Physical Therapy in Sport 46 (2020) 204e213

Contents lists available at ScienceDirect

Physical Therapy in Sport

journal homepage: www.elsevier.com/ptsp

Original Research

Can musculoskeletal screening test findings guide interventions for

injury prevention and return from injury in field hockey?
Mark J. Stokes a, b, *, Jeremy Witchalls a, Gordon Waddington a, Roger Adams a
a
Research Institute for Sport and Exercise, University of Canberra, 11 Kirinari Street, Bruce, ACT, 2617, Australia
b
Centre of Excellence Queensland Academy of Sport, 400 Kessels Road, Nathan, Qld, 4111, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Objectives: The aim of this study is to identify if intrinsic factors tested in the preseason screening (PSS)
Received 23 June 2020 can identify an elevated risk of injury. This aim has two aspects; to assess whether previous injury is
Received in revised form associated with ongoing deficits in performance, and to assess if the PSS can identify differences in
9 September 2020
intrinsic factors that profile risk of future injury.
Accepted 10 September 2020
Design: A cohort of state level field hockey players were tested on a screening test battery including
proprioception, postural stability, muscular strength and range of motion, to establish if these intrinsic
factors were useful in identifying elevated risk of injury. Retrospective injury data was collated to
determine association with previous injury and prospective injury data was collated to determine as-
sociation with future injury.
Participants: A total of 130 field hockey players were included in this study, from state level squads
(age ± SD ¼ 20.96 (3.75); height ¼ 176.09 cm). Groups for prescreening and post screening injury status
(injured/not injured) were established for comparison to screening test results.
Results: Right Active Movement Extent Discrimination Assessment (AMEDA), left AMEDA and right Y-
balance test (YBT) anterior direction (Ant) were significantly associated (p < 0.05) with injury prior to
screening. Right YBTAnt and right and left hip internal rotation (IR) were significantly associated
(p < 0.05) with injury post screening. The YBTAnt and YBT posteromedial (PMed) reach directions and
Hip IR are associated with previous hamstring injury and show a difference between post screening
injured and non-injured groups.
Conclusions: AMEDA, R YBTAnt, Hip IR tests should be a focus for recovery after previous injury and
during season preparation. Full recovery may improve readiness to return to play and reduce risk of
primary injury or re-injury. YBTAnt and YBTPmed and Hip IR show a performance deficit link between
previous injury and subsequent re-injury of hamstrings. Since these are the most common re-injury
types in this cohort, these tests are clinically useful in informing return to play decisions for hockey
players.
© 2020 Elsevier Ltd. All rights reserved.

Funding statement 1. Introduction

Mark Stokes was funded by the Queensland Government Field hockey requires speed, agility and the ability to access a
through the Queensland Academy of Sport for the duration of this ground-level ball with a hockey stick and therefore places consid-
research. No other funding was associated with this study. erable load on the legs. Lower limb injuries are frequently seen in
the injury distribution among field hockey athletes (Lindgren &
Maguire, 1985; Rishiraj, Taunton, & Niven, 2009; Rose, 1981;
Theilen, Mueller-Eising, Bettink, & Rolle, 2015). Preseason testing is
routinely performed in hockey squads, and one of the aims of the
battery of tests is to profile the risk of injuries (McCall, Fanchini, &
* Corresponding author. Research Institute for Sport and Exercise, University of Coutts, 2017). The risk of injury profile has two possible categories:
Canberra, 11 Kirinari Street, Bruce, ACT, 2617, Australia. the prediction of risk of future injury for a previously uninjured
E-mail address: [email protected] (M.J. Stokes).

https://doi.org/10.1016/j.ptsp.2020.09.009
1466-853X/© 2020 Elsevier Ltd. All rights reserved.
M.J. Stokes, J. Witchalls, G. Waddington et al. Physical Therapy in Sport 46 (2020) 204e213

athlete and the risk of subsequent injury for a previously injured A general finding in team sports has been that previous injury
athlete. The present study assessed a battery of existing tests of can be a risk factor for subsequent injury (Arnason et al., 2004;
intrinsic factors to evaluate their usefulness as possible compo- Brockett, Morgan, & Proske, 2004; Ekstrand & Gillquist, 1982;
nents of risk of injury profiles. Freckleton & Pizzari, 2012; H€agglund, Walde n, & Ekstrand, 2006; J.;
Witchalls, Blanch, Waddington, & Adams, 2011).Witchalls et al. (J.
1.1. Predictors of lower limb injury Witchalls et al., 2011) note that preventing the first injury may be
advantageous in injury prevention. Following injury, a player can
Preseason testing in sport is often referred to as screening, carry a functional deficit and this deficit may predispose them to re-
however the process is a combination of screening for existing injury (Waddington & Adams, 1999; J. B.; Witchalls, Waddington,
injury and profiling for potential or future injury. The present study Adams, & Blanch, 2014). Injury history and gender can affect the
uses the term preseason screening (PSS) with the understanding functional movement screening and also the performance of the Y
that it also provides profiling. Varied screening tools have been balance test (YBT) (Chimera, Smith, & Warren, 2015). Both injury
shown to be predictors of lower limb injuries in team sports history and gender therefore need to be considered in injury
(Dallinga, Benjaminse, & Lemmink, 2012). These tools include tests research.
for intrinsic performance characteristics such as balance, strength, Following injury, return-to-play decisions are often made rela-
and range of motion. Hockey is a running and reaching sport, so tive to game performance. History of injury as a risk factor suggests
tests that measure single leg range of motion, control, balance and that even though players have returned to play, a deficit may exist
proprioception, singly or in combination, may be relevant. Evalu- that increases the risk of second injury. To date, this has not been
ations of results from PSS in other sports show association with risk considered in field hockey. Therefore, for the PSS to provide infor-
of injury, including outcomes of tests of single leg squat (Ugalde, mation regarding risk of injury, tests that identify deficits following
Brockman, Bailowitz, & Pollard, 2015), lunge test (Gabbe, Finch, a previous injury would be beneficial.
Wajswelner, & Bennell, 2004), somatosensory skill (propriocep-
tion) using the active movement extent discrimination assessment 1.2. Benefits of screening
(AMEDA) (Cameron, Adams, & Maher, 2003; Han, Anson,
Waddington, Adams, & Liu, 2015; Steinberg et al., 2019), and the The benefit of screening in injury prediction has been ques-
star excursion balance test (SEBT) (Munro & Herrington, 2010; tioned (Bahr, 2016). Further, it has been suggested that there needs
Plisky et al., 2009; Plisky, Rauh, Kaminski, & Underwood, 2006). to be an awareness of the difference between association of a test
Due to the mixed models of screening tests employed in different with an outcome and prediction of the outcome (McCall et al.,
studies, a definitive individual injury risk profile for all lower ex- 2017). Research can therefore be explanatory and clinically orien-
tremity injuries is difficult to establish (Gabbe, Finch, et al., 2004). tated or can be targeted towards prediction utilizing appropriate
There is thus a need for further research with sport specific tests on processes (McCall et al., 2017). Team sports medical staff have a
sport specific common injuries. dual nature to their roles; in both providing immediate interven-
Preliminary analysis of the injury data of the current field tion, and supporting evidence-based research (McCall et al., 2016).
hockey cohort shows that ankle and hamstring injuries are the Screening therefore has a dual function - to provide information on
most frequent. In other physically active sport environments, there immediate possible interventions, and to build the research back-
is evidence for balance tests as a predictor of ankle injuries ground to inform longer term prediction of risk of injury (McCall
(Dallinga et al., 2012) as well as plantarflexion strength asymme- et al., 2016). This can allow team stakeholders to make better de-
tries, running speed, cardiorespiratory endurance, balance, dorsi- cisions on interventions in the daily training environment even if
flexion strength, coordination, muscle reaction, and dorsiflexion they are not predictive. With this in mind this study aims to provide
range of motion (Milgrom et al., 1991; Willems et al., 2005). Using evidence that both contributes to the profile of risk of injury and
the AMEDA, somatosensory acuity has also been shown to be a informs the immediate clinical practice.
valid and reliable measure of differences between injured and non- It is unrealistic in the team sport environment to test all possible
injured ankles (Steinberg et al., 2019). Proprioceptive training, risk factors. Providing useful tests that can contribute to decision
compared with no intervention, has been shown to be an effective making on injury prevention is still beneficial. Further, screening
strategy to reduce the rate of re-occurrence of ankle sprains among can identify tests that can be combined to provide a model that
male soccer players (Mohammadi, 2007). identifies players with a higher risk of injury and in turn informs
In addition to the overall frequency of hamstring injury, there is clinical decisions.
a clinical impression that hamstring re-injury is also common. The aim of this study is to identify if intrinsic factors tested in
Hamstring muscle strains are strongly associated with a history of a the preseason screening can identify an elevated risk of injury
posterior thigh injury (Verrall, Slavotinek, Barnes, Fon, & Spriggins, either individually or in combination. This aim has two aspects; to
2001). Other risk factors for a hamstring muscle strain include assess whether previous injury is associated with ongoing deficits
increasing age, being of Australian Aboriginal descent, having a past in performance during the PSS tests and to assess if the PSS can
history of serious knee injury, and having a past history of osteitis identify differences in intrinsic factors that profile risk of future
pubis (Verrall et al., 2001). Players with increased quadriceps injury (not just following previous injury).
flexibility were found to be less likely to sustain a hamstring injury
(Gabbe, Finch, Bennell, & Wajswelner, 2005). Hip ROM has been 2. Method
associated with a risk of thigh muscle adductor (Ibrahim, Murrell, &
Knapman, 2007) and hamstring injury in soccer players A cohort of state level field hockey players were tested to
(Henderson, Barnes, & Portas, 2010) and in professional soccer establish whether tests of intrinsic performance factors were useful
players, strength asymmetries have been associated with a higher in identifying elevated risk of injury. Retrospective injury data was
risk of sustaining hamstring strains (Fousekis, Tsepis, & Vagenas, collated to determine association between previous injury and PSS
2010). Tests that include hip ROM, flexibility and strength of performance, and prospective injury data was collated to deter-
quadriceps and hamstrings and functional movement ability are mine association between PSS test performance and future injury.
possible inclusions in any hockey PSS, as well as the inclusion of a One hundred and thirty participants were recruited from the inter-
thorough injury history. state level field hockey squads and state level team squads. All
205
M.J. Stokes, J. Witchalls, G. Waddington et al. Physical Therapy in Sport 46 (2020) 204e213

available records were included, but participants were excluded The knee to wall test (KTW) of dorsiflexion range of motion
from the study if they were excluded from training or competition (ROM) was measured using a dorsiflexion lunge test which has
due to a current injury. Participants’ level of attainment was been shown to have good inter- and intra-rater reliability (Intra-
recorded in one of four categories, national senior, state senior, rater ICCs from 0.97 to 0.98: Inter-rater ICC values were 0.97)
national junior and state junior (u18) levels and player positions (Bennell et al., 1998). The participant stood facing a wall and flexed
were also divided into goalkeeper, defense, midfield and attackers. the knee of their test leg forward to touch the wall with their knee.
The research was approved by the University of Canberra Ethics The distance from the tip of their big toe to the wall was measured,
committee and informed consent gained from all participants. as the maximum amount that they could withdraw their foot from
Musculoskeletal measurements were taken between 2015 and the wall while still touching the wall with their bent knee and
2018. Injury data for the preceding 12 months and the following 12 ensuring that their whole foot was kept in contact with the ground.
months of each screening date was collected. Injury data for non- The “Y” balance test (YBT) included an anterior reach (Ant),
contact injuries were collected via the SmartaBase Data Manage- postero-lateral reach (YBTPLat) and postero-medial reach
ment System (AMS) (Fusion Sport, 76 Neon Street, Sumner Park, (YBTPMed). The test has been shown to be reliable (ICC for intra-
QLD 4074) as well as from athlete reporting and questioning by the rater reliability ranged from 0.85 to 0.91 and for interrater reli-
squad physiotherapist at the time of screening. Retrospective injury ability ranged from 0.99 to 1.00) (Munro & Herrington, 2010; Plisky
data was collated from all sources to establish best possible injury et al., 2009). Each participant was allowed 3 practice trials followed
reporting. Follow-up to obtain prospective injury data was for a by 3 trials of measurement and the maximum reach used. The
year post-screening and combined AMS reporting that was maximal reach distance was measured as the point where the most
confirmed by the squad physiotherapist, patient follow up and distal part of the foot reached. The test has further relevance for
treatment history. There are many factors involved and issues field hockey as it has movement patterns that simulate hockey
surrounding the definition of injury as described by Finch et al. movements in reaching to stop, hit and pass the ball. The score is in
(Finch, 1997). Due to the information available in the AMS system centimeters and is adjusted for height (Gribble & Hertel, 2003) as
and the focus of this study on previous injury the definition of an not all particiapnts had limb length measures. Each section of the
injury used was “a recorded episode in AMS that required modified test was scored, for each leg, and a combined score (YBTCombined)
training or competition”. Modified training was defined as the for each participant created for the mean of all directions tested.
injury occurrence required the player to be not available for an The protocol employed for Gastrocnemius Calf Raises (GCR) was
aspect of training or game, not available for the entire training or to repeatedly raise up onto toes off a flat floor, with the measure
game or not able to perform aspects of training or games. This being the number of repetitions. The test was stopped when the
definition is a factor of sports time lost, as a measure of the con- therapist identified a functional change in the movement pattern
sequences to the individual (Finch, 1997) and was chosen as it (Ross & Fontenot, 2000).
provides a basis for comparison at PSS. The AMS system includes Hip internal rotation (Hip IR) was measured in a prone position,
drop down menus for injury data that included body location, and a unilateral measurement taken, using the tibia as a lever so the
hockey mechanism of injury, where and when the injury occurred, foot is moved laterally as far as possible without pelvic movement,
number of treatments and type of treatment required and days that with the angle measured from vertical to the final position. This
modification of training and competition was required. Injury en- method has been shown to be reliable (Simoneau, Hoenig, Lepley, &
tries and injury status of players was cross checked by the main Papanek, 1998).
author and the squad physiotherapist for accuracy. Other injuries The sit and reach test (SR) was conducted in a standard manner
including contact injuries, lower back injuries and upper body in- with a sit and reach over a box (Gabbe, Bennell, Wajswelner, &
juries are measured in the system but were not the focus of this Finch, 2004). A measurement was taken of how far each individ-
study. Musculoskeletal screening was performed by the same ual was able to reach forward in a long sitting position with their
physiotherapist, who remained in place for all years of the data knees held in full extension. Sit and reach has been shown to be a
collection and had over 7 years of experience in treating field reliable measure (Gabbe, Bennell, et al., 2004). Further research
hockey athletes at the commencement of the study years. A battery supports it as a valid measure of hamstring flexibility, but not of
of 17 tests were included (8 are left and right, and 1 bilateral) as lumbar spine flexibility (Liemohn, Sharpe, & Wasserman, 1994).
well as 5 variables calculated from these for a total of 23 tests. The Quality of movement during a single leg squat (SLSq) was rated
current study used data from those tests directly related to lower on a 5-point scale, with one point for each of 5 separate compo-
limb function. nents; anterior alignment (valgus/varus); depth of squat; speed of
squat; trunk alignment and a global control rating. The protocol
2.1. Screening tests was explained to each athlete and the required squat technique and
speed demonstrated prior to testing. Three squats were performed
The Active Movement Extent Discrimination Assessment on each leg and the score for each component provided on the best
(AMEDA) apparatus was used to test proprioceptive acuity, using performance. The measure has been shown to be reliable; inter-
the same protocol as in previous research (Waddington, Adams, & rater reliability for physiotherapists (ICC3,1 ¼ 0.71) and students
Jones, 1999). The device measures ankle inversion somatosensory (ICC3,1 ¼ 0.60); and intra-rater reliability for physiotherapists
accuracy, as a measure of the proprioceptive ability of the athlete. (ICC3,1 ¼ 0.81) and for students (ICC3,1 ¼ 0.71) (Weeks, Carty, &
Over a total of 50 trials, participants are presented with 5 different Horan, 2012).
angles of inversion between 10.5 and 14.5 , randomly sequenced,
10 times each. On each trial, they are asked to identify the angle 2.2. Statistical analysis
experienced, numbered 1e5 based on increased inversion. The
participant’s proprioceptive ability score is the mean area under the Statistical analysis was performed using SPSS version 25 (SPSS
curve (AUC) of the receiver operating characteristic (ROC) curve for Inc., Chicago, IL, USA). Testing for the normality assumption by the
discriminating between adjacent angles of inversion. This gives Shapiro-Wilk test showed that some of the variables were not
scores between 0.5 and 1, where a score of 1 is equivalent to perfect normally distributed. Therefore, Mann-Whitney U (MWU) tests
discrimination, and 0.5 is equivalent to simple chance. The score is were performed comparing screening test results for the dichoto-
multiplied by 100 to show a simplified score in this study. mized injured/non-injured groups (McKnight & Najab, 2010).
206
M.J. Stokes, J. Witchalls, G. Waddington et al. Physical Therapy in Sport 46 (2020) 204e213

Significant associations (p < 0.05) were further examined with 4 cm has been shown to be associated with risk of injury (Smith,
receiver operator characteristic (ROC) curves to determine an effect Chimera, & Warren, 2015). Injury incidence by gender and injury
size and level of discrimination between injured and non-injured side are included in Table 2, to provide an understanding of their
groups. Area Under the Curve (AUC) scores were calculated effect as possible confounders to be included in further analysis.
(0.9e1.00 excellent; 0.8e0.9 good; 0.7e0.8 fair; 0.6e0.7 poor) (Rice Level of attainment, player position and age were assessed as
& Harris, 2005). Raw score optimal cut off points (COP) were possible contributing variables to risk of injury.
determined as the value at which the maximum value of Youden’s Factors that showed association with injury status were further
index was recorded (Fluss, Faraggi, & Reiser, 2005). Tests that analysed in a binary logistic regression to evaluate the likelihood
showed association were further analysed by the creation of an that players who have been previously injured show a difference in
asymmetry metric between the sides. AMEDA asymmetry was the combined factors (deficit) at screening (Table 4) and to assess
created as 0 ¼ no asymmetry, 1 ¼ asymmetry >10% difference the impact of factors on the likelihood that players who show a
between sides; R YBT asymmetry variable was created by 0 ¼ no difference in the combined factors at screening will have a future
asymmetry, 1 ¼ asymmetry > 4 cm difference between sides and injury (Table 5). Factors were included that satisfied binomial
Hip IR was created by 0 ¼ no asymmetry, 1 ¼ asymmetry >5 dif- logical assumptions. Factors that had a p value < 0.1 were included
ference between sides. Asymmetry of the YBT test greater than to reduce the likelihood of a type II error.

Table 1
Descriptive Statistics for pre-screening and post screening injury status.

Pre-screening Post screening

Total Non-Injured Injury Non-Injured Injury

N 130 92 38 105 25

Mean Mean Mean Mean Mean Gender Difference (p)

(S.D) (S.D) (S.D) (S.D) (S.D)

Age 20.96 (3.75) 20.72 (3.84) 21.55 (3.49) 20.86 (3.66) 21.40 (4.12)
(Yrs.)
Height 175.58 (8.36) 175.58 (8.36) 177.34 (8.16) 175.78 (8.13) 177.76 (9.01)
(cm)
R AMEDA 69.38 68.42 71.71 69.45 69.12 .28
(AUC) (8.71) (8.16) (9.63) (8.26) (10.58)
L AMEDA 71.48 69.99 75.08 71.24 72.48 .037
(AUC) (9.93) (9.31) (10.56) (9.90) (10.18)
R KTW 13.02 13.13 12.74 12.92 13.40 0.92
(cm) (2.57) (2.60) (2.51) (2.49) (2.89)
L KTW 12.84 13.04 12.36 12.77 13.12 0.40
(cm) (2.55) (2.43) (2.80) (2.51) (2.76)
R YBT Ant 34.75 35.27 33.50 35.45 31.82 0.40
(cm) (5.62) (5.09) (6.65) (3.56) (10.19)
R YBT PMed 55.60 55.82 55.08 56.56 51.59 0.67
(cm) (9.08) (8.44) (10.58) (5.36) (17.26)
R YBT PLat 55.15 54.98 55.59 55.20 54.95 0.00
(cm) (5.61) (5.78) (5.22) (5.13) (7.42)
L YBT Ant 35.58 35.92 34.74 35.60 35.47 0.49
(cm) (4.05) (4.27) (3.38) (3.94) (4.58)
L YBT PMed 57.10 57.090 57.13 57.00 57.55 0.54
(cm) (4.77) (5.04) (4.13) (4.18) (6.82)
L YBT PLat (cm) 56.05 56.04 56.06 56.29 55.03 0.00
(5.42) (5.74) (4.64) (4.90) (7.26)
R GCR 26.87 26.42 27.97 26.73 27.48 0.16
(raises) (8.82) (9.12) (8.03) (8.79) (9.13)
L GCR 27.05 26.46 28.49 26.87 27.87 0.05
(raises) (8.76) (8.92) (8.29) (8.80) (8.70)
R SLSq 3.72 3.73 3.69 3.71 3.74 0.01
(Rating/5) (0.72) (0.74) (0.67) (0.73) (0.68)
L SLSq 3.66 3.6740 3.63 3.64 3.77 0.06
(Rating/5) (0.75) (0.76) (0.71) (0.76) (0.68)
R Hip IR 34.20 34.53 33.39 35.02 30.76 0.01
(degrees) (9.27) (8.90) (10.19) (8.88) (10.24)
L Hip IR 33.67 34.58 31.47 34.35 30.80 0.00
(degrees) (9.08) (8.37) (10.38) (8.59) (10.61)
R YBT Combined 48.50 48.6876 48.06 49.07 46.12 0.82
(5.40) (5.24) (5.84) (3.83) (9.27)
L YBT Combined 49.58 49.6841 49.31 49.63 49.35 0.93
(3.73) (3.92) (3.25) (3.28) (5.29)
Sit Reach 12.33 12.228 12.59 12.26 12.64 0.11
(cm) (7.58) (8.09) (6.27) (7.70) (7.24)

S.D. ¼ Standard Deviation, AMEDA ¼ Active Movement Extent Discrimination Assessment, AUC ¼ Area under the receiver operating characteristic curve, KTW ¼ Knee-to-wall
measure of dorsiflexion range of motion, YBT ¼Y Balance Test, reach directions; ANT ¼ anterior reach, PMed ¼ postero-medial, PLat ¼ postero-lateral ¼ , YBT
Combined ¼ mean combined reach, GCR ¼ gastrocnemius calf raise measure of calf strength, SR ¼ sit-and-reach, SLSQ ¼ single leg squat, HIP IR Prone ¼ hip internal rotation in
prone.

207
M.J. Stokes, J. Witchalls, G. Waddington et al. Physical Therapy in Sport 46 (2020) 204e213

Table 2
Injury incidence for the whole group, males and females for left and right side, with subsequent injury numbers. Pre/Post: Injury occurred pre ¼ previously, in the 12 months
before screening tests, or post ¼ within the competitive season after screening tests.

WHOLE GROUP MALE FEMALE

No injury Injury R Injury L No injury Injury R Injury L Re-injury No injury Injury R Injury L Re-injury

Pre-Injury 92 21 17 53 13 11 39 11 3
Post-screen Injury 107 10 13 59 6 8 (5)a 44 4 5

Ankle

Pre-Injury 110 11 9 67 5 5 43 9 1
Post-screen Injury 123 2 5 73 1 3 (1)a 50 1 2

Hamstring

Pre-Injury 119 5 6 66 5 6 53 0 0
Post-screen Injury 120 5 5 70 3 4 (4)a 50 2 1
a
Reinjury ¼ recurrence of a previous injury.

3. Results 3.2. Combination of significant associated tests

A total of 130 field hockey players were included in this study The binary logistic regression model for previous injury
(age ± SD ¼ 20.96 (3.75); height ± SD ¼ 176.09 cm (8.31)). Test (Table 4) contained four independent variables from the pre-
descriptive statistics for the whole group and comparison groups liminary analysis (R AMEDA, L AMEDA, L Hip IR and R YBTAnt) and
and are shown in Table 1. Comparisons were performed for pre- included Hip IR difference and gender as a factor as well. The full
screening injured (age (mean ± SD) ¼ 21.55 (3.49); height ¼ 177.34 model containing all predictors was statistically significant chi
(8.16)) and non-injured groups (age ¼ 20.72 (3.84) height ¼ 175.8 squared (9, n ¼ 128) ¼ 17.42, p ¼ 0.05). The model as a whole
(8.36)); as well as injured after screening (age ¼ 21.40 (4.12); explained between 12.2% (Cox and Snell R square) and 17.4%
height ¼ 178.70 cm (8.44)) and not injured after screening (Nagelkerke R square) of the variance in injury status and correctly
(age ¼ 20.86 (3.66); height ¼ 175.53 cm (8.21)). There were 77 classified 76.6% of cases. As shown in Table 4, no independent
males (age ¼ 20.3 (2.91), height ¼ 180.7 cm (6.49) and 53 females variables made a unique statistically significant contribution to the
(age ¼ 20.45 (3.01), height ¼ 169.5 cm (5.85)). (Table 1). To un- model. Sensitivity was 27.0%, specificity was 96.7%, positive pre-
derstand differences results attributed to gender the screening dictive value was 76.9% and negative predictive value was 76.5%.
tests were assessed for differences in gender via a MWU test and The regression model for injury after screening (Table 5) con-
were significant (p < 0.05) for L AMEDA, R YBTPLat, L YBTPLat, R tained two independent variables from the preliminary analysis (L
SLSq, L SLSq, R Hip IR and L Hip IR tests (Table 1). However, there Hip IR and R YBTAnt) and included Hip IR difference, gender and
was not large enough injury incidences for male/female categories previous injury as factors. The full model containing all predictors
to enable separate analysis (Table 2). Univariate testing for associ- was not statistically significant chi squared (8, n ¼ 128) ¼ 11.64,
ation of level of attainment, player position and player age showed p ¼ 0.17). As shown in Table 5, no independent variables made a
no significant associations with injury status or with test results. unique statistically significant contribution to the model.
Table 2 shows the frequency of injury incidence for Male and Reduced performance in the L AMEDA and R YBTPMed were
female, left and right sided injuries and ankle and hamstring in- associated with a prescreening ankle injury, with AUC values 0.676
juries. The numbers are not sufficient for statistical analysis, but and 0.557 respectively. No associations were found with post
they provide information for consideration of sex effect as a screening ankle injuries (Table 4). Reductions in R and L YBTAnt, R
possible confounder. Numbers for ankle and hamstring injury YBTPmed and R YBT Combined were all associated with a pre-
location are included. These are the two most commonly injured screening hamstring injury, with AUC values of 0.781, 0.686, 0.722
areas in this cohort from previous records. The number of post- and 0.736 respectively, in the adequate range of ability to predict a
screening injuries that were re-injury is included; n ¼ 5 of 23 (1 difference in participants who were previously injured and non-
ankle and 4 hamstrings). injured. R YBTAnt, R YBTPmed, both right and left Hip IR and R
YBT Combined showed association with post screening hamstring
injury with AUC values 0.730, 0.730, 0.763, 0.707 and 0.756, also in
the adequate range of predictive ability (Table 6).
3.1. Comparison of individual test results to pre-screening and post
screening injury status 4. Discussion

R AMEDA, L AMEDA and R YBTAnt were significantly associated To reduce non-contact injuries in field hockey it is important to
(p < 0.05) with an injury prior to screening, with AUC values 0.614, identify players with increased risk. The aim of this study was to
0.654 and 0.631 in the moderate range of test ability to predict a identify if intrinsic factors tested in the preseason screening can
difference between injured and non-injured. L Hip IR (p ¼ 0.08) was identify an elevated risk of injury either individually or in combi-
included in binary logistic regression as a possible contributing nation. Injury risk is multifactorial and population specific (Zazulak,
factor. R YBTAnt and right and left Hip IR were significantly asso- Hewett, Reeves, Goldberg, & Cholewicki, 2007). The findings of the
ciated (p < 0.05) with an injury post screening, with AUC values study show individual tests associated with injury status and that
0.643, 0.680 and 0.635 also in the moderate range of predictive combinations of these tests can be useful in an injury risk profile to
ability (Table 3). identify higher risk players. The findings provide clinical usefulness
Assessment of the contribution of asymmetry found no statis- by identifying tests that can be used immediately to inform practice
tically significant differences between those un-injured and and provide a step in the development of evidence-based screening
injured, historically or prospectively (Table 3.). for injury risk profile in this population of field hockey athletes.
208
M.J. Stokes, J. Witchalls, G. Waddington et al. Physical Therapy in Sport 46 (2020) 204e213

Table 3
Pre-Screening and Post Screening Descriptive Statistics; MWUb (p) ¼ Mann Whittney U (p); AUC ¼ Area under the receiver operating characteristic curve; 95% CI ¼ 95%
Confidence Interval ¼ CI; RAW ¼ raw score at Youden’s Index Maximum value.

PRE-SCREENING POST SCREENING

Injury status Injury status

Test No Mean Yes Mean MWUb (p) AUC RAW No Mean Yes Mean MWUb (p) AUC RAW

(Measure) (S.D.) (S.D.) (95% CI) (S.D.) (S.D.) (95% CI)

n ¼ 92 n ¼ 38 n ¼ 107 n ¼ 23

R AMEDA (AUC) 68.42 71.71 0.041 .614 69.5 69.24 70.04 0.915
(8.16) (9.63) (.505-.723) (8.32) (10.51)
L AMEDA (AUC) 69.99 75.08 0.006 .654 78.5 71.09 73.26 0.440
(9.31) (10.56) (.543-.765) (9.88) (10.16)
a
R YBT Ant (ND) 35.27 33.50 0.019 .631 36.03 35.54 31.12 0.031 .643 33.61
(5.09) (6.65) (.523-.739) (3.61) (10.28) (.552-.772)
R YBT PMed 55.82 55.08 0.959 56.57 51.11 0.388
(ND)a (8.44) (10.58) (5.39) (17.82)
R YBT PLat (ND)a 54.98 55.59 0.586 55.16 55.15 0.628
(5.78) (5.22) (5.19) (7.40)
L YBT Ant 35.92 34.74 0.161 35.68 35.10 0.391
(ND)a (4.27) (3.38) (3.95) (4.55)
L YBT PMed 57.09 57.13 0.929 56.97 57.70 0.402
(ND)a (5.04 ( (4.13) (4.15) (7.09)
L YBT PLat 56.04 56.06 0.933 56.27 55.01 0.636
(ND)a (5.74) (4.64) (4.87) (7.55)
R Hip IR 34.53 33.39 0.378 35.09 30.04 0.007 .680 32.5
(degrees) (8.90) (10.19) (8.81) (10.38) (.552-.808)
L Hip IR 34.58 31.47 0.089 34.47 29.96 0.019 .635 28.5
(degrees) (8.37) (10.38) (8.55) (10.65) (.516-.795)
R GCR (repetitions) 26.42 27.97 0.271 26.60 28.24 0.521
(9.12) (8.03) (8.78) (9.10)
L GCR (repetitions) 26.46 28.49 0.133 26.79 28.33 0.527
(8.92) (8.29) (8.76) (8.84)
R SLSq (quality score) 3.73 3.69 0.924 3.70 3.78 0.472
(0.74) (0.67) (0.73) (0.69)
L SLSq (quality score) 3.67 3.63 0.642 3.64 3.75 0.450
(0.76) (0.71) (0.76) (0.69)
R KTW (cm) 13.13 12.74 0.542 12.94 13.35 0.671
(2.60) (2.51) (2.47) (3.01)
L KTW (cm) 13.04 12.36 0.423 12.81 12.96 0.878
(2.43) (2.80) (2.51) (2.81)
Sit Reach 12.23 12.59 0.825 12.29 12.52 0.636
(cm) (8.09) (6.27) (7.63) (7.53)
R YBT Combined (Average) 48.69 48.06 0.778 49.09 45.79 0.316
(5.24) (5.84) (3.89) (9.44)

L YBT Combined (Average) 49.68 49.31 0.707 49.64 49.27 0.758

(3.92) (3.25) (3.26) (5.49)
AMEDA Difference 0.645 0.261
YBTAnt Difference 0.195 0.184
Hip IR Difference 0.052 0.069
a
Normalized Distance¼ (ND).
b
Mann Whitney u Test.

Table 5
Table 4 Logistic Regression predicting likelihood of future injury. Comparison of test results
Logistic Regression predicting likelihood of a deficit in intrinsic factors following with ankle and hamstring injury.
previous injury.
B S.E. SIG. EXP(B) 95% C$I.FOR EXP(B)
Variables B S.E. SIG. EXP(B) 95% C·I.FOR LOWER UPPER
EXP(B)
R YBTAnt 2.741 5.320 0.606 0.065 0.000 2178.616
LOWER UPPER L Hip IR 0.247 0.324 0.445 0.781 0.414 1.473
R YBTAnt 4.090 3.572 0.252 0.017 0.000 18.366 Gender 0.259 0.601 0.667 1.295 0.399 4.205
L Hip IR 0.251 0.181 0.166 0.778 0.546 1.110 Hip IR difference 1.177 0.551 0.033 0.308 0.105 0.908
Gender 0.075 0.503 0.882 0.928 0.346 2.485
Hip IR difference 0.633 0.510 0.214 0.531 0.195 1.442
R AMEDA 0.077 1.673 0.963 1.080 0.041 28.683
of another study that found AMEDA scores were associated with
previous injury (Steinberg et al., 2019) but contrasted the results of
a study that found scores were not associated with self-reported
The study found specifically that performance on the AMEDA is lower limb injury in Snowsport participants (Dickson,
associated with pre-screening lower limb injury and ankle injury, Waddington, & Terwiel, 2019). The latter study included only
but not specifically with hamstring injury. This reflects the findings self-reported injury and the Snowsport population may have

209
M.J. Stokes, J. Witchalls, G. Waddington et al. Physical Therapy in Sport 46 (2020) 204e213

Table 6
Significant Associations with ankle and hamstring injuries.

PRE-SCREENING POST SCREENING

INJURY STATUS INJURY STATUS

b
TEST NO MEAN YES MEAN MWU (P) AUC RAW NO MEAN YES MEAN MWUb (P) AUC RAW

(MEASURE) (S.D.) (S.D.) (95% CI) (S.D.) (S.D.) (95% CI)

N 107 23 123 7

Ankle Injury

L AMEDA 70.29 77.00 0.008 .676 81.5

(AUC) (9.42) (10.54) (.543-.808)

R YBT PMED 54.93 58.73 0.020 .557 35.2

(ND)a (9.69) (4.33) (.536-.679)

Hamstring Injury

R YBTAnt 35.24 29.48 0.002 .781 35.19 29.50 0.016 .730

(ND)a (4.80) 1 (0.11) (.636-.925) (4.80) 1 (0.75) (.560-.899)
R YBTPMed 56.27 48.41 0.042 .686 56.37 46.41 0.016 .730
(ND)a (7.72) 1 (7.25) (.505-.880) (7.62) 1 (7.66) (.559-.901)
L YBTAnt 35.85 32.61 0.010 722
(ND)a (4.03) (3.09) (.526-.9193)
R Hip IR 34.82 26.80 0.006 .763
(Degrees) (9.24) (5.96) (.658-.867)
L Hip IR 34.15 27.90 0.030 .707
(Degrees) (9.08) (7.16) (.547-.866)
RYBT COMBINED (Average) 48.94 43.77 0.015 .736 49.02 42.36 0.007 .756
(4.84) (8.59) (.578-.895) (4.73) (8.78) (.597-.916)
a
ND ¼ normalized distance cm.
b
Mann Whitney u test.

different characteristics than field hockey players in respect to association with future lower limb injury in this cohort. For post
AMEDA based performance. The YBTAnt reach tests are associated screening hamstring injuries, the associations are the same but also
with prescreening lower limb injury, and with hamstring injury. include the R YBTPMed. The associations are weaker with the
The YBTPmed reach is associated with prescreening ankle and prospective component of the study as can be expected with the
hamstring injury. These tests all show association with a previous size of the cohort and the number of injuries. The prospective effect
injury and can be a measure of an existing deficit. They can be used of the tests is confounded by other contributing factors that are
in these squads as one aspect of determining return to play. Further, difficult to control including team changes, training load, coaching,
the YBT reach tests have ecological validity with field hockey and number of games. Any interpretation of the association of the
movement patterns, and the association observed here suggests tests with future injury should be taken carefully. However, there is
that they could also be beneficial for the wider field hockey com- value in identifying potential risk factors to assist daily clinical
munity. The ultimate goal of tests for injury risk is to measure decisions and inform injury prevention strategies that will be un-
effectiveness of injury prevention programs. Intrinsic factors that dertaken. Including the YBT as a practical field test for ongoing
are modifiable are important. Further support for the YBTAnt test monitoring is a simple clinical application.
usefulness is that it has been shown that it can be improved by
performance of the FIFA 11þ program in soccer players (Chimera 4.1. Previous injury association with subsequent injury
et al., 2015). The YBT has a place as an inexpensive test that is
practical in field hockey. It is reliable, valid to hockey and can be The ankle re-injury rate was low, with only 1 of the post
measured longitudinally to monitor changes from baseline per- screening ankle injuries being a re-injury. The clinical interpreta-
formance levels. It may identify persistent deficits after recovery tion of these findings suggests that deficits after ankle injury are
from injury even though the ability to play has returned. The pos- minimal or are rehabilitated well in this squad. Hamstring injuries
sibility is that the deficit is in tissue structure, motor control, ROM, prior to screening however showed more subsequent injuries post
or a combination of these, and single-factor tests are not sensitive screening (4 out of 7 post screening hamstring injuries). Other
enough to demonstrate significant deficits in small squads with studies have found hamstrings a commonly reinjured area with 14%
modest participant numbers. in soccer (Dadebo, White, & George, 2004) and as much as 34% in
The individual tests that show association to injury have another AFL (Orchard & Seward, 2002). The number of injuries here is small,
possible common characteristic, that they are tests of combined however. It does suggest that ankle injuries are well controlled and
contributing intrinsic factors. For example, the AMEDA has multiple have less effect on the squad, but hamstring injuries can incur re-
components that contribute to its final measurement score, injury. This study suggests that for hamstring injury, the return to
including sensory-motor input, central processing and a muscle play decisions should be more conservative.
output that results in response (Han, Anson, Waddington, & Adams,
2014). The YBT reach tests include strength and range of movement 4.2. The usefulness of the tests individually
aspects as well as neuromuscular control.
In respect to associations with future injury the study shows The Youden’s index maximum value provides a cut point that
that the YBTAnt and Hip IR are the only tests that showed an identifies a difference between previously injured and non-injured
210
M.J. Stokes, J. Witchalls, G. Waddington et al. Physical Therapy in Sport 46 (2020) 204e213

groups at time of screening for the respective tests that show as- test has not shown any association with these injury groups. With
sociation. The clinical interpretation of this cut point is a potential the strong association found in this study with the YBT protocol,
measure below which a performance deficit exists in the intrinsic there may be a duplication and the YBT retained as the better
factor that can contribute to an increased risk of injury. A further performing test.
interpretation may be that players with test scores below the cut Asymmetry tests were included in the univariate analysis for
point can have an increased risk of injury, though this requires AMEDA, YBT reach directions and Hip IR and no significant asso-
further study in larger cohorts or longer squad series. In this study ciations were found in this population. This contrasts with previous
the AMEDA test showed association with previous lower limb research that found the asymmetry of YBT as a risk factor for injury
injury and with previous ankle injury. These findings are similar to in college athletics participants (Smith et al., 2015). The lack of
those from other studies (Han et al., 2015; Witchalls, Newman, association in the field hockey cohort may be due to the one-sided
Waddington, Adams, & Blanch, 2013). The AUC values give a cut- nature of the sport, with the result that a small asymmetry is not a
off point for the R AMEDA of 69.5 and for the L AMEDA 78.5 and 81.5 risk factor and could be beneficial. Further analysis on asymmetry is
respectively below which a possible deficit may exist. The differ- required to establish appropriate cut off points for this population
ence in these cut points may reflect the nature of field hockey but is beyond the scope of this study.
where the left, stance leg is more often the pillar for hitting and In summary, in this cohort the tests that show association with
passing. The results can provide immediate benefit in deficit risk of injury are the AMEDA, the YBT reach tests and the Hip IR
guidelines for the squads and indicate that the AMEDA is useful in tests. These should be considered as ongoing targets of research
showing a difference in injured and non-injured groups in this and may be clinically useful in the process of making return-to-
cohort and can be used to help guide return to play decisions. The sport decisions after injury. The results of other intrinsic factors
AMEDA can provide a baseline in identifying possible at-risk ath- in this study and in this cohort show limited usefulness, similar to
letes and the cut-off points used as a clinical guide for intervention, other recent findings for AFL that intrinsic factors alone do not
return to play decisions and targeted treatment. predict risk of injury (Colby, 2018).
The YBT reach tests were found to have differences between The combination of tests in a potential model of injury risk was
injury classification groups in this study, particularly for the R useful in showing a difference between previously injured and not
YBTAnt and R YBTPMed. The cut point values for clinical use in this previously injured players at the time of screening. A number of
cohort indicate that previous injury results in a deficit below these tests identified a sustained difference after an injury and
36.03 cm for the R YBTAnt. For prediction of future injury, cut values could be a measure of an existing deficit. As previous injury is a
for the R YBTAnt below 33.61 cm are associated with an increased known risk factor for non-contact injury, a possible deficit measure
risk of injury. One possible explanation of this is that hockey players can provide clinical relevance in return-to-sport decisions by
reach forward with the left foot more often due to their common providing an objective baseline measure to evaluate when a player
hitting position. has decreased their risk of injury. For this cohort the cutoff values
Both the R Hip IR (32.5 ) and L Hip IR (28.5 ) are associated with for the relevant tests included in the regression analysis are shown
future injury. It is possible that the risk associated with the hip ROM in Table 3 as raw scores in their respective measurement.
below the cut point is due to the requirement in field hockey for
players to achieve depth of hip flexion and rotation to perform 4.3. Limitations
hitting and passing skills, and limited hip ROM can negatively
impact this position and cause increased load on soft tissue struc- The small number of injuries in the categories of interest limit
tures. The clinician should consider whether hypermobility or the statistical comparisons that can be made in a small team cohort
hypomobility is the possible risk factor, and the risk is below the cut of this size. The AUC values for all tests associated with pre-
point or above the cut point. There is also a significant univariate screening injury are low to moderate only, in terms of their pre-
difference in test results between males and females in Hip IR dictive ability. While more data is required to better determine
measurement, suggesting that further research relevant to each sex their predictive ability, the results can be used clinically within the
is required. current squads to inform immediate practice. However, the study
The literature reveals that while a calf raise test is commonly design of comparing screening data from existing team protocols
used, there appears to be no consistent protocol and limited val- for association with injury categories provides clinically useful in-
idity and reliability reporting regarding the test (He bert-Losier, formation for field hockey. The tests selected are reliable as re-
Newsham-West, Schneiders, & Sullivan, 2009; He bert-Losier, ported in existing literature, but dependent on the conduct of
Schneiders, Newsham-West, & Sullivan, 2009). The format used screening by staff in a real-world team environment. Although a
in this study was not associated with injury patterns in its current limitation, this strengthens the validity of the study for real-world
protocol. Because functional tests provide more information team scenarios. This constraint is necessary due to the use of
regarding injury in these squads, this raises questions over the retrospective data. Future efforts to control the reliability of testing
utility of the calf raise for this cohort. can improve the reproducibility of the results in other locations and
The KTW test is easily performed with minimal equipment and squads.
has been regularly employed with field hockey squads. However, The tests were chosen based on suitability for the aim of the
KTW results did not show an association to injury risk in this study and those that were available already in the squad’s standard
sporting cohort. Its association with performance outcomes is protocol. Future research needs to consider what is missing for
outside the scope of this study. future analysis. If we remove tests, then what else needs to be
The SR had no association shown in these squads for these included? Other measurable intrinsic factors were not considered,
injury groups, similar to findings in other research (Bennell, Tully, & and these include lumbar spine testing (which can contribute to
Harvey, 1999). Its usefulness for pre-season screening in these hamstring injury (Orchard, Farhart, & Leopold, 2004)), hamstring
squads is therefore questionable, despite its ease of use as a mea- strength testing and training, and competition load. This study
surement of reach. Possibly its bilateral nature limits its predictive aimed to assess what is currently used and further research on
relevance to hockey, and the unilateral tests such as the AMEDA and relevant tests for hockey is required so that the tests chosen are best
YBT reach tests are more valid. practice.
In this study, the single leg squat test as a functional movement The timing of the testing in pre-season test is also a possible
211
M.J. Stokes, J. Witchalls, G. Waddington et al. Physical Therapy in Sport 46 (2020) 204e213

limitation of the study. Real time information is required to inform Sports Medicine, 38(4), 388e394. https://doi.org/10.1136/bjsm.2002.000044
Dallinga, J. M., Benjaminse, A., & Lemmink, K. A. (2012). Which screening tools can
real time intervention and this study assessed athletes prior to
predict injury to the lower extremities in team sports? Sports Medicine, 42(9),
competition. Possibly more testing per year would provide test- 791e815. https://doi.org/10.1007/bf03262295
retest information regarding changes in baseline levels. The re- Dickson, T. J., Waddington, G., & Terwiel, F. A. (2019). Snowsport experience,
sults of the study however do find association with tests that are expertise, lower limb injury and somatosensory ability. Journal of Science and
Medicine in Sport, 22, S17eS21.
field expedient (the YBT) and this information can be used clinically Ekstrand, J., & Gillquist, J. (1982). Soccer injuries and their mechanisms: A pro-
immediately as an ongoing measure of recovery from injury spective study. Medicine & Science in Sports & Exercise, 15(3), 267e270. https://
(against baseline measures) and return to sport decisions. Further doi.org/10.1249/00005768-198315030-00014
Finch, C. F. (1997). An overview of some definitional issues for sports injury sur-
research could target the usefulness of the tests as ongoing player veillance. Sports Medicine, 24(3), 157e163. https://doi.org/10.2165/00007256-
risk of injury monitoring used in the daily training environment. 199724030-00002
The decision to focus on balance-based testing with some ROM Fluss, R., Faraggi, D., & Reiser, B. (2005). Estimation of the Youden Index and its
associated cutoff point. Biometrical Journal: Journal of Mathematical Methods in
testing limits the scope of the screening but future inclusion of Biosciences, 47(4), 458e472. https://doi.org/10.1002/bimj.200410135
other evidence-based tests and review of the effectiveness of the Fousekis, K., Tsepis, E., & Vagenas, G. (2010). Multivariate isokinetic strength
results to inform practice is essential. asymmetries of the knee and ankle in professional soccer players. The Journal of
Sports Medicine and Physical Fitness, 50(4), 465e474.
Freckleton, G., & Pizzari, T. (2012). Risk factors for hamstring muscle strain injury in
5. Conclusions sport: A systematic review and meta-analysis. British Journal of Sports Medicine,
47(6). https://doi.org/10.1136/bjsports-2011-090664. bjsports-2011-090664.
Gabbe, B. J., Bennell, K. L., Wajswelner, H., & Finch, C. F. (2004a). Reliability of
 R AMEDA, L AMEDA and R YBTAnt tests have association with common lower extremity musculoskeletal screening tests. Physical Therapy in
previous injury and can provide information on readiness to Sport, 5(2), 90e97. https://doi.org/10.1016/s1466-853x(04)00022-7
return to play and risk of re-injury. Gabbe, B. J., Finch, C. F., Bennell, K. L., & Wajswelner, H. (2005). Risk factors for
hamstring injuries in community level Australian football. British Journal of
 R YBTAnt, R Hip IR and L HipIR are associated with risk of future Sports Medicine, 39(2), 106e110. https://doi.org/10.1136/bjsm.2003.011197
injury. Gabbe, B. J., Finch, C. F., Wajswelner, H., & Bennell, K. L. (2004b). Predictors of lower
 The YBTAnt and YBTPMed reach directions and Hip IR are extremity injuries at the community level of Australian football. Clinical Journal
of Sport Medicine, 14(2), 56e63. https://doi.org/10.1097/00042752-200403000-
associated with previous hamstring injury and show a differ- 00002
ence between post screening injured and non-injured groups. Gribble, P. A., & Hertel, J. (2003). Considerations for normalizing measures of the
The injury numbers are limited so interpretation should be star excursion balance test. Measurement in Physical Education and Exercise
Science, 7(2), 89e100. https://doi.org/10.1207/s15327841mpee0702_3
taken carefully. However, as these are the most common re-
H€agglund, M., Walde n, M., & Ekstrand, J. (2006). Previous injury as a risk factor for
injury in this cohort, these tests are clinically useful in inform- injury in elite football: A prospective study over two consecutive seasons.
ing return to play decisions for hockey players. British Journal of Sports Medicine, 40(9), 767e772. https://doi.org/10.1136/
bjsm.2006.026609
Han, J., Anson, J., Waddington, G., & Adams, R. (2014). Sport attainment and pro-
prioception. International Journal of Sports Science & Coaching, 9(1), 159e170.
Ethical statement https://doi.org/10.1260/1747-9541.9.1.159
Han, J., Anson, J., Waddington, G., Adams, R., & Liu, Y. (2015). The role of ankle
proprioception for balance control in relation to sports performance and injury.
The study was approved by the University of Canberra Human BioMed Research International, 842804. https://doi.org/10.1155/2015/842804,
Ethics Committee HREC 16e198 and all athletes were invited and 2015.
Hebert-Losier, K., Newsham-West, R. J., Schneiders, A. G., & Sullivan, S. J. (2009a).
provided informed consent before being included in the study. Raising the standards of the calf-raise test: A systematic review. Journal of
Science and Medicine in Sport, 12(6), 594e602. https://doi.org/10.1016/
Declaration of competing interest j.jsams.2008.12.628
Hebert-Losier, K., Schneiders, A. G., Newsham-West, R. J., & Sullivan, S. J. (2009b).
Scientific bases and clinical utilisation of the calf-raise test. Physical Therapy in
I have no conflicts of interest to disclose. Sport, 10(4), 142e149. https://doi.org/10.1016/j.ptsp.2009.07.001
Henderson, G., Barnes, C. A., & Portas, M. D. (2010). Factors associated with
increased propensity for hamstring injury in English Premier League soccer
References players. Journal of Science oMedicine in Sport, 13(4), 397e402. https://doi.org/
10.1016/j.jsams.2009.08.003
Arnason, A., Sigurdsson, S. B., Gudmundsson, A., Holme, I., Engebretsen, L., & Bahr, R. Ibrahim, A., Murrell, G., & Knapman, P. (2007). Adductor strain and hip range of
(2004). Risk factors for injuries in football. The American Journal of Sports movement in male professional soccer players. Journal of Orthopaedic Surgery,
Medicine, 32(1 suppl), 5Se16S. https://doi.org/10.1177/0363546503258912 15(1), 46e49. https://doi.org/10.1177/230949900701500111
Bahr, R. (2016). Why screening tests to predict injury do not workdand probably Liemohn, W., Sharpe, G. L., & Wasserman, J. F. (1994). Criterion related validity of the
never will…: A critical review. British Journal of Sports Medicine, 50(13), sit-and-reach test. The Journal of Strength & Conditioning Research, 8(2), 91e94.
776e780. https://doi.org/10.1136/bjsports-2016-096256 https://doi.org/10.1519/00124278-199405000-00006
Bennell, K., Talbot, R., Wajswelner, H., Techovanich, W., Kelly, D., & Hall, A. (1998). Lindgren, S., & Maguire, K. (1985). Survey of field hockey injuries. Sports Science and
Intra-rater and inter-rater reliability of a weight-bearing lunge measure of ankle Medicine Quarterly, 1(3), 7e12.
dorsiflexion. Australian Journal of Physiotherapy, 44(3), 175e180. https://doi.org/ McCall, A., Davison, M., Carling, C., Buckthorpe, M., Coutts, A. J., & Dupont, G. (2016).
10.1016/s0004-9514(14)60377-9 Can off-field ‘brains’ provide a competitive advantage in professional football?.
Bennell, K., Tully, E., & Harvey, N. (1999). Does the toe-touch test predict hamstring In BMJ publishing group ltd and British association of sport and exercise medicine.
injury in Australian Rules footballers? Australian Journal of Physiotherapy, 45(2), McCall, A., Fanchini, M., & Coutts, A. J. (2017). Prediction: The modern-day sport-
103e109. https://doi.org/10.1016/s0004-9514(14)60342-1 science and sports-medicine “quest for the holy grail”. International Journal of
Brockett, C. L., Morgan, D. L., & Proske, U. W. E. (2004). Predicting hamstring strain Sports Physiology and Performance, 12(5), 704e706. https://doi.org/10.1123/
injury in elite athletes. Medicine & Science in Sports & Exercise, 36(3), 379e387. ijspp.2017-0137
https://doi.org/10.1249/01.mss.0000117165.75832.05 McKnight, P. E., & Najab, J. (2010). Mann-whitney U test. The Corsini Encyclopedia of
Cameron, M., Adams, R., & Maher, C. (2003). Motor control and strength as pre- psychology, 1-1. https://doi.org/10.1002/9780470479216.corpsy0524
dictors of hamstring injury in elite players of Australian football. Physical Milgrom, C., Shlamkovitch, N., Finestone, A., Eldad, A., Laor, A., Danon, Y. L., et al. (1991).
Therapy in Sport, 4(4), 159e166. https://doi.org/10.1016/s1466-853x(03)00053- Risk factors for lateral ankle sprain: A prospective study among military recruits.
1 Foot & Ankle, 12(1), 26e30. https://doi.org/10.1177/107110079101200105
Chimera, N. J., Smith, C. A., & Warren, M. (2015). Injury history, sex, and perfor- Mohammadi, F. (2007). Comparison of 3 preventive methods to reduce the recur-
mance on the functional movement screen and Y balance test. Journal of Athletic rence of ankle inversion sprains in male soccer players. The American Journal of
Training, 50(5), 475e485. Sports Medicine, 35(6), 922e926. https://doi.org/10.1177/0363546507299259
Colby, M. J. W. (2018). A multifactorial approach to injury risk in elite Australian Munro, A. G., & Herrington, L. C. (2010). Between-session reliability of the star
footballers. Doctoral dissertation), The University of Western Australia, Perth, excursion balance test. Physical Therapy in Sport, 11(4), 128e132. https://doi.org/
Retrieved from api.research-repository.uwa.edu.au. 10.1016/j.ptsp.2010.07.002
Dadebo, B., White, J., & George, K. (2004). A survey of flexibility training protocols Orchard, J., Farhart, P., & Leopold, C. (2004). Lumbar spine region pathology and
and hamstring strains in professional football clubs in England. British Journal of hamstring and calf injuries in athletes: Is there a connection? British Journal of

212
M.J. Stokes, J. Witchalls, G. Waddington et al. Physical Therapy in Sport 46 (2020) 204e213

Sports Medicine, 38(4), 502e504. https://doi.org/10.1136/bjsm.2003.011346 Ugalde, V., Brockman, C., Bailowitz, Z., & Pollard, C. D. (2015). Single leg squat test
Orchard, J., & Seward, H. (2002). Epidemiology of injuries in the Australian football and its relationship to dynamic knee valgus and injury risk screening. PM&R,
league, seasons 1997e2000. British Journal of Sports Medicine, 36(1), 39e44. 7(3), 229e235. https://doi.org/10.1016/j.pmrj.2014.08.361
https://doi.org/10.1136/bjsm.36.1.39 Verrall, G., Slavotinek, J., Barnes, P., Fon, G., & Spriggins, A. (2001). Clinical risk
Plisky, P. J., Gorman, P. P., Butler, R. J., Kiesel, K. B., Underwood, F. B., & Elkins, B. factors for hamstring muscle strain injury: A prospective study with correlation
(2009). The reliability of an instrumented device for measuring components of of injury by magnetic resonance imaging. British Journal of Sports Medicine,
the star excursion balance test. North American Journal of Sports Physical Ther- 35(6), 435e439. https://doi.org/10.1136/bjsm.35.6.435
apy, 4(2), 92. Waddington, G., & Adams, R. (1999). Discrimination of active plantarflexion and
Plisky, P. J., Rauh, M. J., Kaminski, T. W., & Underwood, F. B. (2006). Star Excursion inversion movements after ankle injury. Australian Journal of Physiotherapy,
Balance Test as a predictor of lower extremity injury in high school basketball 45(1), 7e13. https://doi.org/10.1016/s0004-9514(14)60335-4
players. Journal of Orthopaedic & Sports Physical Therapy, 36(12), 911e919. Waddington, G., Adams, R., & Jones, A. (1999). Wobble board (ankle disc) training
doi.org/10.2519/jospt.2006.2244. effects on the discrimination of inversion movements. Australian Journal of
Rice, M. E., & Harris, G. T. (2005). Comparing effect sizes in follow-up studies: ROC Physiotherapy, 45(2), 95e101. https://doi.org/10.1016/s0004-9514(14)60341-x
area, Cohen’s d, and r. Law and Human Behavior, 29(5), 615e620. https://doi.org/ Weeks, B. K., Carty, C. P., & Horan, S. A. (2012). Kinematic predictors of single-leg
10.1007/s10979-005-6832-7 squat performance: A comparison of experienced physiotherapists and stu-
Rishiraj, N., Taunton, J., & Niven, B. (2009). Injury profile of elite under-21 age fe- dent physiotherapists. BMC Musculoskeletal Disorders, 13(1), 207. https://
male field hockey players. The Journal of Sports Medicine and Physical Fitness, doi.org/10.1186/1471-2474-13-207
49(1), 71. Willems, T. M., Witvrouw, E., Delbaere, K., Mahieu, N., De Bourdeaudhuij, I., & De
Rose, C. P. (1981). Injuries in women’s field hockey: A four-year study. The Physician and Clercq, D. (2005). Intrinsic risk factors for inversion ankle sprains in male
Sportsmedicine, 9(3), 97e100. https://doi.org/10.1080/00913847.1981.11711034 subjects a prospective study. The American Journal of Sports Medicine, 33(3),
Ross, M. D., & Fontenot, E. G. (2000). Testeretest reliability of the standing heel-rise 415e423. https://doi.org/10.1177/0363546504268137
test. Journal of Sport Rehabilitation, 9(2), 117e123. https://doi.org/10.1123/ Witchalls, J., Blanch, P., Waddington, G., & Adams, R. (2011). Intrinsic functional
jsr.9.2.117 deficits associated with increased risk of ankle injuries: A systematic review
Simoneau, G. G., Hoenig, K. J., Lepley, J. E., & Papanek, P. E. (1998). Influence of hip with meta-analysis. British Journal of Sports Medicine, 46(7). https://doi.org/
position and gender on active hip internal and external rotation. Journal of 10.1136/bjsports-2011-090137. bjsports-2011-090137.
Orthopaedic & Sports Physical Therapy, 28(3), 158e164. https://doi.org/10.2519/ Witchalls, J. B., Newman, P., Waddington, G., Adams, R., & Blanch, P. (2013). Func-
jospt.1998.28.3.158 tional performance deficits associated with ligamentous instability at the ankle.
Smith, C. A., Chimera, N. J., & Warren, M. (2015). Association of y balance test reach Journal of Science and Medicine in Sport, 16(2), 89e93. https://doi.org/10.1016/
asymmetry and injury in division I athletes. Medicine Science in Sports Exercise, j.jsams.2012.11.037
47(1), 136e141. Witchalls, J. B., Waddington, G., Adams, R., & Blanch, P. (2014). Chronic ankle
Steinberg, N., Adams, R., Ayalon, M., Dotan, N., Bretter, S., & Waddington, G. (2019). instability affects learning rate during repeated proprioception testing. Physical
Recent ankle injury, sport participation level, and tests of proprioception. Therapy in Sport, 15(2), 106e111. https://doi.org/10.1016/j.ptsp.2013.04.002
Journal of Sport Rehabilitation, 1e7. https://doi.org/10.1123/jsr.2018-0164, 00. Zazulak, B. T., Hewett, T. E., Reeves, N. P., Goldberg, B., & Cholewicki, J. (2007).
Theilen, T.-M., Mueller-Eising, W., Bettink, P. W., & Rolle, U. (2015). Injury data of Deficits in neuromuscular control of the trunk predict knee injury risk: Pro-
major international field hockey tournaments. British Journal of Sports Medicine, spective biomechanical-epidemiologic study. The American Journal of Sports
50(11). https://doi.org/10.1136/bjsports-2015-094847. bjsports-2015-094847. Medicine, 35(7), 1123e1130.

213