Journal of Sport Rehabilitation, 2016, 25, 146  -154

http://dx.doi.org/10.1123/jsr.2014-0296
© 2016 Human Kinetics, Inc. ORIGINAL RESEARCH REPORT

Upper-Extremity Physical-Performance Tests

in College Athletes
Jeffrey B. Taylor, Alexis A. Wright, James M. Smoliga, J. Tyler DePew, and Eric J. Hegedus

Context: Physical-performance tests (PPTs) are commonly used in rehabilitation and injury-prevention settings,
yet normative values of upper-extremity PPTs have not been established in high-level athletes. Objective: To
establish normative data values for the Closed Kinetic Chain Upper-Extremity Stability Test (CKCUEST)
and Upper-Quarter Y-Balance Test (UQYBT) in college athletes and compare results between sports and to
analyze the relationship between the 2 tests. Design: Observational. Setting: Laboratory/athletic facility.
Participants: 257 (118 male, 139 female) Division I athletes participating in basketball, soccer, baseball,
lacrosse, volleyball, track and field, and cross-country. Intervention: CKCUEST and UQYBT scores were
recorded as part of a comprehensive injury-risk screening battery. Main Outcome Measure: Pearson correla-
tions assessed the relationship between all measures of the CKCUEST and UQYBT. A factorial ANOVA and
a repeated-measures ANOVA (arm dominance) were used to assess interactions between sex, year in school,
and sport for CKCUEST and UQYBT scores. Results: Normative values for the CKCUEST and UQYBT were
established for 9 men’s and women’s college sports. No significant relationships were found between PPT
scores. Men scored significantly higher than women for the CKCUEST (P = .002) and UQYBT (P = .010).
Baseball players scored significantly higher than athletes from all other sports for the UQYBT (P < .001) but
showed nonsignificant trends of lower scores for the CKCUEST than lower-extremity-dominant athletes such
as runners (P = .063) and lacrosse players (P = .058). Conclusions: Results suggest that average CKCUEST
and UQYBT scores in Division I athletes are distinct from those previously reported in recreationally active
populations and that performance differences exist between sexes and sports. In addition, the CKCUEST and
UQYBT appear to measure different constructs of performance and may complement each other as part of a
screening battery.

Keywords: upper-extremity functional testing, injury risk, college athletics, normative values

Estimating the level of physical function, or the or at field- or courtside.7,8 The data from PPTs are most
ability to perform dynamic tasks during sport activities, often used as outcome measures, although there is also
is critical in making informed decisions when working interest in these tests as prognostic factors.9,10 As outcome
with athletes. Tests that attempt to capture function have measures, PPTs are a proxy measure of an abstract, latent
historically been of 2 types: self-report and physical construct or constructs. Therefore, these tests may mea-
performance. Self-report measures can be affected by sure a single construct or a combination of constructs
memory, candor, pain, and mood and therefore may not of functional performance at the most basic level (eg,
accurately or completely capture the construct of func- strength, power, endurance, agility, mobility, stability) or
tion.1–6 Thus, in an effort to more completely capture at the task-integration level (eg, specific physical move-
function, physical-performance tests (PPTs) are com- ments)11 and can be compared with normative values as
monly used by sports-medicine professionals. As the return-to-play criteria after rehabilitation or to identify
name implies, the athletes are required to physically those who may be at risk for future injury.
perform these tests to demonstrate function rather than The majority of research surrounding PPTs in
simply report their function. PPTs are attractive to a athletes has focused on the lower extremity,8,12 with
broad array of those who work with athletes because little published work regarding upper-extremity PPTs.
they use readily available and affordable equipment and However, upper-extremity injuries are a major concern in
can be easily administered in a sports-medicine clinic college athletics, accounting for 20% to 30% of all ath-
letic injuries.13,14 Risk for upper-extremity injury varies
depending on sport-specific demands, with greater inci-
Taylor, Wright, Smoliga, and Hegedus are with the Dept of dence in sports such as baseball (44–46% of all injuries)
Physical Therapy, and DePew, the Dept of Athletics, High Point and men’s lacrosse (17–26%) while considerably lower
University, High Point, NC. Address author correspondence to in lower-extremity-dominant sports such as men’s (5–7%)
Jeffrey Taylor at [email protected]. and women’s (4–6%) soccer.15–18 Currently, most upper-

146
 Upper-Extremity Physical-Performance Tests   147

extremity PPTS are performed exclusively in the open Methods

chain (eg, shoulder-mobility test,19 single-arm shot put20),
yet many sports, especially those considered contact Subjects
sports, require a combination of open and closed kinetic
chain movements. In addition to being more specific to Subjects were recruited from all NCAA Division I varsity
some sports’ demands, closed-chain PPTs may allow athletic teams across 2 years at the authors’ institution.
the measurement of multiple aspects of physiological In total, 257 college athletes (118 male, 139 female)
performance that are more difficult to assess during open- participated in the study, representing men’s baseball,
chain activities (eg, stability).21 As such, there is a need basketball, lacrosse, cross country, and track and field
to develop PPTs that evaluate physiological performance and women’s basketball, lacrosse, soccer, cross-country,
in closed kinetic chain movements. track and field, and volleyball teams. If an athlete par-
A few tests dedicated to closed-chain function in the ticipated in the screening exam during both years, only
upper extremity have been developed. The Closed Kinetic results from his or her first season were included. Athletes
Chain Upper-extremity Stability Test (CKCUEST) has were excluded from the study if they were less than 18
been found to be a reliable tool to assess upper-extremity years of age without parental consent or had suffered
function.22 This test requires the participant to assume a an acute injury or surgery in the previous 6 weeks that
standard push-up position and perform maximal-speed limited their ability to perform a specific test. Participa-
Downloaded by California State Univ Fresno on 12/03/16, Volume 25, Article Number 2

alternating reaches across midline to targets spaced 36 tion was voluntary, and all subjects who agreed to be
inches apart for 15 seconds.22 While the CKCUEST enrolled in the study provided written informed consent.
appears to be a measure of upper-extremity stability The institutional review board at the authors’ institution
and power, the actual constructs measured by the test approved the study.
remain relatively unknown, although positive correla-
tions with functional tasks such as throwing distance and Testing Procedures
push-up performance have been reported.23 However, the
All subjects were tested with their respective teams as
CKCUEST as originally described may not be appropri-
part of comprehensive preseason athletic screening.
ate for all athletes. The original version of the CKCUEST
Demographic and anthropometric data including sex,
requires athletes to maintain their arms 36 inches apart
age, year in school, sport, height, weight, and limb
throughout the duration of test, yet this does not account
dominance were recorded for each participant. The
for various maturational stages or anthropometric mea-
athletes performed a comprehensive 15-test battery as
surements such as the shoulder width or arm span, poten-
part of a large preseason participation screening that was
tially putting those with narrower builds at a performance
developed to assess symmetry, movement quality, endur-
disadvantage. Thus, modifying the CKCUEST to account
ance, strength, power, flexibility, stability, and balance.
for body size may be warranted.
The tests were performed on both the upper and lower
In addition, the Upper-Quarter Y-Balance Test
extremities, but only the upper-extremity tests were ana-
(UQYBT), with excellent test–retest (ICC = .80–.99)
lyzed in this study. Although formal randomization was
and interrater reliability (ICC = 1.00), requires athletes
not assigned, all PPTs were performed in random order,
to reach maximally in 3 different directions while main-
based on availability of different stations throughout each
taining a unilateral push-up position on the contralateral
session. Before testing athletes, each examiner received
arm.24 The unilateral nature of this test allows the tester
training that consisted of a written description and photos
to differentiate performance scores for each limb. The
of each test, followed by practice in scoring subjects
UQYBT has been suggested to measure a combination
while supervised by one of the principal investigators.
of upper-extremity strength, stability, and mobility,24 yet
Data were collected in pencil-and-paper format or via
positive correlations have only been reported with the Lat-
eral Trunk Endurance Test and push-up performance.25 an iPad application and either hand-entered or exported
Further development of these tests may lead to help into a Microsoft Excel spreadsheet.
with rehabilitation and injury prevention.26 Normative Closed Kinetic Chain Upper-Extremity Stability
data for the CKCUEST and UQYBT have already been Test.  Two pieces of tape 36 inches apart and parallel to
established in recreationally active individuals and low- each other were attached to the floor (Figure 1). In the
level college baseball players,24,25,27 yet additional values originally published description of the test, all subjects
need to be established in athletes of various skill levels began the test with their hands touching the parallel
and sports. Therefore, the purposes of this study were to pieces of tape.22 However, athletic populations can range
establish normative data values for the CKCUEST and considerably in terms of age, sex, sport, and position,
UQYBT in NCAA Division I college athletes and to ana- ultimately resulting in substantially different body sizes.
lyze the relationship between the 2 tests. Subsequently, In our experience, having a narrower shoulder width or
we aimed to compare the results of the CKCUEST and shorter arm span has placed athletes at a disadvantage
UQYBT by sex and sport, hypothesizing that higher when performing the CKCUEST due to the extra
performance for each test would be found in men and stabilization efforts needed while maintaining the arms
athletes participating in upper-extremity-dominant sports 36 inches apart regardless of shoulder width. Thus, the
(eg, baseball). position of upper-extremity support was modified from

JSR Vol. 25, No. 2, 2016

 148  Taylor et al

(a)

Figure 1 — The Closed Kinetic Chain Upper-Extremity

Downloaded by California State Univ Fresno on 12/03/16, Volume 25, Article Number 2

Stability Test.

36 inches and normalized to body size by requiring each

subject to start in a push-up position, with their hands
located directly under their shoulders. Although the
position of the upper extremity was normalized for each
participant, the distance between targets (36 in) remained
constant. Our laboratory has assessed the reliability of the
CKCUEST with the modified upper-extremity position
in a pilot study of 19 subjects, with results indicating (b)
good test–retest (ICC = .73, SEM = 8 touches) and
excellent interrater (ICC = .97, SEM = 6.0) reliability
and precision.28 Once the initial upper-extremity position
was approved by the tester, a signal was given to begin
repetitive, across-the-body reaches, alternating hands
with each touch. Athletes were instructed to touch the
contralateral piece of tape as rapidly as possible for 15
seconds. The score was calculated by tallying the total
number of cross-reaches with both hands, and in the
interest of efficiency, only 1 trial was performed.
Upper-Quarter Y-Balance Test.  Athletes were
instructed to assume a push-up position perpendicular
to the testing device with their feet shoulder width apart
(Figure 2). They then balanced on 1 hand and pushed (c)
a target indicator along a distance-marked PVC pipe Figure 2 — The Upper-Quarter Y-Balance Test in the (a)
in each of 3 directions (lateral, overhead, and crossing overhead, (b) lateral, and (c) underbody directions.
underneath their body) with the opposite hand. Athletes
were prohibited from bearing weight through the reaching
arm until the completion of the test. The process was a ball. Excellent test–retest (ICC .80–.99, SEM 2.2–2.9)
then repeated while balancing on the contralateral arm, and interrater (ICC 1.00) reliability and precision of these
completing 3 trials on each side. The best reach (cm) methods for assessing the UQYBT have been previously
in each direction was recorded and normalized to limb established.24
length using the following formula: greatest reach in
each direction divided by the limb length (distance in Statistical Analysis
centimeters from C7 spinous process to the middle
fingertip with arms held in 90° abduction) times 100. A All data were exported to a Microsoft Excel spreadsheet
normalized composite score was calculated as the mean for organization purposes before being exported to SPSS
of each subject’s best score in all 3 directions.24 The test (IBM Corp, released 2011, IBM SPSS Statistics for Win-
was named for the weight-bearing arm and analyzed in dows, Version 20.0, Armonk, NY) for analysis. Shapiro-
reference to the athletes’ self-reported arm dominance, Wilk tests were performed to assess the normality of each
defined as the arm with which they would prefer to throw dependent variable. Descriptive statistical analysis was

JSR Vol. 25, No. 2, 2016

 Upper-Extremity Physical-Performance Tests   149

performed for the UQYBT and CKCUEST in relation to a relatively normal distribution with mild leptokurtosis
sex, year in school, and sport participation. A 2 × 4 × 6 and minimal skewness, and we determined that these
factorial ANOVA was performed to assess interactions minor issues for this single variable did not preclude use
between sex (male, female), year in school (freshmen, of parametric statistics.
sophomores, juniors, fourth/fifth-year seniors), and
sport (basketball, baseball, lacrosse, soccer, track and Closed Kinetic Chain Upper-Extremity
field/cross country, volleyball) with CKCUEST scores. Stability Test
UQYBT scores for the dominant and nondominant arms
were compared using a repeated-measures ANOVA with Percentile ranks and descriptive statistics of the
arm as the within-subject repeated measure and sex and CKCUEST are reported in reference to sex in Tables 2A
sport serving as the between-subjects variables. Statisti- and 2B and sport in Table 3. ANOVA results revealed
cally significant (α = .05) main effects and interactions statistically significant differences between sex (P =
were identified and further assessed using independent t .002, d = 0.71) and sport (P = .044) and a sex-by-sport
tests for pairwise comparisons. Effect sizes were calcu- interaction (P = .001) in CKCUEST scores. In reference
lated for all statistically significant pairwise comparisons to a sport main effect, no pairwise comparisons were
to more precisely determine the true magnitude of the statistically significant, but non–statistically significant
differences, with 0.2 considered small, 0.5 medium, and trends indicated that baseball players scored lower than
Downloaded by California State Univ Fresno on 12/03/16, Volume 25, Article Number 2

0.8 large effects.29 In addition, Pearson product–moment both lacrosse (P = .058) and track and field and cross-
correlations were used to assess the association between country athletes (P = .063). There was not a statistically
all measures of the UQYBT and CKCUEST scores. significant main effect of year in school on CKCUEST
scores (P = .997). Further analysis of the sex-by-sport
interaction with independent t tests to assess pairwise
Results comparisons suggests that male athletes score higher than
female athletes in lacrosse (P < .001, d = 1.37), and track
Table 1 displays the demographic and descriptive results and field and cross country (P < .001, d = 1.26), but no
of the 257 included subjects, of whom 230 and 250 significant differences were evident between male and
completed the UQYBT and CKCUEST, respectively. female basketball players.
No athletes refused to participate in any of the upper-
extremity PPTs, but some did not participate in 1 or the
Upper-Quarter Y-Balance Test
other of the tests (CKCUEST n = 7, UQYBT n = 27)
due to time conflicts with the screening session. Shapiro- Repeated-measures ANOVA results revealed no statistical
Wilk tests of normality confirmed that all test scores difference between dominant and nondominant arms for
were normally distributed in men and women except for composite (P = .22), overhead (P = .24), lateral (P = .98),
CKCUEST scores in women (P = .002). Transformation or underbody (P = .51) UQYBT scores. UQYBT com-
of the data was attempted but did not change the shape posite (P = .01, d = 1.01), lateral (P = .001, d = 1.13), and
of the distribution. Visual inspection of the data revealed underbody (P = .03, d = 0.85) scores revealed statistically

Table 1  Demographic and Descriptive Subject Data

Year in Schoola Dom Arm
Age (y), Height (cm), Mass (kg),
Sport n Mean (SD) Mean (SD) Mean (SD) 1 2 3 4/5 R L
Men’s 118 19.3 (1.2) 183.2 (7.8) 82.2 (11.9) 51 26 31 11 83 22
 basketball 10 19.7 (1.2) 195.1 (7.1) 91.8 (11.9) 3 3 3 1 8 2
 baseball 37 19.5 (1.0) 184.1 (6.3) 87.9 (9.8) 12 2 21 2 32 5
 lacrosse 36 19.0 (1.2) 181.8 (6.1) 83.1 (7.2) 18 15 1 2 24 9
 TF/CC 35 19.2 (1.4) 180.0 (7.9) 72.0 (11.2) 17 6 6 6 19 6
Women’s 139 19.2 (1.2) 169.6 (8.4) 63.5 (9.6) 48 35 30 27 128 7
 basketball 15 18.9 (1.1) 175.3 (11.5) 71.7 (14.6) 5 5 4 1 15 0
 volleyball 19 19.6 (1.0) 178.8 (7.2) 71.4 (9.3) 3 6 3 7 18 1
 lacrosse 42 19.2 (1.2) 165.7 (7.2) 61.8 (7.3) 13 13 10 6 39 3
 soccer 28 19.5 (1.2) 169.0 (4.8) 63.0 (6.0) 8 6 7 7 27 1
 TF/CC 35 19.0 (1.1) 167.3 (5.6) 57.9 (6.6) 19 5 6 5 28 2
Abbreviations: DOM, dominant; TF/CC, track and field and cross-country.
a 1 = freshman, 2 = sophomore, 3 = junior, 4/5 = senior or fifth-year senior.

JSR Vol. 25, No. 2, 2016

 150  Taylor et al

Table 2A  Mean (SD) and Percentiles (%ile) for the Upper-Quarter Y-Balance Test and Closed
Kinetic Chain Upper-Extremity Stability Test (CKCUEST) in Men
Upper-Quarter Y-Balance Test
Composite Lateral Overhead Underbody
%ile DOM ND DOM ND DOM ND DOM ND CKCUEST
10 90.0 91.0 99.1 98.1 65.8 70.0 93.8 92.6 19.0
20 94.0 95.0 103.0 101.6 72.0 72.1 99.0 99.4 22.0
30 97.0 98.0 105.8 105.3 77.0 78.0 104.0 104.0 23.0
40 99.5 99.9 109.3 108.7 80.8 80.6 107.0 107.0 24.0
50 101.0 102.0 112.0 112.0 84.0 83.7 110.0 111.0 25.0
60 103.0 104.0 115.0 113.0 86.2 88.0 113.0 113.4 26.0
70 105.0 107.0 116.4 115.5 89.0 90.9 116.6 117.0 28.0
80 109.6 109.0 119.6 119.0 93.0 94.1 120.5 123.5 29.0
Downloaded by California State Univ Fresno on 12/03/16, Volume 25, Article Number 2

90 114.0 113.0 124.0 121.2 97.1 99.0 129.0 129.3 30.4

Mean 101.4 101.8 111.4 110.5 82.8 84.2 110.1 110.7 25.0
(SD) (9.1) (8.7) (9.4) (9.0) (11.9) (12.1) (13.1) (13.8) (4.5)
Abbreviations: DOM, dominant arm; ND, nondominant arm.

Table 2B  Mean (SD) and Percentiles (%ile) for the Upper-Quarter Y-Balance Test and Closed
Kinetic Chain Upper-Extremity Stability Test (CKCUEST) in Women
Upper Quarter Y-Balance Test
Composite Lateral Overhead Underbody
%ile DOM ND DOM ND DOM ND DOM ND CKCUEST
10 77.0 80.0 85.0 86.0 61.4 59.0 78.0 81.6 18.6
20 82.0 84.0 90.8 91.1 68.8 67.0 84.0 86.8 20.0
30 86.0 87.0 94.0 94.9 73.0 72.0 89.1 90.6 21.0
40 88.3 89.5 96.6 97.7 74.6 75.0 93.0 94.1 21.0
50 92.0 91.0 99.0 100.0 78.0 78.5 99.0 98.9 23.0
60 95.0 95.0 102.0 103.0 80.0 81.0 103.1 101.8 23.0
70 97.3 97.0 105.0 105.1 82.9 83.0 107.0 106.9 24.2
80 101.0 100.9 109.0 109.6 88.3 88.5 112.0 110.4 26.0
90 105.6 104.4 112.0 114.0 96.2 94.6 118.2 117.8 29.0
Mean 91.7 92.1 99.1 100.4 78.1 77.5 97.9 98.5 22.9
(SD) (10.8) (9.7) (10.5) (10.9) (13.0) (12.9) (16.2) (14.6) (4.2)
Abbreviations: DOM, dominant arm; ND, nondominant arm.

significant main effects for sex, where men scored higher d = 1.00, underbody d = 0.84). Due to the fact that not
than women, although no difference was seen in scores all sports were represented by both sexes, some inherent
for the overhead reach (P = .24). A main effect was also bias exists in the statistical model; therefore, no further
seen for sport for all UQYBT measurements (UQYBT pairwise comparisons of sport were deemed applicable.
composite P < .001, lateral P < .001, overhead P = .01, Finally, no significant interactions of arm dominance, sex,
underbody P < .001). Pairwise comparisons indicated and sport were detected in UQYBT scores.
that baseball players reached significantly higher in the
lateral and underbody directions and achieved a higher Relationship Between CKCUEST
composite score than athletes in all other men’s sports (P
and UQYBT
< .001) including men’s basketball (composite d = 0.99,
lateral d = 0.94, underbody d = 1.28), lacrosse (composite Correlational analysis of the PPTs revealed a minimal
d = 0.82, lateral d = 0.99, underbody d = 0.84), and track relationship between CKCUEST scores and UQYBT
and field and cross-country (composite d = 0.66, lateral composite or specific directional test scores in men

JSR Vol. 25, No. 2, 2016

 Upper-Extremity Physical-Performance Tests   151

Table 3  Mean (SD) Upper Quarter Y-Balance Test and Closed Kinetic Chain Upper-Extremity
Stability Test (CKCUEST) Scores by Sport
Upper Quarter Y-Balance Test
Composite Lateral Overhead Underbody CKCUEST
Sport n DOM ND DOM ND DOM ND DOM ND n Reaches
Men’s
 basketball 6 98.5 95.8 109.8 108.0 83.3 83.0 102.7 96.3 10 22.5
(9.0) (8.9) (6.7) (8.6) (12.5) (12.3) (17.0) (17.5) (3.7)
 baseball 36 106.0 105.1 117.0 115.1 83.9 83.9 117.3 116.8 35 22.6
(8.3) (7.7) (7.5) (7.5) (12.5) (11.4) (9.5) (10.7) (4.7)
 lacrosse 26 98.1 99.6 108.4 108.2 80.1 81.7 106.0 108.7 36 26.5
(7.9) (8.7) (8.8) (7.5) (10.6) (13.2) (12.6) (13.6) (3.6)
 TF/CC 29 99.1 100.9 107.4 107.6 83.6 87.1 106.3 108.1 34 26.6
(9.0) (8.7) (9.3) (10.0) (12.3) (11.7) (13.2) (13.7) (4.1)
Downloaded by California State Univ Fresno on 12/03/16, Volume 25, Article Number 2

Women’s
 basketball 15 93.0 92.5 105.0 105.5 81.9 82.3 91.9 89.9 15 24.7
(11.4) (11.4) (10.9) (13.3) (10.0) (16.2) (23.1) (20.6) (6.1)
 volleyball 19 87.9 90.8 100.1 99.7 70.3 75.9 93.1 96.8 18 24.7
(11.3) (10.6) (9.0) (9.3) (14.3) (16.0) (14.9) (12.1) (3.8)
 lacrosse 42 94.8 94.0 97.4 98.8 80.6 77.7 106.4 105.3 42 22.0
(10.1) (9.3) (11.6) (11.0) (13.0) (11.6) (13.2) (13.8) (3.1)
 soccer 28 85.3 86.9 95.0 95.3 73.0 73.5 87.9 92.0 25 23.6
(10.5) (7.7) (10.2) (9.4) (13.2) (9.2) (12.3) (12.3) (5.3)
 TF/CC 29 95.3 95.2 101.9 105.5 82.6 79.8 101.6 100.9 35 21.9
(7.8) (8.7) (8.1) (9.1) (9.6) (13.2) (12.9) (10.5) (3.3)
Abbreviations: DOM, dominant arm; ND, nondominant arm; TF/CC, track and field and cross-country.

or women (r = .04–.18), suggesting that these 2 PPTs In men, prior results have ranged from 18.5 to 30.4
quantify the measurement of different constructs of reaches,22,25,27,30 with typically lower CKCUEST scores in
performance. female populations, ranging from 15.9 to 20.5 reaches,25,30
although the only study to compare men and women
reported higher scores in women than in men.30 Our
Discussion results are comparable to prior findings in men; however,
This is the first study to establish normative values women scored higher in our study than others previously
for the CKCUEST and UQYBT in male and female reported, yet men (25.0 ± 4.5) still scored significantly
Division I athletes. Relative to previously published higher than women (22.9 ± 4.2, P = .002). The larger
findings,22–25,27,30 our results are consistent, yet distinct, CKCUEST scores exhibited in women in our study may
confirming the importance of establishing normative be attributed to the modification we made to the testing
values of PPTs for different sets of populations based on procedures. In prior screenings using the CKCUEST, we
gender, sport, and skill level. While the ability of these noted extra stabilization efforts for those with a narrower
scores to predict injury or successful return to play has not build when all athletes, regardless of body size, were
been reported, clinicians may be able to use these norma- required to maintain a 36-inch-wide base of support.
tive values to gauge their athletes’ performance relative Thus, we attempted to adjust for body-size heterogene-
to others and may help identify specific weaknesses to ity by allowing each athlete to keep his or her hands at a
target during training or as part of a return-to-play deci- shoulder-width distance throughout the test yet maintained
sion. Although 1 recent study reported the CKCUEST the original distance between targets and did not modify
to be predictive of shoulder injury in football players, a the scoring procedures. While this modification to the
small sample size and large confidence intervals around original version of the CKCUEST test may have allowed
key metrics limit the generalizability of these findings.10 female athletes to perform better than in other studies, the
Further research is warranted to establish the predictive leptokurtic distribution of women’s CKCUEST scores
ability of the CKCUEST and UQYBT for use as injury- exhibited with the modified test may limit the ability to
risk screening tools or return-to-play procedures. differentiate high- and low-performing female athletes.
Normative values for the CKCUEST have been The number of trials of the CKCUEST was also
previously established in various populations.22,23,25,27,30 modified in our study. Traditionally, a warm-up trial

JSR Vol. 25, No. 2, 2016

 152  Taylor et al

is followed by 3 scored trials, which are averaged for power, speed, and stability, whereas the UQYBT further
the athlete’s final score. In comparing our results with challenges the athlete’s limits of stability by focusing on
those of Roush et al,27 who performed the CKCUEST other components including balance, proprioception, and
on lower-level college baseball players, our baseball mobility of the thoracic spine and scapula. In addition,
players scored substantially lower (22.6 ± 4.7 reaches) the UQYBT’s ability to produce unilateral measures may
than baseball players of a lower competitive level (30.4 reveal asymmetries that could account for differences
± 3.9 reaches). Although we would expect CKCUEST between the 2 tests’ scores.
scores to have a positive relationship with competitive While our hypotheses were confirmed in regard
level, the inverse relationship indicated by the 2 studies to performance differences between men and women,
may in part be explained by the difference in trials. It is analyzing results across sports revealed conflicting
possible that the CKCUEST may have a learning effect results. When compared with other athletes, baseball
and that performing repetitive trials may promote better, players achieved significantly higher UQYBT scores
more consistent results. It is also conceivable that multiple than athletes who participate in sports that may not
trials of a test that emphasizes power may promote fatigue require as much unilateral upper-extremity strength and
and lower scores on subsequent trials. Future studies are stability (eg, lacrosse, cross-country, track and field) but
needed to analyze the extent of the differences between trended toward lower CKCUEST scores. These findings,
both the starting positions and the number of trials before in part, might be because the CKCUEST and UQYBT
Downloaded by California State Univ Fresno on 12/03/16, Volume 25, Article Number 2

making recommendations of 1 methodology over the require and measure different constructs of performance.
other for the CKCUEST. Associations between UQYBT results and measures of
To our knowledge, only 1 other study has reported core stability have been reported,25 suggesting that the
normative values for the UQYBT.24 Similar results were UQYBT may quantify both upper-extremity and core
found with respect to arm dominance, as no significant function. Furthermore, it is possible that the demands of
differences between the dominant and nondominant arms the CKCUEST, which is performed in a push-up position,
in composite or individual direction scores in either sex does not replicate the upper-extremity demands of any of
were reported.24 However, in contrast to results from the sports tested and, therefore, its relevance for assess-
the study done by Gorman et al,24 our statistical model ing functional status in athletes may require continued
indicated a significant difference in UQYBT composite investigation.
scores for both the dominant and nondominant arms While normative values are useful for study-to-study
between sexes, with the average normalized composite comparisons, we note that the results of this study and
score in men (dominant 101.4% ± 9.1%, nondominant others mentioned in this manuscript may be immediately
101.8% ± 8.7%) larger than that in women (dominant used in rehabilitation settings. Clinicians frequently use
91.7% ± 10.7%, nondominant 92.1% ± 9.7%). Athletes normative values of PPTs, such as the triple hop for dis-
in our study also revealed sex differences with reaching tance and the Star Excursion Balance Test, during lower-
in the lateral and underbody directions, but no difference extremity rehabilitation and return-to-sport decisions,31,32
was exhibited in overhead scores. Both the men’s and the yet the use of upper-extremity PPT normative values
women’s composite scores in our study were greater than is much less common. Due to the dearth of evidence
the average composite reach previously reported (84.5% regarding upper-extremity PPTs, current return-to-sport
± 8.3%).24 Performance differences may be attributed to algorithms do not i-nclude PPTs in the decision-making
the younger age and greater conditioning of the Divi- process. While our results may provide clinicians with
sion I athletic population than in a recreationally active data that are generalizable and comparable to other Divi-
population. sion I athletes, the normative values of the CKCUEST
In examining the relationship between CKCUEST and UQYBT presented in this article should be used for
and UQYBT scores, there were no statistically significant comparison only, considering that the ability of these
correlations between either the composite or directional tests to predict injury and safe return to sport has not yet
UQYBT scores and CKCUEST scores for either sex. Our been determined.
results conflict with previous findings that showed a mod-
erate positive relationship between the 2 PPTs for both the Limitations
dominant (r = .43, P = .02) and the nondominant arm (r
= .49, P = .01).25 However, differences in methodology, The PPTs analyzed in this study were just a fraction of
especially considering the populations studied (Division a battery of 14 other tests administered in the subject
I athletes compared with healthy college students) may population. Tests were performed in no specific order
explain this disparity, as training experience or higher and based on the athlete’s self-selection, so pure random-
skill levels may lead to distinctions between test results ization was not achieved. However, this methodology
in higher-level athletes. The fact that we found no correla- was chosen because it is similar to the way screening
tion between the CKCUEST and UQYBT may signify examinations are often performed in real-life situations
that in Division I athletes, the 2 tests measure different to efficiently accommodate a large numbers of athletes
sets of constructs and that both may be warranted for in a limited time window. While the normative values
inclusion during screening or rehabilitation activities. reported in this manuscript can be used for comparison
Gorman et al24 contend that the CKCUEST requires against other Division I athletes, it is important to note the

JSR Vol. 25, No. 2, 2016

 Upper-Extremity Physical-Performance Tests   153

makeup of the sample. Athletes from 11 different teams after total knee arthroplasty requires both performance-
composed the study population, yet there was a limited based and patient-report assessments: a longitudinal
number of male basketball players and no football, analysis of outcomes. J Arthroplasty. 2011;26:728–737.
wrestling, softball, or field hockey players included, as PubMed doi:10.1016/j.arth.2010.06.004
these sports are not intercollegiate teams at our institu- 7. Waters E. Suggestions from the field for return to
tion. Given that some sport-specific differences were sports participation following anterior cruciate ligament
observed for each test, it may be more appropriate for reconstruction: basketball. J Orthop Sports Phys Ther.
future studies to gather normative data by sport, rather 2012;42:326–336. PubMed doi:10.2519/jospt.2012.4030
than collectively for heterogeneous groups. In addition, 8. Dallinga JM, Benjaminse A, Lemmink KA. Which
we administered the CKCUEST with 2 modifications screening tools can predict injury to the lower extremi-
relative to prior publications. Based on time constraints ties in team sports?: a systematic review. Sports Med.
and previous experience we only recorded the first 2012;42:791–815. PubMed doi:10.1007/BF03262295
CKCUEST trial and normalized upper-extremity position 9. Plisky PJ, Rauh MJ, Kaminski TW, Underwood FB. Star
to shoulder width for each individual. While we believe Excursion Balance Test as a predictor of lower extremity
these modifications improve the overall nature of the test, injury in high school basketball players. J Orthop Sports
caution is warranted when comparing our results with Phys Ther. 2006;36:911–919. PubMed doi:10.2519/
those previously reported. jospt.2006.2244
Downloaded by California State Univ Fresno on 12/03/16, Volume 25, Article Number 2

10. Pontillo M, Spinelli BA, Sennett BJ. Prediction of in-

season shoulder injury from preseason testing in Division
Conclusion I collegiate football players. Sports Health. 2014;6:497–
503. PubMed doi:10.1177/1941738114523239
Data presented in this study can be regarded as a general 11. Halter JB, Reuben DB. Indicators of function in the geri-
representation of CKCUEST and UQYBT scores in a atric population. In: Finch CE, Vaupel JW, Kinsella K,
heterogeneous population of Division I college athletes. eds. Cells and Surveys: Should Biological Measures Be
CKCUEST and UQYBT results were not correlated, Included in Social Science Research? Washington, DC:
suggesting that the tests measure different sets of per- National Academies Press; 2001.
formance constructs and that both may be valuable in 12. Bolgla LA, Keskula DR. Reliability of lower extrem-
rehabilitation and injury-prevention settings. Statistically ity functional performance tests. J Orthop Sports
significant differences in average values of the UQYBT Phys Ther. 1997;26:138–142. PubMed doi:10.2519/
and CKCUEST scores were found to be present between jospt.1997.26.3.138
sexes and athletes of different sports. Injury-prevention 13. Hootman JM, Dick R, Agel J. Epidemiology of collegiate
efforts would benefit from future research focused on injuries for 15 sports: summary and recommendations for
examining the ability of these PPTs to predict athletic injury prevention initiatives. J Athl Train. 2007;42:311–
injuries in the college population. 319. PubMed
14. Sytema R, Dekker R, Dijkstra PU, Ten Duis HJ, Van Der
Sluis CK. Upper extremity sports injury: risk factors in
References comparison to lower extremity injury in more than 25
1. Stratford PW, Kennedy DM. Performance measures were 000 cases. Clin J Sport Med. 2010;20:256–263. PubMed
necessary to obtain a complete picture of osteoarthritic doi:10.1097/JSM.0b013e3181e71e71
patients. J Clin Epidemiol. 2006;59:160–167. PubMed 15. Agel J, Evans TA, Dick R, Putukian M, Marshall SW.
doi:10.1016/j.jclinepi.2005.07.012 Descriptive epidemiology of collegiate men’s soccer
2. Curb JD, Ceria-Ulep CD, Rodriguez BL, et al. Perfor- injuries: National Collegiate Athletic Association Injury
mance-based measures of physical function for high- Surveillance System, 1988–1989 through 2002–2003. J
function populations. J Am Geriatr Soc. 2006;54:737–742. Athl Train. 2007;42:270–277. PubMed
PubMed doi:10.1111/j.1532-5415.2006.00700.x 16. Dick R, Putukian M, Agel J, Evans TA, Marshall SW.
3. Simmonds MJ. Measuring and managing pain and Descriptive epidemiology of collegiate women’s soccer
performance. Man Ther. 2006;11:175–179. PubMed injuries: National Collegiate Athletic Association Injury
doi:10.1016/j.math.2006.03.002 Surveillance System, 1988–1989 through 2002–2003. J
4. Juhakoski R, Tenhonen S, Anttonen T, Kauppinen T, Aro- Athl Train. 2007;42:278–285. PubMed
koski JP. Factors affecting self-reported pain and physical 17. Dick R, Romani WA, Agel J, Case JG, Marshall SW.
function in patients with hip osteoarthritis. Arch Phys Med Descriptive epidemiology of collegiate men’s lacrosse
Rehabil. 2008;89:1066–1073. PubMed doi:10.1016/j. injuries: National Collegiate Athletic Association Injury
apmr.2007.10.036 Surveillance System, 1988–1989 through 2003–2004. J
5. van den Akker-Scheek I, Zijlstra W, Groothoff JW, Bulstra Athl Train. 2007;42:255–261. PubMed
SK, Stevens M. Physical functioning before and after total 18. Dick R, Sauers EL, Agel J, et al. Descriptive epidemiology
hip arthroplasty: perception and performance. Phys Ther. of collegiate men’s baseball injuries: National Collegiate
2008;88:712–719. PubMed doi:10.2522/ptj.20060301 Athletic Association Injury Surveillance System, 1988–
6. Mizner RL, Petterson SC, Clements KE, Zeni JA, Jr, Irrgang 1989 through 2003–2004. J Athl Train. 2007;42:183–193.
JJ, Snyder-Mackler L. Measuring functional improvement PubMed

JSR Vol. 25, No. 2, 2016

 154  Taylor et al

19. Teyhen DS, Shaffer SW, Lorenson CL, et al. The functional 26. Schneiders AG, Davidsson A, Horman E, Sullivan SJ.
movement screen: a reliability study. J Orthop Sports Functional movement screen normative values in a young,
Phys Ther. 2012;42:530–540. PubMed doi:10.2519/ active population. Int J Sports Phys Ther. 2011;6:75–82.
jospt.2012.3838 PubMed
20. Negrete RJ, Hanney WJ, Kolber MJ, et al. Reliability, 27. Roush JR, Kitamura J, Waits MC. Reference values for
minimal detectable change, and normative values for tests the Closed Kinetic Chain Upper Extremity Stability Test
of upper extremity function and power. J Strength Cond (CKCUEST) for collegiate baseball players. N Am J Sports
Res. 2010;24:3318–3325. Phys Ther. 2007;2:159–163. PubMed
21. Prokopy MP, Ingersoll CD, Nordenschild E, Katch FI, 28. Tarara DT, Hegedus EJ, Taylor JB. Real-time test–retest
Gaesser GA, Weltman A. Closed-kinetic chain upper-body and interrater reliability of select physical performance
training improves throwing performance of NCAA Divi- measures in physically active college-aged students. Int J
sion I softball players. J Strength Cond Res. 2008;22:1790– Sports Phys Ther. 2014;9:874–887. PubMed
1798. PubMed doi:10.1519/JSC.0b013e318185f637 29. Cook C. Use of effect sizes in describing data [Clinimetrics
22. Goldbeck TG, Davies GJ. Test-retest reliability of the Corner]. J Manual Manip Ther. 2008;16(3):54E–57E.
Closed Kinetic Chain Upper Extremity Stability Test: a doi:10.1179/jmt.2008.16.3.54E
clinical field test. J Sport Rehabil. 2000;9:35–45. 30. Ellenbecker T, Manske R, Davies G. Closed kinetic chain
23. Negrete RJ, Hanney WJ, Kolber MJ, Davies GJ, Riemann testing techniques of the upper extremities. Orthop Phys
Downloaded by California State Univ Fresno on 12/03/16, Volume 25, Article Number 2

B. Can upper extremity functional tests predict the softball Ther Clin N Am. 2000;9:219–229.
throw for distance: a predictive validity investigation. Int 31. Abrams GD, Harris JD, Gupta AK, et al. Functional perfor-
J Sports Phys Ther. 2011;6:104–111. PubMed mance testing after anterior cruciate ligament reconstruc-
24. Gorman PP, Butler RJ, Plisky PJ, Kiesel KB. Upper tion: a systematic review. Orthop J Sports Med. 2014;2.
Quarter Y Balance Test: Reliability and performance PubMed doi:10.1177/2325967113518305:1-10
comparison between genders in active adults. J Strength 32. Clanton TO, Matheny LM, Jarvis HC, Jeronimus
Cond Res. 2012;26:3043–3048. PubMed doi:10.1519/ AB. Return to play in athletes following ankle
JSC.0b013e3182472fdb injuries. Sports Health. 2012;4:471–474. PubMed
25. Westrick RB, Miller JM, Carow SD, Gerber JP. Explora- doi:10.1177/1941738112463347
tion of the Y-Balance Test for assessment of upper quarter
closed kinetic chain performance. Int J Sports Phys Ther.
2012;7:139–147. PubMed

JSR Vol. 25, No. 2, 2016