurance Portability and
Accountability Act authorization forms permitting the use of
protected health information for research.
Subjects
The study participants were male candidates aged 18
30 yr, enrolled in officer candidate training during the sum-
mer of 2009, who gave informed consent. There were two
classes: the first was a 6-wk short-cycle (SC: 38 d) and the
second was a 10-wk-long cycle (LC: 68 d). SC participants
are generally enrolled in collegiate ROTC programs, whereas
LC candidates are not and seek direct military commissions.
Although both LCand SCprograms have comparable training
intensities and volumes with candidates expected to be ex-
tremely fit for successful participation, the training is generally
considered to be somewhat more condensed and intensive
during the SC.
Procedures
Before the study, volunteers were briefed on the FMS,
given a demonstration on the movements, and those inter-
ested volunteered to participate. At 1 or 2 d after the brief-
ing, all candidates underwent medical screening at the OCS
medical facility. The screening updated medical records and
immunizations, as well as ensured that candidates had no
significant changes from their commissioning physical ex-
aminations before beginning the rigorous training prog-
ram. Those with a history of orthopedic injuries and/or
surgery received additional evaluation by the physical ther-
apy section to ensure that they had recovered sufficiently
to undergo the upcoming rigorous training. The FMS and a
survey, which asked about age, tobacco use, exercise his-
tory, and prior injury, were incorporated as part of the me-
dical screening for the volunteers.
FMS. The FMS is a comprehensive screen to assess the
quality of fundamental movement patterns for presumably
identifying an individuals physical limitations or asymme-
tries. The FMS includes seven tests that are scored on a 03
ordinal scale. The seven tests are the squat, hurdle step,
lunge, shoulder mobility, active straight leg raise, push-up,
and rotary stability (5,18). A score of 3 indicates that the
movement was completed as instructed and is free of move-
ment compensation and pain. A score of 2 indicates that the
subject could complete the movement pain-free but with
some level of compensation; a score of 1 indicates that the
subject could not complete the movement as instructed; a
0 is assigned if the subject experiences pain with any por-
tion of the movement. Of the seven tests that comprise the
FMS, five of them (hurdle step, lunge, shoulder mobility,
active straight leg raise, and rotary stability) are performed
and scored separately for the right and left sides of the body.
When assigning a score to a test that incorporates both left
and right sides, the lesser of the two scores is used for a final
event score. Overall FMS scores can range from 0 to 21. To
maximize interrater reliability, all members of the research
team were certified in the FMS before participation in this
project by an instructor from Functional Movement Sys-
temsi. FMS certification involved active participation in
a certified workshop, practical application, and successful
performance on a written examination administered through
Functional Movement Systemsi.
The research team set up a total of nine stations; one de-
signated for check-in, seven stations to conduct the FMS,
and the last for check-out. At the check-in station, volunteers
were rebriefed on the research proposal, ensured an in-
formed consent, and issued an FMS assessment sheet, which
was completed by an evaluator at each of the next seven
stations. Participants would then proceed through the seven
FMS evaluation stations, at which time they were reedu-
cated on successful performance of the station and then
assessed.
Physical fitness test score. Within 1 wk of starting
the training program, candidates completed a physical fit-
ness test. The test consisted of pull-ups to exhaustion, 2-min
abdominal crunch, and a 3-mile run for time, conducted in
that order. Points were assigned to various levels of perfor-
mance on each event (100 points maximum), and a compo-
site score was calculated by summing the three event scores.
Details of the test events and scoring system can be found
in a Marine Corps publication (24).
Injury data. Data on injuries were collected daily
during the training cycle at one medical facility. Medical
care providers who were not part of the investigation saw
FUNCTIONAL MOVEMENT SCREENING AND INJURY Medicine & Science in Sports & Exercise
d
2225
C
L
I
N
I
C
A
L
S
C
I
E
N
C
E
S
Copyright 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
subjects with medical problems and electronically recorded
all medical encounters using the militarys electronic medi-
cal record system (Armed Forces Health Longitudinal
Technology Application [AHLTA]). Physicians who were
part of the research team examined each subjects medical
encounter in AHLTA and determined whether the encounter
was for an injury or for other medical care. For each injury
encounter, the diagnosis was extracted from the AHLTA
record. An injury case was a subject who sustained physical
damage to the body secondary to physical training (11,12)
and sought medical care one or more times during the study
period. Injuries were grouped by type, which was deter-
mined from descriptive information in the medical notes
and by the specific diagnosis. Injury types included 1) over-
use injury, 2) traumatic injury, 3) any injury, and 4) serious
injury. Overuse injuries were presumably due to or related
to long-term repetitive energy exchanges that resulted in
cumulative microtrauma. Specific overuse diagnoses in-
cluded musculoskeletal pain (not otherwise specified), stress
fractures, tendonitis, bursitis, fasciitis, muscle injury pre-
sumably due to overuse (strain), joint injury presumably due
to overuse (sprain), retropatellar pain syndrome, impinge-
ment, degenerative joint conditions, and shin splints. A
traumatic injury was presumably due to sudden energy ex-
changes (acute event), which resulted in abrupt overload and
consequent tissue damage. Traumatic injury diagnoses in-
cluded pain associated with an acute event, muscle injury
(strain) due to an acute event, joint injury (sprain) due to
an acute event, dislocation, fracture, blister, abrasion, lacer-
ation, contusions, and/or closed head injury/concussion.
Any injury was considered a combination of overuse and
trauma diagnoses as described above. The any injury type
included primarily musculoskeletal injuries but also in-
cluded dermatological insults (e.g., blisters, abrasions, lac-
erations). A subject could have experienced both traumatic
and overuse injuries and be counted in both categories, but
only once in any injury. Serious injuries were defined as
any type of injury (traumatic or overuse) that was severe
enough to remove the subject from the training program
(i.e., the individual attrited from training due to injury).
Statistical Analysis
A sample size estimate was performed using nQuery by
assuming an > of 0.05, a power of 80%, and projections
from previous OCS injury data. These criteria and data from
prior studies indicated that a sample of 280 candidates would
be required to detect a 20% difference in injury rates (29).
Questionnaire data, physical fitness scores, injuries, and
FMS scores were analyzed by using Statistical Package for
the Social Sciences version 16.0 (SPSS, Inc., Chicago, IL).
Comparisons between LC and SC subjects were made using
t-tests (for continuous variables) or W
2
(for ordinal, nominal,
or discrete variables).
Cumulative injury incidence was defined as the propor-
tion (%) of candidates who had one or more injuries during
their training cycle (injury incidence = ~candidates with Q1
injury/~all candidates). The injury incidence rate was the
proportion of candidates with one or more injuries divided
by the time at risk (incidence rate = ~candidates with Q1
injury/~time (d) in training for all candidates).
Receiver operating characteristic (ROC) curves were cal-
culated by pairing FMS and PT scores with any, overuse,
and serious injuries.
W
2
statistics were used to evaluate differences in injury
risk among those with FMS scores above and below the cut
point and by physical fitness test scores. A W
2
for person
time was used to compare injury incidence rates. Because
the LC and SC subjects had different exposure times, and
similar, although somewhat unique training programs, as-
sociations between FMS and injury were evaluated sepa-
rately in these groups.
RESULTS
A total of 874 male candidates consented, completed
questionnaires, and had the FMS performed during medical
in-processing. Of the 874, 427 volunteers were in the LC
and 447 volunteers were in the SC.
TABLE 1. Characteristics of LC and SC participants.
Variable LC SC P
Age (yr), mean T SD 23.0 T 2.6 21.7 T 2.6 0.001
Physical fitness score, mean T SD 263 T 23.2 268 T 22.2 0.001
Smokers, n (%) 66 (15.6) 78 (17.5) 0.46
Exercise history Q5 wk
j1
, n (%) 290 (68.9) 306 (69.5) 0.60
Prior lower limb injury, n (%) 167 (40.0) 189 (43.3) 0.33
No return to normal activity after injury, n (%) 163 (38.7) 179 (41.1) 0.20
TABLE 2. Cumulative injury incidence and injury incidence rate.
LC SC P
Cumulative injury incidence
(% injured)
Any injury 35.2 27.5 0.02
Overuse injury 12.6 8.2 0.03
Traumatic injury 26.6 20.7 0.04
Serious injury 9.0 4.3 G0.01
Injury incidence rate
(injuries/1000 person-days)
Any injury 6.2 7.9 0.03
Overuse injury 2.2 2.4 0.43
Traumatic injury 4.7 5.9 0.05
Serious injury 1.6 1.2 0.37
FIGURE 1Distribution of FMS scores by cycle length (LC vs SC)
expressed as a percent of sample.
http://www.acsm-msse.org 2226 Official Journal of the American College of Sports Medicine
C
L
I
N
I
C
A
L
S
C
I
E
N
C
E
S
Copyright 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Questionnaire variables. Characteristics of the LC and
SC groups indicated the groups were comparable (Table 1).
Compared with the SC group, the LC group was about 1 yr
older and scored four fewer points on the fitness test than
the SC group. However, no significant differences were no-
ted in baseline exercise frequency, smoking history, or func-
tional return from a prior injury.
Cumulative injury incidence and injury incidence
rates. Because of the longer training cycle and longer ex-
posure time, the LC candidates had significantly higher cu-
mulative injury incidences for any, overuse, traumatic, and
serious injuries (Table 2). When the groups were compared
as a function of injuries per 1000 person-days, the SC can-
didates had higher injury incidence rates for any and trau-
matic injuries; the two groups did not differ on overuse or
serious injury incidence rates.
FMS distribution. The mean FMS score among all
candidates was 16.6 T 1.7 with a range of 621. The mean
FMS scores for the LC and SC were 16.5 T 1.5 and 16.7 T
1.9, respectively (P = 0.07) (Fig. 1). Only 0.2% of can-
didates had a perfect score of 21, and only 0.2% had scores
e10. The most frequent score among candidates was 17
(23%). Figure 2 presents the distribution of scores for each
movement: 8.4% and 12.8% of the LC and SC, respec-
tively, had scores e14, and 25.3% and 36.7% of the LC
and SC, respectively, had scores Q18. Shoulder mobility was
the movement with the highest frequency of 1 as a score
(7.6%), and push-ups was the movement with the highest
frequency of 3 as a score (84.5%). The squat, hurdle step,
and rotary stability were the movements with the highest
frequency of 2.
FMS and injuries. In the SC cohort, candidates with an
FMS score of e14 had a 1.91 times (95% confidence inter-
val (CI) = 1.213.01, P G 0.01) higher any injury incidence
rate compared with a score 914 (Table 3, upper panel).
Candidates in the LC group were 1.65 times more likely
to sustain an injury with an FMS score e14 (95% CI = 1.05
2.59, P = 0.03), compared with those with a score 914.
When the LC and SC groups were combined, the relative
risk was 1.5 times greater for any injury with a FMS score
e14 (P = 0.003): 45.8% of persons with scores e14 suf-
fered an injury compared with 30.6% of those with scores
914. In contrast, FMS scores were not associated with the
incidence of overuse injuries. Overall, 12.5% of persons
with scores e14 had overuse injuries compared with 10.6%
with scores 914 (P = 0.6). Mean FMS scores of 16.7 T 1.7
for those who had no injury were comparable to those
who suffered any injury (16.7 T 1.8).
Although the intent of the study was to dichotomize the
data, as has been done in the past, we found that cumula-
tive injury incidence was higher at FMS scores of 18 (LC =
46.7%, n = 75; SC = 32.2%, n = 90) compared with FMS
scores of 17 (LC = 27.6%, n = 116; SC = 18.1%, n = 94).
Thus, injury risk of LC and SC groups was examined
according to FMS categories of e14, 1517, and Q18. When
grouped in this way, and using the 1517 as the reference,
the risk of injury was significantly higher in the e14 group,
as before, but also significantly higher for the Q18 category
for the LC group, which suggests a bimodal distribution
(Table 3, lower panel).
FMS cut point determination. ROC curves were de-
veloped for overuse, serious, and any injury. Analysis of the
ROC curves yielded areas under the curve of 0.58, 0.52, and
0.53 for any, overuse, and serious injuries, respectively. No
ROC curve provided a point that maximized specificity and
sensitivity. Table 4 presents the odds ratios (OR), CI, sen-
sitivity, and specificity for the dichotomized FMS score
(e14 vs 914) as a predictor of injury. In addition, we exam-
ined whether a combination of only four movements would
yield comparable ORs: we chose the deep squat, shoulder
mobility, active straight leg, and hurdle step because they had
the highest frequency of scores less than 3.
FIGURE 2Distribution of individual FMS movements expressed as
a percent of sample scoring 1, 2, or, 3.
TABLE 3. Injury rates by cycle length: incidence rates among those with FMS e14 and 914 and percent injured and risk ratios among those with FMS scores e14, 1517, and Q18.
Cycle Score n Total Time in Training (d) Injured (n) Injury Rate per 1000 Person-Days Risk Ratio, e14 / 914 (95% CI) P
LC e14 36 1961 19 9.69 1.65 (1.052.59) 0.03*
914 391 22,280 131 5.89
SC e14 57 1714 23 13.42 1.91 (1.213.01) G0.01*
914 390 13,822 97 7.02
Cycle Score n Injured (%) Risk Ratio 95% CI P
LC e14 36 52.8 1.76 1.232.51 0.001*
1517 283 29.3 1.0
Q18 108 44.4 1.61 1.242.08
SC e14 57 40.4 1.88 1.282.75 0.015*
1517 223 22.2 1.0
Q18 166 28.9 1.32 0.951.84
* Statistically significant result.
FUNCTIONAL MOVEMENT SCREENING AND INJURY Medicine & Science in Sports & Exercise
d
2227
C
L
I
N
I
C
A
L
S
C
I
E
N
C
E
S
Copyright 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Physical fitness scores, FMS scores, and
injury. The relationship between physical fitness (PT) and
FMS scores was also examined. PT scores were dichotomized
as Q280 (high fitness) versus G280 (moderate fitness): the
best possible score was 300 and 33.3% of the sample had
scores Q280. Analysis of the ROC curves for PT yielded
areas under the curve comparable to FMS scores0.57,
0.53, and 0.52 for any, overuse, and serious injuries, re-
spectively. Again, no ROC curve provided a point that
maximized specificity and sensitivity. Table 4 presents the
association between fitness, FMS scores, and injuries. As
shown candidates with PT scores G280 were 2.2 times more
likely to have FMS scores e14 and significantly more likely
to sustain an injury across all types of injuries. No biomodal
distribution was noted for PT scores when classified by
quintiles (data not shown).
DISCUSSION
Because musculoskeletal injuries are so common and as-
sociated with significant morbidity, clinicians and research-
ers have been seeking approaches for identifying those as
highest risk. To date, no large prospective study using FMS
had been conducted. In the present study, we applied the
FMS, which has been used to identify functional imbalances
and weaknesses in football players, to candidates undergo-
ing training to determine whether scores would predict in-
jury. The risk of any injury was 2.0 times higher among
those with FMS scores e14, but it remains to be determined
whether treating the identified imbalances would have pre-
vented those injuries. Although the limited predictability
might reflect the low percentage of persons with scores e14,
this cannot be determined from the present data. Impor-
tantly, PT scores were just as predictive of future injury as
FMS scores and had a higher sensitivity.
The FMS, first introduced in 2001 by Cook (4) and de-
scribed in more detail in 2006 (6,7), has altered the paradigm
of screening for static biomechanical deficiencies: com-
prehensive functional movements and core stability are as-
sessed to establish an individuals functional platform. Core
stability may serve a role in injury prediction and preven-
tion (13,22,23,33), but its role in predicting injury or im-
proving performance is questionable (13,27). Limited data
are available on the use of the FMS for screening, but the
interrater reliability of the FMS has been established in re-
cent studies, with weighted J values ranging from 0.45
to 1.00 (25,32). Preliminary studies with FMS have dem-
onstrated injury predictability in a small number of NFL
football players (17,28). Kiesel et al. (18) retrospectively
analyzed the relationship between FMS scores for National
Football League (NFL) football players and the likelihood
of serious injury. FMS scores were obtained before the start
of the season for 46 NFL players, and a score of e14 was
found to positively predict serious injury with a specificity
of 0.91 and sensitivity of 0.54; the odds of sustaining a se-
rious injury was 11.7 times higher in those with an FMS
score e14 compared with those with a score 914. Kiesel
et al. (18) also noted lower scores among those who had
been injured (14.3 T 2.3) compared with those without in-
jury (17.4 T 3.1). In the present study, when we compared
entry FMS scores by no injury versus any injury, the scores
were the same (16.7 T 1.7) and the OR for sustaining a se-
rious injury was 2.0 (95% CI = 1.03.8); the sensitivity and
specificity were 0.19 and 0.90, respectively. Interestingly,
two groups have reported that FMS did not predict in-
jury: one study was with 60 marathon runners (14) and an-
other was on 112 basketball players (31). Hoover et al. (14)
reported an 8.3% sensitivity and 94.5% specificity for mar-
athon runners, whereas Sorensons (31) data yielded a sen-
sitivity and specificity of 53.8% and 52.3%, respectively, for
basketball players. The low sensitivity is problematic be-
cause a sensitivity above 50% is desirable so those predis-
posed to injury can be identified early on and potentially
rehabilitated before injury. Although specificity seems to be
high, this is in large part explained by the small proportion
of the cohort with scores e14.
One major difference between our study and those of
Kiesel et al. (17,18) was the percent with scores e14. In one
study, 21.7% of the football players had low scores com-
pared with only 10.6% of candidates (18). In another study,
Kiesel et al. (17) found that 90.1% of football players had
scores e14. It is possible that the low percentage of scores
e14 limited the predictive value of our data. One other im-
portant observation from the present study was that, although
TABLE 4. Percent of persons incurring injuries, with OR, CI, and sensitivity/specificity of predicting Any, Overuse, and Serious Injuries by FMS Scores for Seven and Four Movements and
PT scores.
FMS Score: 7 Movements 914 (%) e14 (%) OR (CI; P) Sensitivity Specificity
Any injury 29.2 45.2 2.0 (1.33.1; 0.002) 45.2 78.2
Overuse injury 9.9 12.9 1.4 (0.712.6; 0.35) 12.9 90.1
Serious injury 6.1 11.8 2.0 (1.04.1; 0.05) 11.8 93.9
FMS Score: 4 Movements
a
96 e6 OR (CI; P) Sensitivity Specificity
Any injury 29.7 41.5 1.5 (0.992.35; 0.06) 41.4 70.3
Overuse injury 9.9 12.8 1.2 (0.642.29; 0.55) 12.7 90.1
Serious injury 6.1 11.7 2.2 (1.144.2; 0.018) 11.7 93.8
PT Scores Q280 G280 OR (CI; P) Sensitivity Specificity
Any injury 21.8 35.6 2.1 (1.52.9; 0.000) 35.8 78.2
Overuse injury 6.0 11.9 2.4 (1.34.3; 0.003) 11.8 94.1
Serious injury 2.5 5.9 2.5 (1.15.7; 0.01) 5.9 97.5
a
Movements included deep squat, shoulder mobility, active straight leg, and hurdle step.
http://www.acsm-msse.org 2228 Official Journal of the American College of Sports Medicine
C
L
I
N
I
C
A
L
S
C
I
E
N
C
E
S
Copyright 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
no single test alone was significantly associated with inju-
ries, combinations of selected movements yielded ORs
comparable to those for the seven tests. Likewise, although
asymmetries were assessed and recorded, no statistical
evidence supported asymmetry as a risk factor for injury
in this cohort. However, correcting asymmetries and reme-
diating problematic movements might be essential for any
intervention to mediate risk of injury.
Several studies have investigated whether using FMS
scores to dictate treatment for identified weaknesses is ef-
fective. Peate et al. (28) assessed core strength and flexibil-
ity in 443 firefighters and found lost time to and the number
of injuries was reduced by 62% and by 42%, respectively,
after an intervention to improve flexibility and strength in
those with FMS scores G17. In addition, Goss et al. (10)
used FMS scores to dictate training, and after 6 wk, perfor-
mance on functional tests and FMS scores improved sig-
nificantly, but no improvement in pain scores was noted.
Kiesel et al. (17) examined the effect of a 7-wk off-season
intervention program on changes in the number of asym-
metries and FMS scores in football players. After interven-
tion, 51.6% had scores 914 compared with the 9.9% at
preintervention and the number of asymmetries declined by
35%. However, 20 subjects undergoing the intervention
failed to improve their score enough to exceed the thresh-
old of 14. The authors concluded that, although a standard-
ized intervention did change overall movement patterns,
further research would be necessary to determine whether
changes in movement patterns translated into reduced in-
jury risk (17). The high percentage of persons in their study
with scores e14 is striking, given only 10% of our volun-
teers had low scores.
Among our SC and LC groups, no significant differences
in injury risk factors were noted, to include baseline exer-
cise frequency, smoking history, or functional return from a
prior injury. Whereas the LC groups scored slightly higher
(about 1%) on the PT test, this small difference was unlikely
to have contributed to injury risk. However, scores on the
PT test were significantly associated with injury, such that
those with high PT scores were significantly less likely to
suffer any, overuse, or serious injuries compared with those
with lower PT scores, which is consistent with previous
military studies (15,16,19,20,29). In addition, PT scores
were comparable to FMS scores about predicting injury:
the OR for PT scores ranged from 2.1 to 2.5 and combining
the two scores did not significantly improve the prediction.
Although the ability to predict injury in this population
seems to be comparable between PT and FMS scores, the
FMS offers the potential advantage of a rehabilitative inter-
vention. However, further study will be needed to address
the higher injury rates in persons with scores Q18.
Several factors should be considered in interpreting the
data from the present study. First, our sample of candidates
represents highly fit young men who have been previously
challenged and screened in the Marine Corps and, as such,
represent a relatively homogeneous population. In the pres-
ent study, the average FMS score was 16.7 T 1.7, whereas
the average score was only 12.6 T 2.1 in one NFL study (17).
However, mean scores of 16.7 T 3.0 (18) and 15.1 (10) have
also been reported for football players and military person-
nel, respectively. Another consideration is the distribution
of scores. In our study, 10.3% had scores e14 with only
1.3% of scores e12; among the NFL football players stud-
ied, 22% had scores e14 such that the distribution of their
scores was skewed to the left (17). Clearly, more work will
be needed with different populations.
In summary, FMS scores e14 were associated with in-
creased injury risk, although the sensitivity was low. None-
theless, data from this study suggest that further investigations
are warranted. Future studies should seek to evaluate a mili-
tary cohort entering basic training because they are likely to
represent a more heterogeneous population than Marine can-
didates. In addition, the present study has demonstrated that
FMS screening can be accomplished as part of medical in-
processing.
The authors acknowledge that this research was performed un-
der a research grant award from the American Medical Society
for Sports Medicine.
The authors report no conflict of interest.
The authors thank Dr. Bruce Jones from the US Army Public
Health Command for his insights, careful reading, and assistance
with the statistics; Drs. Megan Raleigh and Devin McFadden for
screening of the medical records; and Mr. Tyson Grier for his assis-
tance with constructing the database. The authors also thank COL
Richard Mancini, MAJ Brad Kroll, and Mr. Jess Vera Cruz from Offi-
cer Candidate School, Quantico, VA, as well as Francesca Cariello,
PhD, RN, CCRN, and Richard Blumling, MSN, CDR, USN, from the
Naval Health Clinic Quantico. Finally, the authors also thank Mr.
Brian McGuire, Marine Corps Training and Education Command, for
his assistance in facilitating this project at Quantico.
The results of this study do not constitute endorsement of func-
tional movement screening by the authors, the US Department of
Defense, or the American College of Sports Medicine.
REFERENCES
1. Almeida S, Williams K. A Physical Training Program to Reduce
Musculoskeletal Injuries in U.S. Marine Corps Recruits. San
Diego, CA: Center NHR, National Technical Information Service,
US Department of Commerce; 1997. p. 89.
2. Almeida SA, Williams KM, Shaffer RA, Brodine SK. Epidemi-
ological patterns of musculoskeletal injuries and physical training.
Med Sci Sports Exerc. 1999;31(8):117682.
3. Cohen SP, Brown C, Kurihara C, Plunkett A, Nguyen C, Strassels
SA. Diagnoses and factors associated with medical evacuation
and return to duty for service members participating in Operation
Iraqi Freedom or Operation Enduring Freedom: a prospective co-
hort study. Lancet. 2010;375(9711):3019.
4. Cook G. Athletic body balance. In: Athletic Body Balance.
Champaign (IL): Human Kinetics; 2001. p. 232.
5. Cook G, Burton L. The Functional Movement Screen [Internet].
www.performbetter.com; [cited 2010 Oct 25]. Available from: http://
www.functionalmovement.com/SITE/publications/fmscreening.
php.
FUNCTIONAL MOVEMENT SCREENING AND INJURY Medicine & Science in Sports & Exercise
d
2229
C
L
I
N
I
C
A
L
S
C
I
E
N
C
E
S
Copyright 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
6. Cook G, Burton L, Hogenboom B. The use of fundamental
movements as an assessment of function part 1. N Am J Sports
Phys Ther. 2006;1(2):6272.
7. Cook G, Burton L, Hogenboom B. Pre-participation screening:
the use of fundamental movements as an assessment of function
part 2. N Am J Sports Phys Ther. 2006;1(3):1329.
8. Craig BN, Congleton JJ, Kerk CJ, Lawler JM, McSweeney KP.
Correlation of injury occurrence data with estimated maximal aer-
obic capacity and body composition in a high-frequency manual
materials handling task. Am Ind Hyg Assoc J. 1998;59(1):2533.
9. Ekstrand J, Gillquist J. Soccer injuries and their mechanisms:
a prospective study. Med Sci Sports Exerc. 1983;15(3):26770.
10. Goss DL, Christopher GE, Faulk RT, Moore J. Functional training
program bridges rehabilitation and return to duty. J Spec Oper
Med. 2009;9(2):2948.
11. Haddon W Jr. On the escape of tigers: an ecologic note. Am J
Public Health Nations Health. 1970;60(12):222934.
12. Haddon W Jr. The changing approach to the epidemiology, pre-
vention, and amelioration of trauma: the transition to approaches
etiologically rather than descriptively based. 1968. Inj Prev. 1999;
5(3):2315.
13. Hibbs AE, Thompson KG, French D, Wrigley A, Spears I. Opti-
mizing performance by improving core stability and core strength.
Sports Med. 2008;38(12):9951008.
14. Hoover D, Killian CB, Bourcier B, Shannon L, Jenny T, Willis R.
Predictive validity of the Functional Movement Screeni in a
population of recreational runners training for a half marathon.
Med Sci Sports Exerc. 2008;40(5):S219.
15. Jones BH, Canham-Chervak M, Canada S, Mitchener TA,
Moore S. Medical surveillance of injuries in the U.S. Military de-
scriptive epidemiology and recommendations for improvement.
Am J Prev Med. 2010;38(1 suppl):S4260.
16. Kaufman KR, Brodine S, Shaffer R. Military trainingrelated in-
juries: surveillance, research, and prevention. Am J Prev Med.
2000;18(3 suppl):5463.
17. Kiesel K, Plisky P, Butler R. Functional movement test scores
improve following a standardized off-season intervention program
in professional football players. Scand J Med Sci Sports. 2011;
21(2):28792.
18. Kiesel K, Plisky P, Voight M. Can serious injury in professional
football be predicted by a preseason functional movement screen?
N Am J Sports Phys Ther. 2007;2(3):14750.
19. Knapik JJ, Darakjy S, Hauret KG, et al. Increasing the physical
fitness of low-fit recruits before basic combat training: an evalua-
tion of fitness, injuries, and training outcomes. Mil Med. 2006;
171(1):4554.
20. Knapik JJ, Sharp MA, Canham-Chervak M, Hauret K, Patton JF,
Jones BH. Risk factors for training-related injuries among men
and women in basic combat training. Med Sci Sports Exerc. 2001;
33(6):94654.
21. Knapik JJ, Trone DW, Swedler DI, et al. Injury reduction effec-
tiveness of assigning running shoes based on plantar shape in ma-
rine corps basic training. Am J Sports Med. 2010;38(9):175967.
22. Lederman E. The myth of core stability. J Bodyw Mov Ther.
2010;14(1):8498.
23. Leetun DT, Ireland ML, Willson JD, Ballantyne BT, Davis IM.
Core stability measures as risk factors for lower extremity injury
in athletes. Med Sci Sports Exerc. 2004;36(6):92634.
24. Headquarters Marine Corps. Marine Corps Physical Fitness Test and
Body Composition Program Manual. Washington, DC. 2002. 131 p.
25. Minick KI, Kiesel KB, Burton L, Taylor A, Plisky P, Butler RJ.
Interrater reliability of the functional movement screen. J Strength
Cond Res. 2010;24(2):47986.
26. Mottram S, Comerford M. A new perspective on risk assessment.
Phys Ther Sport. 2008;9(1):4051.
27. Okada T, Huxel KC, Nesser TW. Relationship between core
stability, functional movement, and performance. J Strength Cond
Res. 2011;25(1):25261.
28. Peate WF, Bates G, Lunda K, Francis S, Bellamy K. Core strength:
a new model for injury prediction and prevention. J Occup Med
Toxicol. 2007;2:3.
29. Piantanida NA, Knapik JJ, Brannen S, OConnor F. Injuries dur-
ing Marine Corps officer basic training. Mil Med. 2000;165(7):
51520.
30. Sacks JJ, Nelson DE. Smoking and injuries: an overview. Prev
Med. 1994;23(4):51520.
31. Sorenson EA. Functional movement screen as a predictor of in-
jury in high school basketball athletes [dissertation]. Eugene (OR):
University of Oregon; 2009. p. 89.
32. Teyhen DS, Donofry DF, Shaffer SW, et al. Functional move-
ment screen: a reliability study in service members. US Army Med
Dep J. 2010;33(7):71.
33. Willson JD, Dougherty CP, Ireland ML, Davis IM. Core stability
and its relationship to lower extremity function and injury. J Am
Acad Orthop Surg. 2005;13(5):31625.
http://www.acsm-msse.org 2230 Official Journal of the American College of Sports Medicine
C
L
I
N
I
C
A
L
S
C
I
E
N
C
E
S
Copyright 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.