Original Research

Intrarater Reliability of Muscle Strength and

Hamstring to Quadriceps Strength Imbalance
Ratios During Concentric, Isometric, and
Eccentric Maximal Voluntary Contractions
Using the Isoforce Dynamometer
Anett Mau-Moeller, PhD,*† Martin Gube, MS,† Sabine Felser, PhD,† Frank Feldhege, Dipl.-Ing.,*
Matthias Weippert, PhD,† Florian Husmann, MS,† Thomas Tischer, MD,* Rainer Bader, Dipl.-Ing., MD,*
Sven Bruhn, PhD,† and Martin Behrens, PhD†

Abstract
Objective: To determine intrasession and intersession reliability of strength measurements and hamstrings to quadriceps
strength imbalance ratios (H/Q ratios) using the new isoforce dynamometer. Design: Repeated measures. Setting: Exercise
science laboratory. Participants: Thirty healthy subjects (15 females, 15 males, 27.8 years). Main Outcome Measures:
Coefficient of variation (CV) and intraclass correlation coefficients (ICC) were calculated for (1) strength parameters, that is peak
torque, mean work, and mean power for concentric and eccentric maximal voluntary contractions; isometric maximal voluntary
torque (IMVT); rate of torque development (RTD), and (2) H/Q ratios, that is conventional concentric, eccentric, and isometric H/Q
ratios (Hcon/Qcon at 60 deg/s, 120 deg/s, and 180 deg/s, Hecc/Qecc at 260 deg/s and Hiso/Qiso) and functional eccentric antagonist
to concentric agonist H/Q ratios (Hecc/Qcon and Hcon/Qecc). High reliability: CV ,10%, ICC .0.90; moderate reliability: CV between
10% and 20%, ICC between 0.80 and 0.90; low reliability: CV .20%, ICC ,0.80. Results: (1) Strength parameters: (a) high
intrasession reliability for concentric, eccentric, and isometric measurements, (b) moderate-to-high intersession reliability for
concentric and eccentric measurements and IMVT, and (c) moderate-to-high intrasession reliability but low intersession reliability for
RTD. (2) H/Q ratios: (a) moderate-to-high intrasession reliability for conventional ratios, (b) high intrasession reliability for functional
ratios, (c) higher intersession reliability for Hcon/Qcon and Hiso/Qiso (moderate to high) than Hecc/Qecc (low to moderate), and (d) higher
intersession reliability for conventional H/Q ratios (low to high) than functional H/Q ratios (low to moderate). Conclusions: The
results have confirmed the reliability of strength parameters and the most frequently used H/Q ratios.
Key Words: torque, work, asymmetry, H/Q ratio
(Clin J Sport Med 2017;0:1–9)

INTRODUCTION (RTD). Before a new dynamometer can be used in sports,

research, or clinical settings, its reliability must be established to
Isokinetic and isometric assessments of muscle function are
ensure the consistency of measurements.11 Several studies
commonly used in sports, clinical practice, and research to
analyzed the intrasession and intersession reliability of various
monitor training and therapy interventions. The first dyna-
brands of isokinetic devices and most have demonstrated that
mometer with isokinetic mode was developed in the late
the device tested has high reliability.1–10 Most reliability studies
1960s1 and, since then, various manufactures have introduced
have concentrated on the measurement of knee extension and
different devices.2–10 Isokinetic dynamometers are used to
flexion strength. However, only a few have investigated the
measure the concentric and eccentric muscle strength of single
reliability of intralimb hamstrings to quadriceps muscle
muscle groups throughout the full range of motion at
strength imbalance ratios (H/Q ratios) that are frequently used
a constant angular velocity. Furthermore, static strength
to describe muscle function at the knee joint.4,7,10,12
measurements can be performed to assess isometric maximal
The purpose of this study was to investigate the absolute and
voluntary torque (IMVT) and the rate of torque development
relative intrasession and intersession reliability of isokinetic and
isometric strength measurements and the less-investigated ham-
Submitted for publication October 13, 2016; accepted April 6, 2017.
strings to quadriceps strength imbalance ratios (H/Q ratios) using
From the *Department of Orthopedics, University Medicine Rostock, Rostock,
Germany; and †Department of Sports Science, University of Rostock, Rostock,
the new isoforce dynamometer (TUR GmbH, Berlin, Germany).
Germany.
The authors report no conflicts of interest.
METHODS
Corresponding Author: Anett Mau-Moeller, PhD, Department of Orthopaedics,
University of Rostock, Doberaner Strasse 142, 18057 Rostock, Germany Subjects
(anett.mau-moeller@uni-rostock.de).
Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved. Thirty healthy subjects (15 females, 15 males, age: 27.8 6 3.3
http://dx.doi.org/10.1097/JSM.0000000000000493 years, height: 1.80 6 0.10 m, weight: 73.5 6 12.4 kg, physical

Volume 0
· Number 0
· Month 2017 www.cjsportmed.com 1

Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
A. Mau-Moeller et al. (2017) Clin J Sport Med

activity: 5.0 6 3.4 hours per week) with no history of each test series, subjects performed 2 to 3 submaximal
cardiovascular, neurological and musculoskeletal disorders or familiarization trials.
injuries participated in 2 experimental sessions separated by 7 The concentric protocol consisted of 3 maximal knee
days; both sessions took place at the same time of the day extensions and flexions at 3 different angular velocities; these
(within-subject design; repeated measures design). Subjects were were performed in the following order: 60, 120, and 180 deg/
instructed to maintain their regular physical activities and not to s.13 Range of motion was set at 70 degrees (from 10 to 80
take part in any intensive training for the duration of the study degrees of knee flexion; 0 degrees 5 full extension). The
period. They were further instructed to refrain from consuming subjects were thoroughly instructed to push (for the extension)
alcohol and caffeine in the 72 hours preceding the experiments and pull (for the flexion) as forcefully and as fast as possible
and not to perform any strenuous exercise in the 72 hours before and to complete the full range of motion. There was a 1-minute
the measurements. Before taking part in the study, all rest period between the test series.
participants provided written consent to participation. Isometric measurements were made at 60-degree knee
flexion.14 The subjects performed 2 knee extension trials,
followed by 2 knee flexion trials with a 1-minute rest period in
Procedures between. They were asked to push and pull as fast and as
All data were collected by the same investigator to avoid forcefully as possible and to maintain maximal force for 4 to 5
intertester variability. Subjects were comfortably seated in seconds.
a standardized position on the isoforce dynamometer with Maximal eccentric voluntary contractions were also
a hip angle of 90 degrees. Measurements were taken performed between 10 and 80 degrees of knee flexion at
unilaterally on the dominant leg [right leg: n 5 24 (80%)], a single velocity of 260 deg/s. The eccentric protocol consisted
which was defined as the limb used to kick a ball. The axis of of 3 knee extension and 3 knee flexion trials. The subjects were
the dynamometer was aligned with the knee joint rotation axis asked to use maximal force to resist the lever arm over the
(lateral femoral epicondyle). Velcro straps across the waist complete range of motion of 70 degrees. After an eccentric
and shoulder prevented excessive movements. The shank was trial was completed, the lever arm was moved passively back
fixed to the lever arm of the dynamometer 2 to 3 cm above the to the starting position at a velocity of 30 deg/s. There was
lateral malleolus. The configurations of the seat and a 1-minute rest period between the test series.
dynamometer were documented to ensure that the subjects The subjects reported no discomfort or pain throughout the
were positioned identically during the second experimental experiments. The torque signals were digitized with a sampling
session. Gravity correction was performed before each test frequency of 1000 Hz and analyzed using the isoforce
series according to the manufacturer’s instructions. software (TUR GmbH, Berlin, Germany).
Subjects performed a warm-up protocol consisting of 8 to
10 submaximal concentric knee extensions and flexions at Outcome Measures
a velocity of 60 deg/s. The experimental protocol included Isokinetic and Isometric Strength Parameters
concentric, isometric, and eccentric muscle contractions and is
presented in Table 1. Peak torque, mean work, and mean power were calculated for
Subjects were instructed to fold their arms across their concentric and eccentric maximal voluntary contractions.
chests and to exert maximal force to extend and flex the knee Isometric maximal voluntary torque and RTD were de-
against the lever arm of the dynamometer. Standardized termined for isometric maximal voluntary contractions.8
strong, verbal encouragement was given by the investigator
throughout the execution of the movement. An observer H/Q Ratios
checked that the movement was performed without any
visible pretension or countermovement. Feedback on strength The conventional H/Q ratios were calculated by dividing the
performance was given to the subjects after each trial. Before peak torque produced by the hamstrings by the peak torque
produced by the quadriceps muscle during maximal concen-
tric (Hcon/Qcon at 60, 120, and 180 deg/s), isometric (Hiso/
Qiso), and eccentric (Hecc/Qecc at 260 deg/s) voluntary
TABLE 1. Experimental Protocol contractions.15,16 The H/Q ratio was expressed functionally
Contraction Muscle as the ratio of eccentric antagonist peak torque to concentric
Mode Group Velocity Trials ROM/Position agonist peak torque (knee extension: Hecc 260/Qcon 60; knee
Concentric KE 1 KF 60 deg/s 8-10 70-degree ROM flexion: Hcon 60/Qecc 260).17,18
warm-up
Concentric KE 1 KF 60 deg/s 3 70-degree ROM Statistical Analysis
KE 1 KF 120 deg/s 3 70-degree ROM
Further analysis was conducted on the 2 trials in each session
KE 1 KF 180 deg/s 3 70-degree ROM with the highest peak torque for each contraction mode and
Isometric KE 0 deg/s 2 60-degree knee angular velocity. The normality of the distribution of these
flexion data was checked with the Kolmogorov–Smirnov test. Intra-
KF 0 deg/s 2 60-degree knee session reliability (trial 1 vs trial 2) and intersession reliability
flexion (mean of trials 1 and 2 from session 1 vs the mean of trials 1
Eccentric KE 260 deg/s 3 70-degree ROM and 2 from session 2) were calculated using an Excel
KF 260 deg/s 3 70-degree ROM spreadsheet developed by Hopkins.19 The typical error of
measurement and coefficient of variation (CV) were calcu-
KE, knee extensors; KF, knee flexors; ROM, range of motion.
lated to provide measures of absolute reliability, representing

Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Volume 0
· Number 0
· Month 2017 www.cjsportmed.com

TABLE 2. Intrasession Reliability of the Knee Extension Measurements

Trial 1, Mean (SD) Trial 2, Mean (SD) Mean Difference (95% CI) SDDiff TE (95% CI) CV (95% CI), % ICC (95% CI)
Peak torque, N·m
60 deg/s 190.7 (51.0) 187.9 (48.4) 22.8 (25.6 to 0.0) 7.5 5.3 (4.2-7.2) 2.8 (2.2-3.8) 0.99 (0.98-0.99)
120 deg/s 157.6 (43.7) 156.4 (41.8) 21.2 (23.8 to 1.5) 7.2 5.1 (4.0-6.8) 3.4 (2.7-4.6) 0.99 (0.97-0.99)
180 deg/s 143.4 (42.5) 141.3 (39.5) 22.1 (24.5 to 0.3) 6.4 4.5 (3.6-6.1) 3.6 (2.9-4.9) 0.99 (0.98-0.99)
260 deg/s 232.2 (71.4) 235.0 (73.4) 2.8 (23.8 to 9.5) 17.8 12.6 (10.0-16.9) 5.7 (4.5-7.8) 0.97 (0.94-0.99)
Average work, J
60 deg/s 152.4 (45.8) 151.6 (44.2) 20.8 (23.9 to 2.3) 8.2 5.8 (4.6-7.8) 3.9 (3.1-5.3) 0.98 (0.97-0.99)
120 deg/s 134.7 (42.9) 137.1 (40.0) 2.4 (21.6 to 6.3) 10.7 7.6 (6.0-10.2) 7.0 (5.5-9.5) 0.97 (0.94-0.99)
180 deg/s 118.9 (39.4) 120.0 (36.2) 1.1 (22.1 to 4.4) 8.7 6.1 (4.9-8.3) 6.4 (5.1-8.7) 0.98 (0.95-0.99)
260 deg/s 173.2 (53.5) 171.4 (57.1) 21.8 (27.0 to 3.3) 13.8 9.8 (7.8-13.1) 7.0 (5.5-9.5) 0.97 (0.94-0.99)
Average power, W
60 deg/s 117.0 (31.9) 117.7 (32.8) 0.7 (21.7 to 3.1) 6.5 4.6 (3.7-6.2) 4.2 (3.3-5.6) 0.98 (0.96-0.99)
120 deg/s 169.5 (53.5) 172.3 (51.2) 2.8 (22.6 to 8.3) 14.6 10.4 (8.3-13.9) 6.2 (4.9-8.4) 0.96 (0.93-0.98)
180 deg/s 192.7 (69.1) 194.0 (63.8) 1.3 (24.5 to 7.2) 15.6 11.0 (8.8-14.8) 7.0 (5.5-9.5) 0.97 (0.95-0.99)
260 deg/s 138.9 (46.6) 137.4 (49.4) 21.5 (26.0 to 3.0) 12.0 8.5 (6.8-11.5) 7.3 (5.8-10.0) 0.97 (0.94-0.99)
IMVT, N·m
207.9 (59.0) 208.6 (59.8) 0.7 (25.0 to 6.3) 15.0 10.6 (8.5-14.3) 5.3 (4.2-7.1) 0.97 (0.94-0.99)
RTD, N·m·s21
720.4 (292.6) 685.0 (304.8) 235.4 (279.1 to 8.9) 117.8 83.3 (66.4-112.0) 13.2 (10.4-18.2) 0.93 (0.85-0.96)
SDDiff, SD of the difference between trials 1 and 2; TE, typical error; SD, standard deviation; CI, confidence interval; CV, coefficient of variation; ICC, intraclass correlation coefficient; IMVT,
isometric maximal voluntary torque; RTD, rate of torque development during isometric maximum voluntary contraction.

the degree of variability in repeated measurements for a given in the rank order of all the subjects in a retest.11 Intraclass
individual (within-subject variation).11 Relative reliability correlation coefficients and CV values were classified as
was estimated using the intraclass correlation coefficient follows: CV ,10% and ICC .0.90 were considered as high,
(ICC) with 95% confidence level that represents the variation CV between 10% and 20% and ICC between 0.80 and 0.90 as

TABLE 3. Intrasession Reliability of the Knee Flexion Measurements

Trial 1, Mean (SD) Trial 2, Mean (SD) Mean Difference (95% CI) SDDiff TE (95% CI) CV (95% CI), % ICC (95% CI)
Peak torque, N·m
60 deg/s 108.4 (35.0) 107.6 (35.6) 20.8 (22.2 to 0.6) 3.7 2.6 (2.1-3.5) 2.8 (2.2-3.8) 0.99 (0.99-1.00)
120 deg/s 95.8 (31.3) 93.4 (29.6)* 22.4 (24.2 to 20.6) 4.8 3.4 (2.7-4.5) 3.9 (3.1-5.3) 0.99 (0.98-0.99)
180 deg/s 87.1 (28.8) 86.6 (28.3) 20.5 (22.4 to 1.4) 5.0 3.6 (2.8-4.8) 3.9 (3.1-5.2) 0.99 (0.97-0.99)
260 deg/s 133.2 (47.3) 131.2 (45.7) 22.0 (24.9 to 0.8) 7.7 5.4 (4.3-7.3) 4.3 (3.4-5.8) 0.99 (0.97-0.99)
Average work, J
60 deg/s 93.5 (35.1) 92.2 (35.5)* 21.3 (22.4 to 20.2) 2.9 2.1 (1.7-2.8) 2.7 (2.2-3.7) 1.00 (0.99-1.00)
120 deg/s 80.9 (31.0) 79.4 (29.8) 21.5 (23.4 to 0.5) 5.1 3.6 (2.9-4.9) 4.6 (3.7-6.2) 0.99 (0.97-0.99)
180 deg/s 68.6 (26.1) 68.9 (27.3) 0.3 (21.7 to 2.4) 5.6 3.9 (3.1-5.3) 5.2 (4.2-7.1) 0.98 (0.96-0.99)
260 deg/s 108.6 (42.3) 106.9 (40.6) 21.7 (24.2 to 0.7) 6.6 4.7 (3.7-6.3) 4.9 (3.9-6.7) 0.99 (0.98-0.99)
Average power, W
60 deg/s 71.4 (25.0) 69.1 (25.2)** 22.3 (23.7 to 21.0) 3.7 2.6 (2.1-3.5) 4.0 (3.2-5.4) 0.99 (0.98-1.00)
120 deg/s 105.4 (39.9) 101.1 (37.8)* 24.3 (27.9 to 20.8) 9.5 6.7 (5.4-9.0) 6.7 (5.3-9.1) 0.97 (0.94-0.99)
180 deg/s 118.1 (50.1) 114.7 (46.9) 23.4 (27.5 to 0.6) 10.8 7.7 (6.1-10.3) 6.1 (4.9-8.3) 0.98 (0.95-0.99)
260 deg/s 90.6 (36.6) 90.0 (35.4) 20.6 (22.7 to 1.5) 5.7 4.1 (3.2-5.4) 5.4 (4.3-7.3) 0.99 (0.98-0.99)
IMVT, N·m
104.1 (33.4) 101.3 (33.5)* 22.8 (25.1 to 20.4) 6.3 4.4 (3.5-6.0) 4.0 (3.2-5.4) 0.98 (0.97-0.99)
21
RTD, N·m·s
403.2 (164.3) 379.8 (180.9) 223.4 (253.0 to 6.1) 79.1 56.0 (44.6-75.2) 14.6 (11.5-20.1) 0.90 (0.80-0.95)
*P , 0.050; **P , 0.010 (significantly lower than trial 1).
SDDiff, SD of the difference between trials 1 and 2; TE, typical error; SD, standard deviation; CI, confidence interval; CV, coefficient of variation; ICC, intraclass correlation coefficient; IMVT,
isometric maximal voluntary torque; RTD, rate of torque development during isometric maximum voluntary contraction.

Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
A. Mau-Moeller et al. (2017) Clin J Sport Med

TABLE 4. Intrasession Reliability of the H/Q Strength Imbalance Ratios

Trial 1, Mean (SD) Trial 2, Mean (SD) Mean Difference (95% CI) SDDiff TE (95% CI) CV (95% CI) ICC (95% CI)
Conventional H/Q ratios
Hcon/Qcon at 60 deg/s 0.57 (0.09) 0.57 (0.08) 0.00 (20.01 to 0.01) 0.04 0.03 (0.02-0.04) 4.4 (3.5-6.0) 0.91 (0.82-0.96)
Hcon/Qcon at 120 deg/s 0.61 (0.09) 0.59 (0.08) 20.02 (20.03 to 0.00) 0.04 0.03 (0.02-0.04) 4.3 (3.4-5.9) 0.90 (0.80-0.95)
Hcon/Qcon at 180 deg/s 0.61 (0.09) 0.61 (0.09) 0.00 (20.02 to 0.02) 0.05 0.03 (0.03-0.04) 5.8 (4.6-7.9) 0.87 (0.74-0.94)
Hecc/Qecc at 260 deg/s 0.58 (0.14) 0.57 (0.13) 20.01 (20.03 to 0.01) 0.06 0.04 (0.03-0.05) 6.7 (5.3-9.1) 0.92 (0.84-0.96)
Hiso/Qiso at 0 deg/s 0.51 (0.10) 0.49 (0.09) 20.02 (20.03 to 0.00) 0.05 0.03 (0.03-0.05) 6.3 (5.0-8.6) 0.89 (0.78-0.95)
Functional H/Q ratios
Hecc 260/Qcon 60 0.69 (0.12) 0.69 (0.11) 0.00 (20.02 to 0.02) 0.05 0.03 (0.03-0.02) 5.2 (4.1-7.1) 0.92 (0.84-0.96)
Hcon 60/Qecc 260 0.48 (0.12) 0.47 (0.12) 20.01 (20.03 to 0.00) 0.04 0.03 (0.02-0.04) 6.5 (5.2-8.9) 0.95 (0.90-0.98)
ecc, eccentric; H/Q ratio, hamstring/quadriceps ratio; iso, isometric; SDDiff, SD of the difference between trials 1 and 2; TE, typical error; SD, standard deviation; CI, confidence interval;
CV, coefficient of variation; ICC, intraclass correlation coefficient.

moderate and CV .20% and ICC ,0.80 as low reliabil- RESULTS

ity.20,21 Between-trial and between-session differences were
assessed with paired Student’s t tests. Data were analyzed
using the SPSS statistical package 20.0 (SPSS Inc, Chicago,
Illinois) and the criterion for significance was set at P # 0.050.
Ethical Considerations
The study was performed according to the principles of the
Declaration of Helsinki and approved by the Ethical Review
Committee of the University of Rostock.
Intrasession Reliability
Isokinetic and Isometric Strength Parameters
Absolute and relative intrasession reliability for knee extension
and flexion measurements was high: CV values ranged from
2.7% to 7.3% and ICC values ranged from 0.96 to 0.99 for
concentric and eccentric peak torque, average work, and average
power at all angular velocities and IMVT (Tables 2 and 3). The
highest CV values (.10%) were those for RTD, indicating that
absolute intrasession reliability for knee extension and flexion
measurements was moderate although relative intrasession
reliability for RTD was high, with ICC values of $0.90
(Tables 2 and 3). There were significant between-trial differences
for knee flexion measurements with respect to peak torque at 120
deg/s, average work at 60 deg/s, average power at 60 deg/s and
120 deg/s and IMVT (Table 3). A summary of data on the
intrasession reliability of concentric, eccentric, and isometric
knee extension and flexion measurements is given in Table 8.

H/Q Ratios
Coefficient of variation values for conventional H/Q ratios
ranged between 4.3% and 6.7% and ICC values ranged from
0.87 to 0.92, indicating moderate-to-high intrasession re-
liability. Functional H/Q ratios also had high absolute and
relative intrasession reliability, with CV values below 6.5%
and ICC values above 0.92 (Table 4 and Figure 1). A summary
of data on the intrasession reliability of conventional and
functional H/Q ratios is given in Table 9.

Figure 1. Absolute and relative intrasession (A) and intersession (B) re-
liability of the conventional H/Q ratios. CON; concentric; CV, coefficient of
variation; ECC, eccentric; ICC, intraclass correlation coefficient; ISO,
isometric.
Intersession Reliability
Isokinetic and Isometric Strength Parameters
Relative intersession reliability of knee extension and flexion
measurements was high with respect to concentric peak
torque, average work, and average power at 60, 120, and 180
deg/s, with ICC values between 0.91 and 0.98. The absolute
intersession reliability of concentric knee flexion and exten-
sion measurements at all angular velocities was high for peak
torque (CV: 4.9%-8.7%) and moderate for average work
(CV: 7.0%-13.3%). However, the absolute intersession
reliability was high with respect to the average power of knee
extension measurements (CV: 8.6%-9.8%) and moderate

Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Volume 0
· Number 0
· Month 2017 www.cjsportmed.com

TABLE 5. Intersession Reliability of the Knee Extension Measurements

Mean Difference
Session 1, Mean (SD) Session 2, Mean (SD) (95% CI) SDDiff TE (95% CI) CV (95% CI), % ICC (95% CI)
Peak torque, N·m
60 deg/s 185.6 (52.1) 189.3 (49.6) 3.7 (23.6 to 11.0) 19.5 13.8 (11.0-18.6) 7.7 (6.1-10.5) 0.93 (0.86-0.97)
120 deg/s 151.5 (44.2) 157.0 (42.6)* 5.5 (0.9 to 10.1) 12.2 8.6 (6.9-11.6) 6.5 (5.1-8.8) 0.96 (0.92-0.98)
180 deg/s 138.4 (39.0) 142.3 (40.9)* 3.9 (0.5 to 7.4) 9.2 6.5 (5.2-8.8) 4.9 (3.9-6.6) 0.98 (0.95-0.99)
260 deg/s 225.8 (81.4) 233.6 (71.9) 7.8 (24.6 to 20.3) 33.2 23.5 (18.7-31.6) 12.4 (9.7-16.9) 0.91 (0.82-0.96)
Average work, J
60 deg/s 153.7 (47.7) 152.0 (44.8) 21.7 (29.2 to 5.8) 20.1 14.2 (11.3-19.1) 10.4 (8.2-14.2) 0.91 (0.82-0.96)
120 deg/s 132.6 (43.0) 135.9 (41.1) 3.3 (22.7 to 9.4) 16.2 11.5 (9.1-15.4) 10.3 (8.1-14.1) 0.93 (0.86-0.97)
180 deg/s 118.2 (38.2) 119.5 (37.6) 1.3 (22.8 to 5.3) 10.9 7.7 (6.2-10.4) 7.0 (5.5-9.5) 0.96 (0.92-0.98)
260 deg/s 163.4 (59.5) 172.3 (54.9) 8.9 (20.4 to 18.3) 24.9 17.6 (14.1-23.7) 13.3 (10.5-18.3) 0.91 (0.82-0.96)
Average power, W
60 deg/s 112.9 (34.1) 117.4 (32.2) 4.5 (20.5 to 9.4) 13.2 9.4 (7.5-12.6) 8.9 (7.0-12.1) 0.93 (0.85-0.96)
120 deg/s 164.2 (59.2) 170.9 (51.9) 6.7 (20.6 to 13.9) 19.5 13.8 (11.0-18.5) 9.8 (7.8-13.5) 0.94 (0.88-0.97)
180 deg/s 189.3 (71.2) 193.4 (66.0) 4.1 (23.0 to 11.0) 18.7 13.3 (10.6-17.8) 8.6 (6.8-11.7) 0.97 (0.93-0.98)
260 deg/s 130.1 (51.3) 138.2 (47.6)* 8.1 (0.2 to 16.0) 21.2 15.0 (11.9-20.1) 13.7 (10.8-18.9) 0.91 (0.83-0.96)
IMVT, N·m
208.5 (60.0) 208.2 (58.9) 20.3 (28.3 to 7.7) 21.4 15.1 (12.1-20.3) 8.3 (6.6-11.4) 0.94 (0.88-0.97)
RTD, N·m·s21
676.6 (266.0) 702.8 (292.9) 26.2 (244.6 to 97.1) 189.8 134.2 (106.9-180.4) 20.1 (15.7-27.9) 0.78 (0.59-0.89)
* P , 0.050 (significantly higher than session 1).
SDDiff, SD of the difference between sessions 1 and 2; TE, typical error; SD, standard deviation; CI, confidence interval; CV, coefficient of variation; ICC, intraclass correlation coefficient; IMVT,
isometric maximal voluntary torque; RTD, rate of torque development during isometric maximum voluntary contraction.

with respect to the average power of knee flexion measure-
ments (CV: 0.9%-13.8%) (Tables 5 and 6).
Absolute intersession reliability for knee flexion and
extension measurements was moderate with respect to
eccentric peak torque, average work, and average power at
a velocity of 260 deg/s (CV: 10.3%-19.2%) whilst relative
reliability was high (ICC: 0.91-0.94). Measurements of IMVT
showed moderate absolute intersession reliability and high
relative intersession reliability (Tables 5 and 6).
Rate of torque development had the highest CV values
(.20%) and lowest ICCs (,0.78), indicating that intersession
reliability for both measurements was low (Tables 5 and 6).
There were significant between-session differences in peak
torque, average work, and average power at all angular
velocities of knee flexion measurements (Table 6) and in peak
torque at 120 and 180 deg/s of knee extension measurements
(Table 5). A summary of the data on the intersession reliability
of concentric, eccentric, and isometric knee extension and
flexion measurements is given in Table 8.
H/Q Ratios
The absolute and relative intersession reliability of conven-
tional H/Q ratios was lower than the corresponding values for
intrasession reliability (Figure 1). The absolute reliability of
Hcon/Qcon at 60, 120, and 180 deg/s and Hiso/Qiso was high
(CV: 7.4%-8.3%), whilst relative intersession reliability was
low to moderate (ICC: 0.79-0.85). Hecc/Qecc at 260 deg/s had
moderate absolute and low relative intersession reliability
(Table 7 and Figure 1). There were significant between-session
differences in Hcon/Qcon at 60, 120, and 180 deg/s (Table 7).
The CVs and ICCs for functional H/Q ratios indicated that
intersession reliability was moderate for Hcon 60/Qecc 260
and low to moderate for Hecc 260/Qcon 60 (Table 8). A
summary of intersession reliability data for conventional and
functional H/Q ratios is given in Table 9.
DISCUSSION
The aim of this study was to determine the intrasession and
intersession reliability of measurements of isokinetic and
isometric strength and different hamstrings to quadriceps
strength imbalance ratios using the new isoforce dynamometer.
Isokinetic and Isometric Strength Parameters
Various studies have investigated the reliability of different
types of isokinetic dynamometers such as Cybex,7 Biodex,5
Con-Trex,8 Kin Com,10 Lido,6 Merac,3 iSAM,9 and Tech-
nogym.4 The values found for the reliability of isometric and
isokinetic strength parameters in this study are comparable to
those observed in previously.7,8 Measurements of peak
torque, average work, and average power for concentric and
eccentric knee extension and flexion measurements at angular
velocities of 60, 120, 180, and 260 deg/s showed high
intrasession reliability. Intersession reliability was high for
measurements of concentric knee extension and moderate to
high for measurements of concentric knee flexion. However,
whilst eccentric strength parameters had high relative in-
tersession reliability, their absolute intersession reliability was
only moderate; this might be related to difficulties in
maximally activating the muscles during eccentric

Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
A. Mau-Moeller et al. (2017) Clin J Sport Med

TABLE 6. Intersession Reliability of the Knee Flexion Measurements

Mean Difference
Session 1, Mean (SD) Session 2, Mean (SD) (95% CI) SDDiff TE (95% CI) CV (95% CI), % ICC (95% CI)
Peak torque, N·m
60 deg/s 100.7 (35.0) 108.0 (35.3)** 7.3 (3.3 to 11.4) 10.8 7.6 (6.1-10.3) 8.7 (6.9-11.9) 0.96 (0.91-0.98)
120 deg/s 89.3 (31.8) 94.6 (30.4)** 5.3 (2.5 to 8.0) 7.4 5.2 (4.2-7.2) 8.1 (6.4-11.0) 0.97 (0.95-0.99)
180 deg/s 82.5 (29.3) 86.8 (28.4)** 4.3 (1.7 to 7.0) 7.1 5.0 (4.0-6.8) 7.5 (6.0-10.3) 0.97 (0.94-0.99)
260 deg/s 130.9 (46.6) 132.2 (46.3) 1.3 (25.1 to 7.7) 17.1 12.1 (9.7-16.3) 10.3 (8.1-14.0) 0.94 (0.87-0.97)
Average work, J
60 deg/s 87.1 (35.2) 92.8 (35.3)* 5.7 (0.7 to 10.8) 13.5 9.6 (7.6-12.9) 12.2 (9.6-16.8) 0.93 (0.86-0.97)
120 deg/s 75.2 (31.3) 80.1 (30.3)* 4.9 (1.5 to 8.4) 9.2 6.5 (5.2-8.8) 12.9 (10.2-17.8) 0.96 (0.91-0.98)
180 deg/s 64.7 (28.7) 68.7 (26.5)* 4.0 (1.0 to 7.1) 8.0 5.7 (4.5-7.6) 13.3 (10.5-18.3) 0.96 (0.92-0.98)
260 deg/s 104.0 (40.3) 107.7 (41.3) 3.7 (22.0 to 9.5) 15.3 10.8 (8.6-14.6) 12.4 (9.8-17.0) 0.93 (0.87-0.97)
Average power, W
60 deg/s 62.3 (23.8) 70.2 (25.0)** 7.9 (4.2 to 11.7) 10.1 7.2 (5.7-9.6) 13.8 (10.9-19.0) 0.92 (0.84-0.96)
120 deg/s 97.8 (40.2) 103.3 (38.5)* 5.5 (1.8 to 9.2) 10.0 7.1 (5.6-9.5) 10.9 (8.6-14.9) 0.97 (0.94-0.99)
180 deg/s 110.2 (48.9) 116.4 (48.2)* 6.2 (0.7 to 11.7) 14.8 10.5 (8.3-14.1) 13.3 (10.5-18.3) 0.96 (0.91-0.98)
260 deg/s 86.9 (36.8) 90.3 (35.9) 3.4 (22.6 to 9.4) 16.1 11.4 (9.1-15.3) 19.2 (15.0-26.7) 0.91 (0.82-0.95)
IMVT, N·m
101.7 (36.8) 102.7 (33.3) 1.0 (23.9 to 5.9) 13.2 9.3 (7.4-12.5) 10.8 (8.5-14.8) 0.93 (0.87-0.97)
RTD, N·m·s21
359.2 (173.1) 391.5 (168.2) 32.3 (212.2 to 76.8) 119.2 84.3 (67.1-113.3) 29.9 (23.2-42.2) 0.77 (0.57-0.88)
*P , 0.050; **P , 0.010 (significantly higher than session 1).
SDDiff, SD of the difference between sessions 1 and 2; TE, typical error; SD, standard deviation; CI, confidence interval; CV, coefficient of variation; ICC, intraclass correlation coefficient; IMVT,
isometric maximal voluntary torque; RTD, rate of torque development during isometric maximum voluntary contraction.

contractions. Measurements of IMVT for isometric knee
extension and flexion showed high intrasession reliability and
moderate to high intersession reliability. However, the
absolute intrasession reliability of RTD measurements for
both knee extension and flexion was only moderate although
the relative reliability was high; the intersession reliability of
RTD was low.
In conclusion, we found that measurements of isokinetic
and isometric strength parameters had high repeatability
within and between sessions. Intersession reliability was
generally slightly lower than intrasession reliability. The
results show (1) high intrasession reliability for
measurements of concentric, eccentric, and isometric knee
extension and flexion, (2) moderate-to-high intersession
reliability for concentric and eccentric measurements and
IMVT and (3) moderate-to-high intrasession reliability,
but low intersession reliability for RTD during isometric
measurements.
Conventional H/Q Ratios
The conventional H/Q ratios are frequently used to assess
muscle function at the knee joint,16,18,22 although relatively
few studies have examined the reliability of this

TABLE 7. Intersession Reliability of H/Q Strength Imbalance Ratios

Session 1, Session 2, Mean Difference,
Mean (SD) Mean (SD) (95% CI) SDDiff TE (95% CI) CV (95% CI) ICC (95% CI)
Conventional H/Q ratios
Hcon/Qcon at 60 deg/s 0.54 (0.08) 0.57 (0.08)* 0.03 (0.01 to 0.05) 0.06 0.04 (0.03-0.05) 7.4 (5.8-10.0) 0.79 (0.60-0.89)
Hcon/Qcon at 120 deg/s 0.58 (0.08) 0.60 (0.08)* 0.02 (0.00 to 0.04) 0.05 0.03 (0.03-0.05) 6.1 (4.9-8.3) 0.82 (0.66-0.91)
Hcon/Qcon at 180 deg/s 0.59 (0.08) 0.61 (0.09)* 0.02 (0.00 to 0.04) 0.05 0.03 (0.03-0.05) 6.1 (4.9-8.3) 0.85 (0.72-0.93)
Hecc/Qecc at 260 deg/s 0.60 (0.15) 0.57 (0.13) 20.03 (20.06 to 0.01) 0.10 0.07 (0.06-0.09) 11.8 (9.3-16.2) 0.76 (0.56-0.88)
Hiso/Qiso at 0 deg/s 0.49 (0.09) 0.50 (0.10) 0.01 (20.01 to 0.03) 0.05 0.04 (0.03-0.05) 8.3 (6.5-11.3) 0.84 (0.69-0.92)
Functional H/Q ratios
Hecc 260/Qcon 60 0.70 (0.11) 0.69 (0.11) 20.01 (20.05 to 0.03) 0.10 0.07 (0.06-0.09) 10.6 (8.3-14.4) 0.63 (0.36-0.81)
Hcon 60/Qecc 260 0.47 (0.14) 0.47 (0.12) 0.00 (20.02 to 0.04) 0.08 0.05 (0.04-0.07) 13.6 (10.7-18.6) 0.83 (0.68-0.92)
* P , 0.050 (significantly higher than session 1).
SDDiff, SD of the difference between sessions 1 and 2; TE, typical error; H/Q ratio, hamstring/quadriceps ratio; con, concentric; ecc, eccentric; iso, isometric; SD, standard deviation; CI,
confidence interval; CV, coefficient of variation; ICC, intraclass correlation coefficient.

Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Volume 0
· Number 0
· Month 2017 www.cjsportmed.com

TABLE 8. Summary of Results for Intrasession and Intersession Reliability of Concentric, Eccentric,
and Isometric Knee Extension and Flexion Measurements
Concentric Eccentric Isometric
Intrasession Intersession Intrasession Intersession Intrasession Intersession
abs. rel. abs. rel. abs. rel. abs. rel. abs. rel. abs. rel.
Knee extension measurements
Peak torque H H H H H H M H — — — —
Average work H H H H H H M H — — — —
Average power H H H H H H M H — — — —
IMVT — — — — — — — — H H H H
RTD — — — — — — — — M H L L
Knee flexion measurements
Peak torque H H H H H H M H — — — —
Average work H H M H H H M H — — — —
Average power H H M H H H M H — — — —
IMVT — — — — — — — — H H M H
RTD — — — — — — — — M H L L
abs., absolute reliability represented by the coefficient of variation (CV); H, high reliability (CV ,10%; ICC .0.90); L, low reliability (CV .20%; ICC ,0.80); M, moderate reliability (CV
between 10% and 20%; ICC between 0.80 and 0.90); rel., relative reliability represented by the intraclass correlation coefficient (ICC); IMVT, isometric maximal voluntary torque; RTD, rate of
torque development during isometric maximum voluntary contraction.

parameter.4,7,10,12 The most frequently reported conventional Overall, our findings indicate that for most of the H/Q
strength imbalance ratio is the concentric H/Q ratio at an ratios intrasession reliability is higher than that of intersession
angular velocity of 60 deg/s. A minimum Hcon/Qcon ratio of reliability. The absolute and relative intrasession reliability of
0.60 seems to be generally accepted as indicative of normative the various conventional H/Q ratios was high, with the
muscle balance in the context of injury prevention and exception of Hcon/Qcon at 180 deg/s and Hiso/Qiso, which
rehabilitation.15 Yeung et al16 have shown that athletes demonstrated moderate relative reliability.
with a Hcon/Qcon of less than 0.60 at an angular velocity of The absolute intersession reliability of Hcon/Qcon at 60,
180 deg/s have a 17 times higher risk of hamstring injury. 120, and 180 deg/s was high. However, the CVs (6.1%-
Most studies of the reliability of the H/Q ratio have 7.4%) are slightly higher than those reported by Impellizzeri
analyzed intersession reliability. However, different cutoff et al (5.2%-5.4%) in an investigation of the intersession
criteria for the interpretation of ICC values have been reliability of Hcon/Qcon at the same angular velocities as we
used.4,7,10,12 We have used the criteria provided by Vincent used.7 The same authors also found that the relative
when comparing our results with those of previous studies.21 intersession reliability of Hcon/Qcon was low, with ICCs
ranging between 0.34 and 0.79; they found that measure-
ments taken at slower angular velocities had better
TABLE 9. Summary of Results for Intrasession reliability than those taken at faster velocities. In contrast,
and Intersession Reliability of the we found that Hcon/Qcon had low-to-moderate relative
Conventional and Functional H/Q intersession reliability at all 3 angular velocities (ICC:
Ratios 0.79-0.85) and that velocity-related differences in reliability
Intrasession Intersession were less pronounced.
abs. rel. abs. rel. A more recent study of the intersession reliability of Hcon/
Qcon at 60, 180, and 240 deg/s by Ayala et al12 reported that
Conventional H/Q ratios
these parameters had moderate absolute reliability (CV:
Hcon/Qcon at 60 deg/s H H H L 16.3%-20.6%) and low relative reliability (ICC: 0.64-0.69).
Hcon/Qcon at 120 deg/s H H H M Ayala et al consistently reported lower repeatability for
Hcon/Qcon at 180 deg/s H M H M measurements made at angular velocities of 60 and 180 deg/s
Hecc/Qecc at 260 deg/s H H M L than that we found in this study. The lowest relative reliability
for Hcon/Qcon at 60 deg/s was reported by Sole et al10: an ICC
Hiso/Qiso at 0 deg/s H M H M
of 0.43.
Functional H/Q ratios To the authors’ knowledge only one study, by de Carvalho
Hecc 260/Qcon 60 H H M L Froufe Andrade et al,4 has investigated the intersession
Hcon 60/Qecc 260 H H M M reliability of the conventional H/Q ratio with respect to the
abs., absolute reliability represented by the coefficient of variation (CV); con, concentric; 3 different contractions modes. This study reported that Hcon/
ecc, eccentric; H, high reliability (CV ,10%; ICC .0.90); H/Q ratio, hamstring/ Qcon had high absolute reliability and moderate relative
quadriceps ratio; iso, isometric; L, low reliability (CV .20%; ICC ,0.80); M, moderate reliability at 60 deg/s (CV 5 4.7%; ICC 5 0.89), Hecc/Qecc had
reliability (CV between 10% and 20%; ICC between 0.80 and 0.90); rel., relative reliability
high reliability at 260 deg/s (CV 5 5.6%; ICC 5 0.91), and
represented by the intraclass correlation coefficient (ICC).
isometric measurements at 60 degrees of knee flexion had

Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
A. Mau-Moeller et al. (2017)
moderate reliability (CV 5 13.8%; ICC 5 0.87). We also
found, however, that the reliability of the H/Q ratios for
concentric and, especially, eccentric measurement was lower
than the values reported by Carvalho et al, with moderate
absolute reliability and low relative intersession reliability for
Hecc/Qecc at 260 deg/s.
Functional H/Q Ratios
The conventional H/Q ratio has some limitations as it is based
on the peak torque of knee extensors and flexors in the same
contraction mode. This situation does not occur during
functional movement because coactivation of these muscle
groups is characterized with opposing contraction modes.
Consequently, the “dynamic control ratio” is generally
regarded as being more useful than the conventional H/Q
ratio for determining an injury risk.17,18
The results indicate that Hecc 260/Qcon 60 and Hcon60/Qecc
260 have high absolute and relative intrasession reliability;
however, the intersession reliability of these functional
ratios was low to moderate. Most previous studies have
reported a higher reliability for Hecc 260/Qcon 60.4,7,10 The
reliability of this ratio was firstly studied by Sole et al10, who
reported that it had low relative reliability (ICC 5 0.73).
Impellizeri et al demonstrated that it had high absolute
reliability (CV 5 6.3%) and moderate relative reliability
(ICC 5 0.87). The highest reported value for the reliability of
Hecc 260/Qcon 60 is that found by de Carvalho et al (CV 5
6.9%; ICC 5 0.92). These authors actually found that the
functional H/Q ratio had higher reliability than conventional
ratio. Only one study, by Ayala et al,12 reported a lower
reliability for Hecc 260/Qcon 60 (CV 5 20.2; ICC 5 0.34)
than that we found in this study.
Study Limitations
Eccentric measurements and RTD during isometric contrac-
tions showed the highest intersession variability. The re-
liability of these measurements might be improved by a more
intense familiarization protocol. The subjects experienced
some difficulty in achieving maximal eccentric activation of
the muscles and this may have influenced the repeatability of
the measurements.
CONCLUSIONS
Isokinetic and Isometric Strength Parameters
We found that measurements of isokinetic and isometric
strength parameters had high repeatability within and
between sessions. Intersession reliability was generally slightly
lower than intrasession reliability. These results conclusively
showed (1) high intrasession reliability for measurements of
concentric, eccentric and isometric knee extension and
flexion, (2) moderate-to-high intersession reliability for
concentric and eccentric measurements and IMVT, and (3)
moderate-to-high intrasession reliability, but low intersession
reliability for RTD during isometric measurements.
Conventional and Functional H/Q Ratios
In summary, the results showed that (1) conventional ratios
had moderate-to-high intrasession reliability, (2) functional
Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Clin J Sport Med
ratios had high intrasession reliability, (3) Hcon/Qcon and Hiso/
Qiso had higher intersession reliability (moderate to high) than
Hecc/Qecc (low to moderate), and (4) conventional H/Q ratios
had higher intersession reliability (low to high) than functional
H/Q ratios (low to moderate). In other words, these data
indicate that Hcon/Qcon and Hiso/Qiso had adequate reliability.
The eccentric measurements showed the highest intersession
variability. The relatively greater variability of Hecc/Qecc and
the functional ratios might be reduced using a more intense
familiarization procedure, as discussed above.
Our findings are likely to be of great interest to the
researchers, clinicians, and trainees who want to use the
isoforce dynamometer in the future. The results confirm
the reliability of isokinetic and isometric strength measure-
ments and the most frequently used strength imbalance ratios
(Hcon/Qcon and Hiso/Qiso) for knee extension and flexion
measurements. The new dynamometer can therefore be
recommended for use in research and clinical applications.
ACKNOWLEDGMENTS
The authors thank Ulrike Balzer for supporting data
acquisition.
References
1. Hislop HJ, Perrine JJ. The isokinetic concept of exercise. Phys Ther. 1967;
47:114–117.
2. Bandy WD, McLaughlin S. Intramachine and intermachine reliability for
selected dynamic muscle performance tests. J Orthop Sports Phys Ther.
1993;18:609–613.
3. Capranica L, Battenti M, Demarie S, et al. Reliability of isokinetic knee
extension and flexion strength testing in elderly women. J Sports Med Phys
Fitness. 1998;38:169–176.
4. de Carvalho Froufe Andrade AC, Caserotti P, de Carvalho CM, et al.
Reliability of concentric, eccentric and isometric knee extension and
flexion when using the REV9000 isokinetic dynamometer. J Hum Kinet.
2013;37:47–53.
5. Feiring DC, Ellenbecker TS, Derscheid GL. Test-retest reliability of the
biodex isokinetic dynamometer. J Orthop Sports Phys Ther. 1990;11:
298–300.
6. Iga J, George K, Lees A, et al. Reliability of assessing indices of isokinetic leg
strength in pubertal soccer players. Pediatr Exerc Sci. 2006;18:436–445.
7. Impellizzeri FM, Bizzini M, Rampinini E, et al. Reliability of isokinetic
strength imbalance ratios measured using the Cybex NORM
dynamometer. Clin Physiol Funct Imaging. 2008;28:113–119.
8. Maffiuletti NA, Bizzini M, Desbrosses K, et al. Reliability of knee
extension and flexion measurements using the Con-Trex isokinetic
dynamometer. Clin Physiol Funct Imaging. 2007;27:346–353.
9. Orri JC, Darden GF. Technical report: reliability and validity of the iSAM
9000 isokinetic dynamometer. J Strength Cond Res. 2008;22:310–317.
10. Sole G, Hamren J, Milosavljevic S, et al. Test-retest reliability of isokinetic
knee extension and flexion. Arch Phys Med Rehabil. 2007;88:626–631.
11. Atkinson G, Nevill AM. Statistical methods for assessing measurement
error (reliability) in variables relevant to sports medicine. Sports Med.
1998;26:217–238.
12. Ayala F, De Ste Croix M, Sainz de Baranda P, et al. Absolute reliability of
isokinetic knee flexion and extension measurements adopting a prone
position. Clin Physiol Funct Imaging. 2012;33:45–54.
13. Wilhite MR, Cohen ER, Wilhite SC. Reliability of concentric and eccentric
measurements of quadriceps performance using the KIN-COM
dynamometer: the effect of testing order for three different speeds.
J Orthop Sports Phys Ther. 1992;15:175–182.
14. Thorstensson A, Grimby G, Karlsson J. Force-velocity relations and fiber
composition in human knee extensor muscles. J Appl Physiol. 1976;40:
12–16.
15. Coombs R, Garbutt G. Developments in the use of the hamstring/
quadriceps ratio for the assessment of muscle balance. J Sports Sci Med.
2002;1:56–62.
16. Yeung SS, Suen AM, Yeung EW. A prospective cohort study of hamstring
injuries in competitive sprinters: preseason muscle imbalance as a possible
risk factor. Br J Sports Med. 2009;43:589–594.
Volume 0
· Number 0
· Month 2017
17. Aagaard P, Simonsen EB, Magnusson SP, et al. A new concept for
isokinetic hamstring: quadriceps muscle strength ratio. Am J Sports Med.
1998;26:231–237.
18. Dvir Z, Eger G, Halperin N, et al. Thigh muscle activity and anterior cruciate
ligament insufficiency. Clin Biomech (Bristol, Avon). 1989;4:87–91.
19. Hopkins WG. Precision of measurement. In: A New View of Statistics.
2011. Available at: newstats.org/precision.html. Accessed April 3, 2015.
Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
www.cjsportmed.com
20. Stokes M. Reliability and repeatability of methods for measuring muscle
in physiotherapy. Physiother Pract. 1985;1:71–76.
21. Vincent WJ. Statistics in Kinesiology. Champaign, IL: Human Kinetics;
1999.
22. Behrens M, Mau-Moeller A, Wassermann F, et al. Effect of fatigue on
hamstring reflex responses and posterior-anterior tibial translation in men
and women. PLoS One. 2013;8:e56988.