Training & Testing Thieme

Borms Dorien, Cools Ann. Upper-Extremity Functional Performance Tests: … Int J Sports Med 2018; 00: 00–00

Upper-Extremity Functional Performance Tests: Reference Values

for Overhead Athletes

Authors
Dorien Borms, Ann Cools

Affiliation Abs tr ac t
Rehabilitation Sciences and Physiotherapy, Ghent University, This study aimed to provide an age, gender and sport-based
Gent, Belgium normative database for three functional shoulder tests: Y Bal-
ance Test – Upper Quarter (YBT-UQ), Closed Kinetic Chain Up-
Key words per Extremity Stability Test (CKCUEST), and Seated Medicine
normative database, physical performance test, Ball Throw (SMBT). A second aim was to discuss gender, age
­upper-extremity functional testing and sports differences. Finally, correlation between tests was
evaluated. Overhead athletes (106 male, 100 female) between
accepted 29.01.2018 18 and 50 years old and from three different sports (volleyball,
tennis, handball) performed all functional tests. A linear mixed
Bibliography or regression model was applied to determine significant dif-
DOI https://doi.org/10.1055/a-0573-1388 ferences in test scores between gender, age and sports. Pear-
Published online: 21.3.2018 son correlation coefficients were analyzed to determine the
Int J Sports Med 2018; 39: 433–441 relationship between tests. Normative values were established
© Georg Thieme Verlag KG Stuttgart · New York and divided by gender, sports and age. Results showed signifi-
ISSN 0172-4622 cant gender and age differences for all tests. For YBT-UQ, also
significant side and sports differences were recorded. CKCUEST
Correspondence is moderately correlated with SMBT and YBT-UQ. Weak correla-
Drs. Dorien Borms tion was found between SMBT and YBT-UQ. In conclusion, this
Rehabilitation Sciences and Physiotherapy study provides normative data for YBT-UQ, CKCUEST and SMBT,
Ghent University which is clinically relevant for functionally screening overhead
Campus Heymans athletes and benchmark their performance to others from the
Gent, 9000 same gender, age and sports. A combination of included tests
Belgium is recommended.
Tel.: + 32/9/332 26 32, Fax: + 32/09/332 38 11
[email protected]

Introduction Nevertheless, overhead athletes have to perform in the entire

The overhead throwing shoulder is at high risk for injuries kinetic chain suggesting that functional screening tools should be
[1, 6, 10, 20, 22, 25]. More specifically, prevalence of shoulder inju- part of the athletes’ screening program [7]. Functional tests for the
ries range from 2.1 to 42 % in volleyball [20], 4 to 17 % in tennis [1] lower limb have been investigated extensively and are used widely
and up to 28 % in handball [6], making it one of the most common by researchers, physicians and coaches for pre-season screening,
injuries in these sports. Shoulder pain can occur due to an sudden in-season evaluation, and in rehabilitation to determine the return
event (e. g. player contact) or because of chronic overuse related to play after lower body injury [17, 18]. In contrast, upper extrem-
to local deficits in the shoulder region or more extensively in the ity functional tests are rather scarce and the application of these
kinetic chain. Several individual risk factors for throwing-related tests as a screening tool for injury prevention or return to play after
shoulder injuries are already described in volleyball, tennis and injury is limited [7].
handball such as GIRD or loss of total ROM, decrease in external ro- However, some field tests have been developed to assess upper
tation strength and scapular dyskinesia [6, 19, 26, 29, 31]. Evalua- body function in a closed kinetic chain such as the Y balance test
tion methods are already available to assess these deficits and Upper Quarter (YBT-UQ) and Closed Kinetic Chain Upper Extrem-
screen for athletes at risk [7] such as isokinetic shoulder strength ity Stability Test (CKCUEST) [14, 15]. Although the position of these
testing [11, 13, 36]. tests is not functional for throwing (plank position instead of stand-

Borms D, Cools A. Upper-Extremity Functional Performance Tests: … Int J Sports Med 2018; 39: 433–441 433
 Training & Testing Thieme

ing), the functionality lays in the involvement of the entire kinetic between September 2016 and February 2017. The included
chain in test performance. For open kinetic chain assessment, the ­subjects had to perform volleyball, tennis or handball on a com-
Seated Medicine Ball Throw (SMBT) is often used to assess bilater- petitive level at least 3 h per week, were aged between 18 and 50
al upper body strength or power in overhead athletes [2, 9]. In con- years, and were in good general health. Participants were exclud-
trast to laboratory tests, all of these field tests have the advantage ed if they had (1) a history of shoulder dislocation, or surgery in the
of being portable, low-cost, non time consuming, and easy-to-ad- upper quadrant, lower quadrant or spine, or (2) reported pain with
minister [14, 17]. Moreover, these test have been proven to be re- time-loss in sports participation in the past 6 months. Anthropo-
liable with ICC values ranging from 0.92 to 0.97 for the YBT-UQ [2], metric data of all subjects are summarized in ▶Table 1. Each sports
0.85 to 0.96 for the CKCUEST [35], and 0.98 for the SMBT [2]. How- discipline was initially divided into four age categories, each with
ever, little or no normative reference values for these tests are avail- an 8-year interval: 18–25, 26–33, 34–41 and 42–50 years [8]. Due
able for overhead athletes despite the clinical need for cut-off val- to the small sample size in the oldest age categories, they were
ues as criteria for return to training and return to play [7]. Studies merged into one age category of 34–50 years. Before participa-
that have reported sport-specific data for the CKCUEST and YBT- tion, all subjects read and signed an informed consent form. This
UQ have been performed with male baseball players [21, 27], male study was approved by the Ethical Committee of XXXX (2016/0963),
wrestlers [21] and athletes participating in baseball, lacrosse and and the procedures were performed in accordance with the ­ethical
volleyball [33]. All of these studies were performed with healthy standards of the journal [16].
college-aged athletes, and no research is available providing refer-
ence values for the SMBT in healthy overhead athletes. For func- Testing procedure
tional testing to be optimally applied, it is often beneficial to have The three functional tests were performed in random order and
normative data associated with athletes’ anthropometric and the subjects were given a verbal explanation and visual demonstra-
sports participation characteristics [21]. tion of each test by the investigator. All tests were executed bare-
Therefore, the primary aim of this study was to provide age, gen- foot and standardized verbal encouragement was given in order to
der and sport-based (volleyball, tennis, handball) reference values enhance a maximal performance. After a short and standardized
for three functional shoulder tests: YBT-UQ, CKCUEST and SMBT. A warm-up with multidirectional shoulder movements (10 repeti-
second purpose was to determine whether there were differences tions of forward flexion, abduction, wall push-ups), the testing pro-
in functional test scores based on age, gender and overhead sports. cedure started.
A third purpose was to evaluate correlation coefficients and coef-
ficients of determination in order to evaluate the relationship be- Y Balance Test-Upper Quarter
tween functional tests. The YBT-UQ is a closed kinetic chain functional screening tool using
the Y Balance Test kit (Move2Perform, Evansville, IN, USA). Before
testing, upper limb length was determined in a standing position
Methods with the arm in 90 ° of abduction, full elbow extension and thumb
facing upwards. The distance between the middle of the C7 spinous
Subjects process and the most distal point of the middle finger was then
In total, 206 healthy overhead athletes (106 male, 100 female) measured using a tapeline.
­ articipated in this cross-sectional study. Sports clubs were con-
p The YBT-UQ was performed on the non-dominant (ND) side, fol-
tacted for participation or subjects were recruited by flyers from lowed by the dominant (D) side according to previously published
July 2016 until February 2017. All tests were performed in the field protocols [2, 15]. The dominant side was determined as the throw-
ing side during sports performance. The tested side was named
▶Table 1 Anthropometric data subdivided for male and female partici- based on the weight-bearing hand (left or right). Subjects had to
pants.
stand in a three-point plank position with the tested shoulder per-
Age Males Females pendicular to the hand and with the feet shoulder-width apart. The
category testing hand was placed on the stance platform with the thumb
Age (Years) 18–25 21.45 ± 2.24 22.04 ± 2.09 behind the red line. From this position, the free hand had to push
26–33 28.60 ± 2.30 29.63 ± 2.64 the reach indicator as far as possible in the medial (M), inferolater-
34–50 38.02 ± 3.64 38.84 ± 4.56 al (IL) and superolateral (SL) reach direction, and subsequently re-
Height (m) 18–25 183.18 ± 6.11 168.51 ± 6.03 turn to the starting position in a controlled manner (▶Fig. 1). After
26–33 184.87 ± 6.78 169.00 ± 7.02 two practice trials, three test trials were performed on each side
34–50 181.64 ± 7.35 165.38 ± 4.54 with 30 s of rest in between each trial. The trial had to be repeated
Weight (kg) 18–25 77.28 ± 9.97 66.89 ± 9.57
when the subject failed to maintain the three-point contact,
pushed out the reach indicator (i. e. kicking it away) or when the
26–33 82.83 ± 10.30 70.12 ± 13.48
ground or reach indicator was used for support. For further analy-
34–50 81.38 ± 9.16 67.90 ± 11.86
sis, the mean distance for each individual reach direction was cal-
BMI (kg/m²) 18–25 23.00 ± 2.50 23.55 ± 3.10
culated and normalized for upper limb length together with a nor-
26–33 24.30 ± 3.36 24.62 ± 4.83
malized composite score, which is the mean of the averaged dis-
34–50 24.69 ± 2.59 24.86 ± 4.41
tance in all 3 reach directions.

434 Borms D, Cools A. Upper-Extremity Functional Performance Tests: … Int J Sports Med 2018; 39: 433–441
 ▶Fig. 1 Y Balance Test Upper Quarter.

SMBT

The SMBT is an open kinetic chain functional screening test to as-

sess bilateral upper body power and strength [2, 9]. The subjects
were instructed to sit on the floor with the head, shoulder and back
against the wall (▶Fig. 2). The legs were extended and a 2 kg med-
icine ball was held with both arms in 90 ° of shoulder abduction and
elbows flexed. The medicine ball was covered in gymnastic chalk in
order to leave a clear print on the floor after each throw. A measur- ▶Fig. 2 Seated Medicine Ball Throw.

ing tape was placed on the floor and stretched out over a distance
of 10 m. Participants had to throw the medicine ball forward, in a
straight line and as far as possible with head, shoulders and back
maintaining full wall contact. After three practice trials, four test
trials were executed with a 1-min rest between each trial. To ac-
count for different arm lengths, the medicine ball was dropped
down with the arms extended in front of the body. The distance
between the wall and the most proximal end of the chalk mark was
subtracted from the total throwing distance. For further analysis,
the mean distance was calculated.

CKCUEST
For the CKCUEST, subjects had to adopt a push-up position with
the hands 91.4 cm apart (marked with two stripes of tape on the
floor) and with both shoulders perpendicular to the hands
[12, 14, 27, 30, 34, 35] (▶ Fig. 3). The back and lower body were ▶Fig. 3 Closed Kinetic Chain Upper Extremity Stability Test.
aligned, and the feet were shoulder-width apart. From this posi-
tion, the dominant hand reached across the body, touched the non-
dominant hand, and returned to the starting position. Subsequent-
ly, the same movement was performed by the non-dominant hand. Statistical analysis
The participant was instructed to perform as many alternating All data were entered into SPSS 24 (IBM Corporation, Armonk, NY,
touches as possible in 15 s while maintaining the correct push-up USA) for statistical analysis. Mean and standard deviations were
position. The first investigator assured correct timing by using a calculated for all dependent variables: CKCUEST (number of touch-
digital stopwatch, the second investigator counted the number of es), SMBT (cm) and YBT-UQ (cm) for the medial, inferolateral, su-
touches out loud. The test started when the second investigator perolateral and composite scores on both sides. Trial-to-trial reli-
said “go” and ended when this investigator said “stop”. After one ability of performance scores was calculated for all three function-
submaximal familiarization trial, 3 maximal performance trials were al tests with intraclass correlation coefficients (ICCs; 2-way random,
executed with 45 s of rest in between. For further analysis, the absolute agreement, average measure) and corresponding 95 %
mean number of touches was calculated. confidence intervals (95 %CI).

Borms D, Cools A. Upper-Extremity Functional Performance Tests: … Int J Sports Med 2018; 39: 433–441 435
 Training & Testing Thieme

For the SMBT and CKCUEST, a linear regression model was ap-

343.44 (46.98)
298.19 (55.43)
306,17 (36.15)
296.30 (38.80)
307.08 (21.61)
316.68 (49.81)
311.33 (40.46)
298.09 (22.99)
306.21 (37.58)
plied with random intercept per patient and fixed factors “gender”

SMBT
(male or female), “sports discipline” (volleyball, tennis or handball)
and “age category” (18–25, 26–33, 34–50) in order to determine

▶Table 3 Descriptive analysis (mean and SDs) of the results for YBT-UQ (cm), CKCUEST (number of touches) and SMBT (cm) test scores for male athletes divided by sports discipline and age category.
significant differences in functional test score between these fixed
factors. Additionally, side differences (dominant or non-dominant)

27.00 (2.42)
27.42 (2.51)
27.88 (2.59)
28.33 (3.45)
25.33 (1.58)
26.06 (1.82)

27.63 (2.32)
27.95 (3.49)
27.42 (2.53)
were also checked for all YBT-UQ test scores (three reach directions

CKCUEST
and composite score) using a linear mixed model by adding a fixed
factor “side”. For all statistical analyses, a backwards stepwise re-
gression was applied by starting with the full factorial model and
then gradually leaving out non-significant interaction effects so

89.61 (8.04)
96.99 (7.23)
90.11 (6.89)
88.57 (6.22)
95.03 (8.83)
86.30 (8.92)
90.41 (6.91)
89.80 (5.74)
86.92 (4.77)
that only the highest significant interaction-effect (or mean effect

ND
in absence of an interaction effect) was used in the model for inter-
preting the results. The residuals of the linear mixed models or lin-

Composite
ear regression were checked for normal distribution. Post hoc pair-

84.19 (10.23)
wise comparisons were performed using a Bonferroni correction.

87.70 (7.82)
95.77 (7.68)
88.12 (8.03)
91.83 (5.73)
85.60 (2.21)

87.70 (5.62)
90.00 (7.48)
89.89 (5.27)
Pearson correlation coefficient (r) was used to investigate the

D
possible relationship between the three functional tests. Only the
YBT-UQ composite scores were analyzed for this comparison. We
categorized r values as weak (0.499), moderate (0.50–0.707) or

66.78 (13.47)
79.08 (13.14)
66.53 (10.28)

71.66 (10.48)
73.08 (11.46)
strong (0.707) [32]. Additionally, coefficients of determination (R²)

77.35 (7.95)
68.10 (7.17)
71.52 (9.12)
66.23 (8.49)
were calculated to determine shared variance between functional

ND
tests. Superolateral
An alpha level of.05 was applied to all data for determining sig-
nificant differences.

64.22 (10.93)
71.08 (10.93)
75.64 (10.51)
63.86 (12.26)
65.58 (11.95)

72.72 (9.03)
69.10 (8.69)
72.14 (9.58)

Results 73.34 (7.40)

D

The results of the reliability study showed good-to-excellent trial-

to-trial reliability for all functional tests, and are summarized in
YBT- UQ

▶ Table 2. Mean results of the YBT-UQ (three reach directions and

96.16 (10.08)
92.08 (10.47)

94.05 (11.88)
91.71 (10.34)

94.08 (10.30)
93.47 (10.20)
95.68 (10.84)

102.60 (6.98)
101.29 (7.61)
composite scores), SMBT and CKCUEST are shown in ▶ Table 3
ND

(males) and ▶Table 4 (females) and are divided by sports discipline,

age category and side (only for YBT-UQ).
Inferolateral

The results of the linear mixed model (for YBT-UQ) and linear re-
gression model (for SMBT and CKCUEST) analyses and post hoc
94.70 (t11.88)
84.94 (11.94)

91.82 (12.63)
90.55 (10.46)
93.65 (12.93)
95.19 (10.10)

91.68 (6.83)
103.12 (7.87)
90.84 (7.80)

tests are summarized in ▶ Table 5.

For the YBT-UQ medial reach, results of the linear mixed model
D

showed a significant main effect for sports (p = 0.019) and gender

(p < 0.001). Post-hoc tests revealed that volleyball players have sig-
nificantly lower test scores than handball players (99.109 compared
n, number of participants; D, dominant; ND, non-dominant
104.73 (13.39)

to 102.416 cm) (p = 0.017) and that males reach significantly fur-

105.90 (6.09)
111.02 (7.75)
100.29 (9.35)

104.61 (4.84)
102.83 (4.46)

103.23 (5.77)
104.67 (5.08)
102.21 (2.74)

ther than females (103.585 compared to 97.484 cm) (p < 0.001).

ND

For YBT-UQ IL reach, a three-way interaction was significant for gen-

Medial

▶Table 2 Trial-to-trial reliability: intraclass correlation coefficients (ICC)

with 95 % CI for all three functional tests.
107.21 (8.30)
111.47 (7.72)
102.23 (5.63)
105.17 (8.16)
103.77 (9.72)
100.39 (2.15)

103.45 (5.87)
104.21 (5.30)
101.12 (3.62)

YBT-UQ
D

Medial Dominant 0.973 (0.965 – 0.979)

Non-dominant 0.972 (0.958 – 0.980)
Inferolateral Dominant 0.975 (0.968 – 0.981)
(n)

24
20
19
11

Non-dominant 0.976 (0.971 – 0.981)

9
4
6
7
6

Superolateral Dominant 0.978 (0.972 – 0.983)

Volleyball

Handball

26–33
34–50
18–25
26–33
34–50
34–50

18–25
18–25
26–33

Non-dominant 0.978 (0.972 – 0.983)

Tennis

CKCUEST 0.962 (0.880 – 0.982)

SMBT 0.958 (0.947 – 0.966)

436 Borms D, Cools A. Upper-Extremity Functional Performance Tests: … Int J Sports Med 2018; 39: 433–441
der x sports x side (p = 0.027). In general, post hoc tests showed

233.89 (38.91)

253.22 (37.06)
246.83 (29.41)
247.38 (21.39)
248.89 (30.92)
238.36 (26.30)
222.38 (26.37)

181.25 (5.17)
that males scored significantly higher than females in all sports dis-

SMBT
ciplines and for all sides (p values ranging from 0.001 to 0.018). Ad-

212.50
ditionally, male (p < 0.001) tennis players have significantly higher

▶Table 4 Descriptive analysis (mean and SDs) of the results for YBT-UQ (cm), CKCUEST (number of touches) and SMBT (cm) test scores for female athletes divided by sports discipline and age category.
test scores on ND compared to D side. For YBT-UQ SL reach, a sig-
nificant three-way interaction effect for gender x sports x age was

19.83 (0.24)
19.67 (4.94)

21.10 (3.54)
21.48 (4.19)
21.10 (4.01)
20.33 (2.73)
22.39 (3.76)
18.89 (5.24)
found (p = 0.050). For volleyball, males scored significantly higher

CKCUEST
than females in age category 26–33 (p = 0.003) and 34–50

14.67
(p = 0.032). Male tennis players scored significantly higher than fe-
males in the second age category (p = 0.015), and in the last age
category male handball players scored significantly higher than fe-

79.94 (10.89)

83.66 (18.15)
82.27 (11.02)

82.54 (4.97)
81.50 (9.20)

84.97 (9.28)
84.43 (9.04)
77.49 (9.23)
males (p = 0.048). Regarding the age category, the first category

ND
scored significantly higher than the second (p = 0.038) and oldest

70.87
category (p = 0.046) for female volleyball players. For the YBT-UQ

Composite
composite scores, a significant main effect for gender (p < 0.001)
and side (p = 0.001) was found. Post-hoc tests revealed that test

79.10 (10.36)

84.26 (15.10)
81.78 (11.47)
82.46 (10.59)
76.06 (11.49)

78.26 (5.86)
77.90 (9.94)

85.34 (9.61)
scores on the ND side are significantly higher than on D side (85.843
compared to 84.734 cm) (p = 0.001) and that males reach signifi-

63.28
cantly further than females (88.684 compared to 81.895 cm)
(p < 0.001).
For the SMBT, results of the linear regression model showed a

63.48 (15.45)

68.53 (12.73)
67.40 (20.88)
64.98 (14.09)
59.63 (15.93)
68.57 (11.11)
56.38 (15.46)

55.01 (8.28)
significant two-way interaction effect for gender x age (p = 0.016).

ND
In general, post hoc test showed that males have significantly high-

43.13
Superolateral
er performance scores than females for all separate age categories
(all p-values < 0.001). Additionally, female athletes score signifi-
cantly higher between 18–25 years compared to the second age

62.23 (14.35)

69.68 (12.94)
66.07 (17.08)
63.24 (15.34)
53.61 (16.18)
66.73 (13.59)
52.81 (12.07)

55.07 (2.64)
category between 26–33 years (p = 0.026).
For CKCUEST, results of the linear regression model showed a
D

34.46
significant main effect for gender (p < 0.001) and age (p = 0.016).
YBT- UQ

Post-hoc tests revealed that males score higher than females

(27.048 compared to 20.598) (p < 0.001) and that athletes between

78.81 (15.43)

85.13 (11.27)
84.69 (20.73)
83.63 (12.02)
86.07 (12.54)
82.51 (11.28)

92.63 (0.41)
85.59 (9.82)
18–25 years have higher test scores than athletes between 34–50
ND

years (24.728 compared to 22.758 touches) (p = 0.016).

77.02

Pearson correlation coefficients demonstrate weak correlation

Inferolateral

between the YBT-UQ and the SMBT with r = 0.350 for D and r = 0.362
for ND side. The coefficient of determination was R² = 0.123 for D
79.14 (13.92)

83.65 (11.25)
87.24 (18.55)
83.23 (12.44)
83.12 (10.42)
84.28 (12.42)
82.13 (16.67)

83.64 (2.34)
and R² = 0.131 for the ND side. Moderate correlation was shown be-
tween YBT-UQ and CKCUEST (r = 0.524 for D and r = 0.558 for ND
D

63.47

side) with coefficient of determination R² = 0.275 for D and

R² = 0.311 for ND side. Also moderate correlation was found be-
tween CKUEST and SMBT (r = 0.616) with a coefficient of determi-
n, number of participants; D, dominant; ND, non-dominant
95.89 (14.80)
91.02 (10.94)

96.43 (11.26)

97.19 (3.10)
101.56 (3.02)
96.46 (6.61)

100.55 (6.66)
96.65 (7.89)

nation of R² = 0.379.
ND

92.01

Discussion
Medial

This is the first study providing normative reference values for the
99.48 (13.37)
96.07 (22.54)
93.23 (11.87)

95.94 (7.31)

102.65 (8.57)
98.86 (9.95)
97.27 (3.89)
96.37 (8.81)

YBT-UQ, CKCUEST and SMBT in overhead athletes based on age,

gender and sports (volleyball, tennis, handball). Although the abil-
D

91.90

ity of these tests to predict injury or to guaranty successful return

to training or return to play has not yet been investigated, the data
of this study could be useful for clinicians, trainers or physicians as
part of an athlete’s screening program in order to benchmark the
(n)

26
14
30

2
9

9
6
1

athlete’s performance relative to others from the same gender, age

and sports. Moreover, all three functional tests could be applied in
Volleyball

Handball
34–50

18–25
26–33
34–50
34–50

18–25
26–33
18–25
26–33

the field, because they are portable, inexpensive, and easy and
Tennis

quick to administer or interpret.

Borms D, Cools A. Upper-Extremity Functional Performance Tests: … Int J Sports Med 2018; 39: 433–441 437
 Training & Testing Thieme

▶Table 5 Results from the linear mixed model or linear regression model and post hoc tests for YBT-UQ (cm), CKCUEST (number of touches) and SMBT
(cm) test scores for all variables.

YBT-UQ
MED IL SL COMP CKCUEST SMBT
Four-way interaction NS NS NS NS NA NA
Side x gender x sport x age
Three-way interaction NS NS NS NS
Side x gender x sport V,T,H; ND: ♂ > ♀ * NS
V,T,H; D: ♂ > ♀ *
T; ♂: ND > D
Side x gender x age NS NS
Side x gender x sport NS NS
Gender x sport x age NS V,T; 26-33: ♂ > ♀ *
V, H; 34-50: ♂ > ♀ *
V; ♀: 18-25 > 26-33,
34-50
Two-way interaction NS NA NA NS NS
Side x gender NA
Side x sport NA
Side x age NA
Gender x sport NS
Gender x age 18-25, 26-33,
34-50: ♂ > ♀ *
♀: 18-25 > 26-33
Main effects NA NA NA
Side NS ND > D * NS
Gender ♂>♀* ♂>♀* ♂>♀*
Sport H>V* NS NS
Age NS NS 18-25 > 34-
50 *

D, dominant; ND, non-dominant; NS, not significant; NA, not applicable; V, volleyball; T, tennis; H = handball

* p < 0.05

Results for the YBT-UQ have been previously reported in various except for IL and composite scores. For male tennis volleyball play-
populations [15, 33], or comparison between sports have been in- ers, performance on non-dominant side was significantly higher
vestigated in small populations participating in overhead sports than on dominant side when reaching IL. Also composite scores
such as baseball [5, 21, 33], softball [5], lacrosse [33], swimming were significantly higher on the non-dominant side, disregarding
[4] and volleyball [33]. According to Gorman et al. [15] no signifi- sport or age, with a 1.11 cm side difference. With respect to sport
cant gender differences exist, which is in contrast to our study and disciplines, volleyball players had significantly lower medial reach
Taylor et al. [33] where male athletes reach further than female ath- performance compared to handball which suggests that sport spe-
letes for M, IL and composite scores. This finding is probably due to cific normative values for the included sports in this study are only
the difference in populations included (recreational fitness versus important for YBT-UQ medial reach. In our study, female volleyball
competitive overhead sports or division 1 athletes). Additionally, players (aged 18–26 years) have lower performance scores than
we also found higher male compared to female performance scores female Division I volleyball players (with mean age of 19.6 years)
in the SL reach direction in all included sports and for both sides [33]. These differences may be attributed to lower competition lev-
but not for all age categories. Especially for the younger athletes els and higher mean age in our study. Detailed comparative stud-
(18–25 years), no gender differences could be found. Moreover, ies are lacking for male volleyball players and for tennis or handball
­female volleyball players reach significantly further between the players of both sexes. Moreover, all available studies of athletes are
ages of 18–25 compared to 26–33 or 34–50 years. These age dif- performed with young athletes (mean age less than 20 years)
ferences have not been identified for male athletes or for other [4, 5, 21, 33].
sport disciplines. Since no study has investigated age differences Normative values for SMBT display significantly higher scores
for YBT-UQ performance, these results could not be compared. Re- for men compared to women in all three age categories, and fe-
garding sides, no significant differences between dominant and male athletes aged 18–25 years have better performance than fe-
non-dominant arm have been previously reported in various over- males aged 26–33 years (all independent of sport discipline). These
head sports [4, 5, 21, 33]. Similar results were found in this study results suggests that SMBT data should be age and gender-specif-

438 Borms D, Cools A. Upper-Extremity Functional Performance Tests: … Int J Sports Med 2018; 39: 433–441
▶Table 6 Reference values (mean and SDs) for YBT-UQ composite scores (cm), CKCUEST (number of touches) and SMBT (cm) test scores for male athletes
divided by age category.

(n) YBT-UQ CKCUEST SMBT

D ND
18–25 63 88.56 (7.11) 89.71 (6.63) 27.82 (2.77) 304.57 (38.22)
26–33 26 89.49 (6.31) 90.78 (7.28) 27.49 (2.66) 315.87 (37.98)
34–50 17 87.42 (8.66) 89.04 (8.22) 26.01 (1.96) 308.63 (44.78)

n, number of participants; D, dominant; ND, non-dominant

▶Table 7 Reference values (mean and SDs) for YBT-UQ composite scores (cm), CKCUEST (number of touches) and SMBT (cm) test scores for female
­athletes divided by age category.

YBT-UQ
(n) CKCUEST SMBT
D ND
18–25 70 83.411 (10.36) 84.23 (9.43) 21.66 (3.73) 245.99 (31.86)
26–33 18 81.24 (14.08) 81.24 (14.08) 20.43 (4.32) 227.75 (34.63)
34–50 12 77.65 (10.09) 79.62 (9.86) 19.28 (4.45) 234.35 (34.89)

n, number of participants; D, dominant; ND, non-dominant

ic. For males, athletes in our study have similar throwing distance more variety with lower test scores in the oldest age category com-
compared to basketball players [28], but lower scores than previ- pared to younger athletes. Regarding the latter age category, our
ously reported in volleyball players [32] or golfers [24]. For females, results show lower scores than female Division 1 volleyball players
average SMBT test performance with a 2 kg medicine ball in our [33], which could be explained by differences in competition level.
study is much lower than the average throw in female netball play- Furthermore, only one study prospectively investigated the use of
ers with 400 g netball. For both genders, comparison with other the CKCUEST to identify football players at risk for in-season shoul-
studies is difficult because of different test positions or perfor- der injury [23]. A score of less than 21 touches was established as
mance, different weights of medicine ball used, and different sports cut-off for higher shoulder injury risk. These results should be in-
tested. Further research is needed regarding optimal standardiza- terpreted with caution since no gender difference was made and
tion in literature on SMBT performance. because of the small sample size, the specific sports population and
Only two studies have investigated gender differences in CK- competition level.
CUEST performance, but the results were contradictory [12, 33]. The results of the correlation analysis can be discussed in rela-
Similar to our study, Taylor et al. [33] found that male athletes have tion to the characteristics of test performance and, more specifi-
significantly higher scores than female athletes, whereas in the cally, the kinetic chain involvement (open versus closed) and the
study of Ellenbecker [12], female performance was higher than speed at which the test is conducted (slow versus fast). The CK-
male performance. This difference can probably be attributed to CUEST is executed in a closed chain at a rather high speed and
the modified – easier – test position for female subjects (knees sup- therefore has moderate relationship with the SMBT (open chain but
ported instead of feet). Additionally, our results also showed age fast) and YBT-UQ tests (slow but closed chain). This reasoning could
differences with significantly higher values in the category 18–25 also explain the weak correlation between SMBT and YBT-UQ. Fur-
compared to 34–50 years suggesting that CKCUEST performance thermore, coefficients of determination showed that YBT-UQ com-
decreases with increasing age, although no significant differences posite scores can only account for 12.3 % to 13.1 % of the variance
were found for the middle age category. Moreover, no significant in the SMBT performance and 27.5 % to 31.1 % in CKCUEST perfor-
differences exist between sports suggesting that athletes involved mance. Additionally, the CKCUEST and SMBT have a shared vari-
in volleyball, tennis and handball have similar CKCUEST perfor- ance of 37.9 %. Since none of the included tests are strongly relat-
mance despite the differences in sport-specific biomechanics and ed to each other and because of limited shared variance, we sug-
loading of the kinetic chain. When looking to the descriptive results gest to implement all three tests when screening the athlete.
for male athletes, values are higher than previously reported for Since no statistically significant difference exists between sports
baseball players in one study [33] but slightly lower than those of for YBT-UQ composite scores, SMBT and CKCUEST, ▶ Table 6, 7
Roush et al. [27]. Comparison between these studies and ours is were added and display the reference values for all three function-
difficult due to different sports, ages and methods. Although no al test scores for, respectively, male and female athletes divided by
differences exists between the sports included in our study, there age category (sports were merged). These tables are clinically more
could be variation between other overhead sports, which might in- relevant and are easy to use in the field. Moreover, reference inter-
dicate that reference values for CKCUEST should be subclassified vals (instead of mean values) could be of higher clinical implication
into type of sport. Further research is needed on this topic. Regard- when interpreting performance results. It is recommended to spec-
ing descriptive values for female athletes, performance scores show ify the reference interval by the mean value ± 2 SD [3].

Borms D, Cools A. Upper-Extremity Functional Performance Tests: … Int J Sports Med 2018; 39: 433–441 439
 Training & Testing Thieme

[5] Butler RJ, Myers HS, Black D, Kiesel K, Plisky P, Moorman CT, Queen R.
Limitations Bilateral differences in the upper quarter function of high school aged
baseball and softball players. Int J Sports Phys Ther 2014; 9: 518–524
Interpretation of the results must be viewed within the limitations
[6] Clarsen B, Bahr R, Andersson SH, Munk R, Myklebust G. Reduced
of the study. First, only three overhead sports were included so ex-
glenohumeral rotation, external rotation weakness and scapular
trapolation of these results to other (overhead) sports, should be dyskinesis are risk factors for shoulder injuries among elite male
done with caution. Future studies could focus on providing norma- handball players: A prospective cohort study. Br J Sports Med 2014;
tive data for other shoulder-loading sports such as swimming, gym- 48: 1327–1333
nastics, hockey, badminton, softball or javelin throwing. Moreover, [7] Cools AM, Johansson FR, Borms D, Maenhout A. Prevention of
no healthy non-athletic population was defined with which to com- shoulder injuries in overhead athletes: a science-based approach. Braz
J Phys Ther 2015; 19: 331–339
pare our results. Second, the last age category was small in sample
size, even though the ages of 34–41 years and 42–50 years were [8] Cools AM, Vanderstukken F, Vereecken F, Duprez M, Heyman K,
Goethals N, Johansson F. Eccentric and isometric shoulder rotator cuff
merged into one age category. Although this should be considered
strength testing using a hand-held dynamometer: Reference values for
as a limitation, we believe that the number of participants in each overhead athletes. Knee Surg Sports Traumatol Arthrosc 2016; 24:
age category and for each gender represents the actual trends in 3838–3847
the population. Third, besides the inclusion and exclusion criteria [9] Cronin JB, Owen GJ. Upper-body strength and power assessment in
used in this study with respect to the training and competition vol- women using a chest pass. J Strength Cond Res 2004; 18: 401–404
ume, no subdivision into competition level was made, which could [10] Dines JS, Bedi A, Williams PN, Dodson CC, Ellenbecker TS, Altchek DW,
influence our results. On the other hand, this could lead to includ- Windler G, Dines DM. Tennis injuries: Epidemiology, pathophysiology,
and treatment. J Am Acad Orthop Surg 2015; 23: 181–189
ing only young athletes, since older athletes often perform on rec-
reational or low competitive levels. Finally, the use of these func- [11] Edouard P, Degache F, Oullion R, Plessis JY, Gleizes-Cervera S, Calmels P.
Shoulder strength imbalances as injury risk in handball. Int J Sports Med
tional tests for injury-prevention or for return to play decision-mak-
2013; 34: 654–660
ing has not yet been proven. Further research needs to explore the
[12] Ellenbecker T, Manske R, Davies G. Closed kinetic chain testing
possible injury predicting character of these tests via prospective techniques of the upper extremities. Orthop Phys Ther Clin N Am
studies. Nevertheless, this normative database might be clinically 2000; 9: 219–229
useful for functional evaluation of overhead athletes with specific [13] Forthomme B, Wieczorek V, Frisch A, Crielaard JM, Croisier JL. Shoulder
gender and age-based reference values. pain among high-level volleyball players and preseason features. Med
Sci Sports Exerc 2013; 45: 1852–1860
[14] Goldbeck T, Davies G. Test-retest reliability of the closed kinetic chain
Conclusion upper extremity stability test. J Sport Rehabil 2000; 9: 35–45

This study provides normative data for YBT-UQ, CKCUEST and SMBT [15] Gorman PP, Butler RJ, Plisky PJ, Kiesel KB. Upper quarter Y balance test:
based on age, gender and sports, which is clinically relevant for Reliability and performance comparison between genders in active
adults. J Strength Cond Res 2012; 26: 3043–3048
functionally screening overhead athletes and benchmarking their
[16] Harriss DJ, Macsween A, Atkinson G. Standards for ethics in sport and
performance. In general, all tests showed significant gender and
exercise science research: 2018 update. Int J Sports Med 2017; 38:
age differences. For YBT-UQ, also some significant side and sports 1126–1131
differences were found. Additionally, a combination of all included [17] Hegedus EJ, McDonough S, Bleakley C, Cook CE, Baxter GD. Clinician-
test is recommended when screening the athlete. friendly lower extremity physical performance measures in athletes:
A systematic review of measurement properties and correlation with
injury, part 1. The tests for knee function including the hop tests. Br J
Conflict of Interest Sports Med 2015; 49: 642–648
[18] Hegedus EJ, McDonough SM, Bleakley C, Cook CE. Clinician-friendly
The author has no conflict of interest to declare. lower extremity physical performance tests in athletes: A systematic
review of measurement properties and correlation with injury. Part
2--the tests for the hip, thigh, foot and ankle including the star
excursion balance test. Br J Sports Med 2015; 49: 649–656
References
[19] Hjelm N, Werner S, Renstrom P. Injury risk factors in junior tennis
players: a prospective 2-year study. Scand J Med Sci Sports 2012; 22:
[1] Abrams GD, Renstrom PA, Safran MR. Epidemiology of musculoskel- 40–48
etal injury in the tennis player. Br J Sports Med 2012; 46: 492–498 [20] Kilic O, Maas M, Verhagen E, Zwerver J, Gouttebarge V. Incidence,
[2] Borms D, Maenhout A, Cools AM. Upper quadrant field tests and aetiology and prevention of musculoskeletal injuries in volleyball: A
isokinetic upper limb strength in overhead athletes. J Athl Train 2016; systematic review of the literature. Eur J Sport Sci 2017; 17: 765–793
51: 789–796 [21] Myers H, Poletti M, Butler RJ. Difference in functional performance on
[3] Boyd JC. Defining laboratory reference values and decision limits: the upper-quarter y-balance test between high school baseball players
populations, intervals, and interpretations. Asian J Androl 2010; 12: and wrestlers. J Sport Rehabil 2017; 26: 253–259
83–90 [22] Myklebust G, Hasslan L, Bahr R, Steffen K. High prevalence of shoulder
[4] Butler R, Arms J, Reiman M, Plisky P, Kiesel K, Taylor D, Queen R. Sex pain among elite Norwegian female handball players. Scand J Med Sci
differences in dynamic closed kinetic chain upper quarter function in Sports 2013; 23: 288–294
collegiate swimmers. J Athl Train 2014; 49: 442–446 [23] Pontillo M, Spinelli BA, Sennett BJ. Prediction of in-season shoulder
injury from preseason testing in division I collegiate football players.
Sports Health 2014; 6: 497–503

440 Borms D, Cools A. Upper-Extremity Functional Performance Tests: … Int J Sports Med 2018; 39: 433–441
[24] Read PJ, Lloyd RS, De Ste Croix M, Oliver JL. Relationships between [31] Seminati E, Minetti AE. Overuse in volleyball training/practice: A
field-based measures of strength and power and golf club head speed. review on shoulder and spine-related injuries. Eur J Sport Sci 2013; 13:
J Strength Cond Res 2013; 27: 2708–2713 732–743
[25] Reeser JC, Joy EA, Porucznik CA, Berg RL, Colliver EB, Willick SE. Risk [32] Stockbrugger BA, Haennel RG. Contributing factors to performance of
factors for volleyball-related shoulder pain and dysfunction. PM R a medicine ball explosive power test: A comparison between jump and
2010; 2: 27–36 nonjump athletes. J Strength Cond Res 2003; 17: 768–774
[26] Reeser JC, Verhagen E, Briner WW, Askeland TI, Bahr R. Strategies for [33] Taylor JB, Wright AA, Smoliga JM, DePew JT, Hegedus EJ. Upper-ex-
the prevention of volleyball related injuries. Br J Sports Med 2006; 40: tremity physical-performance tests in college athletes. J Sport Rehabil
594–600 2016; 25: 146–154
[27] Roush JR, Kitamura J, Waits MC. Reference values for the closed kinetic [34] Tucci HT, Felicio LR, McQuade KJ, Bevilaqua-Grossi D, Camarini PM,
chain upper extremity stability test (CKCUEST) for collegiate baseball Oliveira AS. Biomechanical analysis of the closed kinetic chain
players. N Am J Sports Phys Ther 2007; 2: 159–163 upper-extremity stability test. J Sport Rehabil 2017; 26: 42–50
[28] Santos EJ, Janeira MA. The effects of resistance training on explosive [35] Tucci HT, Martins J, Sposito Gde C, Camarini PM, de Oliveira AS. Closed
strength indicators in adolescent basketball players. J Strength Cond kinetic chain upper extremity stability test (CKCUES test): A reliability
Res 2012; 26: 2641–2647 study in persons with and without shoulder impingement syndrome.
[29] Schmidt-Wiethoff R, Rapp W, Mauch F, Schneider T, Appell HJ. BMC Musculoskelet Disord 2014; 15: 1
Shoulder rotation characteristics in professional tennis players. Int J [36] Wang HK, Cochrane T. Mobility impairment, muscle imbalance,
Sports Med 2004; 25: 154–158 muscle weakness, scapular asymmetry and shoulder injury in elite
[30] Sciascia A, Uhl T. Reliability of strength and performance testing volleyball athletes. J Sports Med Phys Fitness 2001; 41: 403–410
measures and their ability to differentiate persons with and without
shoulder symptoms. Int J Sports Phys Ther 2015; 10: 655–666

Borms D, Cools A. Upper-Extremity Functional Performance Tests: … Int J Sports Med 2018; 39: 433–441 441