Volume 88, Issue 9 , Pages 1164-1170, September 2007
Reliability of a New Instrument for Measuring Plantarflexor Muscle Strength
Article Outline
Abstract
Örtqvist M, Gutierrez-Farewik EM, Farewik M, Jansson A, Bartonek Å, Broström E. Reliability of a new instrument for measuring plantarflexor muscle strength.
Objectives
To test the reliability of a new muscle strength testing instrument (the Strength Measuring Chair [SMC]) designed to quantify isometric strength in the lower extremities, and to determine the agreement between the SMC and an isokinetic dynamometer (Biodex).
Design
Isometric strength tests were performed in plantarflexors with 2 different knee positions (60°, 30°). Measurements were taken at 3 different sessions.
Setting
Strength testing laboratory.
Participants
Twenty-three able-bodied adults and 15 able-bodied children.
Interventions
Not applicable.
Main Outcome Measure
Isometric plantarflexor strength.
Results
The reliability of isometric strength measurements of plantarflexors taken in the SMC was excellent for both the adult and children groups (intraclass correlation coefficient range, .84−.87). A Bland-Altman 95% limit of agreement test showed no systematic variation in 3 of the 4 SMC test observations; systematic variation was only observed in the adult group at a knee position of 30°. There was no systematic difference in the adult group between the SMC and the isokinetic dynamometer, but there was a systematic variation in the children’s group.
Conclusions
The SMC reliably measured isometric plantarflexor strength in the tested populations.
Key Words: Gastrocnemius muscle, Isometric contraction, Rehabilitation, Reproducibility of results, Soleus muscle
MEASURING MUSCLE STRENGTH is a routine procedure in the assessment of children and adults with pathology that affects muscle strength or motion.1 Traditionally in clinical practice, manual muscle testing (MMT) is the most commonly used method of evaluating muscle strength.2 The disadvantages of this method, namely, the subjective nature of assessing the amount of resistance applied during the test and the noninterval data scale—which complicates and rules out many statistical analyses—have led to more frequent use of quantitative instruments such as hand-held dynamometers and computerized isokinetic dynamometers.3 Hand-held dynamometers are used to assess isometric strength at isolated joint positions. Studies with hand-held dynamometers have found that they have good reliability in various populations.4, 5, 6, 7
Even so, this method has some important limitations, such as difficulties in stabilizing the patient, accurately assessing the joint position, and using the examiner’s strength as the estimated resistance, particularly when testing large muscle groups.5, 8, 9, 10 A more sensitive and valid testing technique may be achieved by using a computerized strength-testing isokinetic dynamometer. Such dynamometers reliably measure strength in the calf and knee muscles in different populations.1, 11, 12, 13, 14, 15, 16 Several studies have shown that one commonly used isokinetic dynamometer, the Biodex System 3, is a reliable device that has a negligible learning effect.17, 18, 19 Isokinetic dynamometers are more commonly used in larger physiotherapy clinics than in smaller clinics or gait laboratories.
The reliability of any new instrument must be tested before it can be used clinically. Reliability refers to the consistency of repeatability of a measurement. It is difficult to find a clinical measurement that is totally reliable because all instruments and raters have some inconsistencies. Any observed value may, therefore, be considered a function of one true value and one error component. The difference between the true value and the observed value is called “measurement error.”6 “Agreement” refers to how well results from 2 different methods or instruments agree.20
Testing muscle strength in the lower extremities is not an easy task because strength is dependent on posture and joint position.3 This is particularly true for the plantarflexors, which are normally capable of lifting several times body weight and which are important for ankle and knee stability21 and for propelling the body forward in gait.22 To our knowledge, relatively few studies7, 14, 15, 23, 24, 25 have measured isometric plantarflexor strength in healthy adults and children. Because of their strength, plantarflexors of adolescents and adults are impossible to measure reliably with MMT3 and hand-held dynamometers.4, 7 The demands on an instrument for testing this muscle group while controlling the joint position are many.26 In this project, a new muscle strength testing instrument, the Strength Measuring Chair (SMC), was designed to restrict joint motion during testing.
Our purpose in this study was to test the SMC’s intrasubject, intrasession, and intersession reliability and to determine the agreement between this new instrument and a commonly used isokinetic dynamometer, the Biodex System 3.
Methods
Participants
An adult group consisted of 23 able-bodied subjects (13 women, 10 men) between 23 and 60 years of age. A second group consisted of 15 able-bodied children (8 girls, 7 boys) ranging in age from 5 to 10 years (table 1). None of the subjects had current or recent injuries in the lower extremities at time of testing. Subjects were recruited from among colleagues, friends, and acquaintances. Participation was voluntary and written and verbal information was given to all subjects. The Karolinska University Hospital ethics committee approved the study.
Table 1. Characteristics of the Adult and Children Groups
| Group | N | Weight (kg) | Height (cm) | Age (y) |
|---|---|---|---|---|
| Adult group | ||||
 Women | 13 | 62.3±6.5 | 168.8±4.5 | 30(24−60) |
| Men | 10 | 79.0±7.3 | 180.2±3.9 | 34(23−53) |
| Total | 23 | 69.6±10.8 | 173.8±7.1 | 30(23−60) |
| Children group | ||||
| Girls | 8 | 26.0±2.9 | 131.3±6.0 | 8(6−8) |
| Boys | 7 | 28.8±5.2 | 130.8±8.8 | 8(5−10) |
| Total | 15 | 27.3±4.3 | 131.0±7.2 | 8(5−10) |
Instrumentation
The SMCThis new instrument is designed for seated subjects undergoing testing with hip and ankle joints in the 90° positions and the knee joint adjustable from 90° to 30° positions (0° is extended leg). The chair is adjustable in size so as to accommodate subjects from approximately 5 years of age to adulthood. Four force sensors, which measure force generation in both tension and compression, are placed at adjustable distances from joint centers (fig 1). The same sensors can be used to test opposing muscle groups; for example, by pushing on a sensor, plantarflexor strength can be tested, and dorsiflexor strength can be tested by pulling up on a strap attached to the same sensor. The instrument is therefore able to measure force generation in 4 different muscle groups: plantarflexors, ankle dorsiflexors, knee flexors, and knee extensors on both sides simultaneously. This study was focused only on plantarflexors.
Fig 1.
The SMC.
To restrict motion during plantarflexor strength testing, 5 straps were attached to the subjects: one around the waist, one over the thigh, and one around the lower shank. The foot was fixed by 2 straps to restrict movement at the ankle, to prevent harmful loading to the equipment, and to ensure against obtaining inaccurate data because of a nonaxial movement. Plantarflexor strength was tested when force was applied to the sensor by pushing the sensor plate under the foot.
The moment arm was measured from the lateral malleolus to the sensor’s middle along the sensor plate axis. The voltage signal from the force sensors is converted to a digital signal and sent to a personal computer (PC). The instrument measures plantarflexor strength as the product of the compression force (in newtons) of the sensor and the moment arm (in meters). A user-friendly PC interface was designed in LabView.a The user interface implements moment arms and displays the generated torque in newton-meters. The data are then collected and exported to Microsoft Excel.b Taring for each sensor was implemented to offset the limb weight on the sensor. Data were sampled at 1000Hz and analyzed as a mean over a 10-second interval. The mean of the collected data were then considered the tare value of the limb weight. Sensors were calibrated shortly before this study began.
Isokinetic dynamometerThe Biodex System 3c can be used for both muscle strength testing and rehabilitation. It accommodates several positions and exercises. Five different modes are available for muscular testing or rehabilitation: isokinetic resistance, eccentric resistance, passive motion, isotonic, and isometric mode. The manufacturer has provided recommendations for testing isometric plantarflexor strength.27 The leg is attached via straps and strength is measured by a central torque-measuring device.
Testing Procedure and Data Collection
Subjects attended 3 trial sessions that lasted about 1 hour each. Isometric torque was measured with both devices. Subjects were tested first in the SMC and then with the isokinetic dynamometer by the same examiner to evaluate agreement, and then again in the SMC. In the SMC the plantarflexors were tested with both 60° and 30° angles in the knee. Because of the instrument setup conditions for the isokinetic dynamometer, the plantarflexors were only tested with a 60° angle in the knee. During all tests, subjects were seated with their backs supported, hips at right angles, and arms across their chests. The right leg only was tested, with 3 repetitions per testing position and a minimum resting period of 2 minutes between each trial to avoid muscle fatigue. Before each trial, subjects were told to push as hard as they could for 5 seconds. The lateral malleolus of the fibula was used as the reference landmark and aligned with the axis of rotation of the machine. There was a 1-week interval between trials with the different devices to avoid muscle fatigue and prevent a learning effect (fig 2). Subjects were given similar verbal encouragement during all tests. Visual feedback on the computer monitor during the tests was not provided. The same examiner (MÖ) conducted all tests.
Fig 2.
Test design. Abbreviation: ID, isokinetic dynamometer.
Statistical Analysis
Descriptive statistics of muscle strength are presented as torque (in newton-meters). The differences in peak torque measured in the 2 testing positions (60°, 30°) were calculated with repeated-measures analysis of variance (ANOVA).
The degree of correlation between the 3 repetitions during session 2 in the isokinetic dynamometer and the 3 repetitions during session 3 in the SMC (intrasubject reliability) was calculated with an intraclass correlation coefficient, model 1,1 (ICC1,1).28 We used a Bland-Altman 95% limit of agreement29 and an ICC2,1 to test the intersession reliability and evaluate systematic variations between session 1 and 3. The Bland-Altman test is an approach to assessing agreement between 2 different methods of clinical measurements. This test involves calculating the mean result from each method and using it in a series of agreement tests. The results are presented in a Bland-Altman plot. The measurement error was calculated with a standard error of measurement and a coefficient of variation (CV).20 ICC was interpreted according to recommendations by Fleiss.30 An ICC greater than .75 represents excellent reliability; 0.4 to .75 represents fair to good reliability.30 In the adult group, reliability was additionally calculated separately for men and women but sex was not considered individually in children.31
One child could not be tested with the knee at 30° during the first session because of technical problems. That subject’s data were not included in the calculation of statistics for plantarflexor strength with the knee at 30°.
We performed a paired t test, a Bland-Altman 95% limit of agreement test,29 and an ICC2,128 to test the agreement and to evaluate systematic variation between the 2 instruments. To study the agreement between the 2 instruments, we used data from session 3 because of the possible learning effect from the first session. We used the commercially available statistical software packaged for all statistical analyses.
Results
Muscle Strength
Table 2 shows that with the knee in the 30° position while seated in the SMC, the mean plantarflexor peak torque was 137.1Nm in the adult group and 37.3Nm in the children’s group. Both groups showed higher plantarflexor strength with the knee in the 30° position than in the 60° position (adults in session 3, 117.0Nm, P<.001; children in session 3, 30.3Nm, P<.05) (fig 3).
Table 2. Mean Values of 3 Peak Torque Measurements (in Nm) in Plantarflexion at the 3 Different Sessions, in the SMC, and in the Isokinetic Dynamometer, With 60° and 30° Knee Angles
| Group | n | Session 1 SMC | Session 2 Biodex | Session 3 SMC |
|---|---|---|---|---|
| Adult group | ||||
| Knee 60° all | 23 | 120.0±40.0 | 112.5±39.0 | 117.0±40.4 |
| Women | 13 | 92.3±28.5 | 87.5±23.3 | 98.3±33.6 |
| Men | 10 | 135.7±31.9 | 144.9±30.4 | 141.4±36.3 |
| Knee 30° all | 23 | 131.6±43.4 | ND | 137.1±42.2 |
| Women | 13 | 99.6±32.6 | ND | 111.0±35.6 |
| Men | 10 | 154.7±30.7 | ND | 156.7±39.0 |
| Children group | ||||
| Knee 60° | 15 | 30.7±12.5 | 39.3±17.3 | 30.3±13.6 |
| Knee 30° | 14 | 35.6±12.4 | ND | 37.3±15.3 |
Fig 3.
Differences in mean plantarflexor peak torque between the 2 different testing positions with knee angles of 60° and 30° in the SMC during session 3.
Instrument Reliability
The ICC values indicate excellent intersession reliability (.84−.87) of the plantarflexor strength measurements with the 30° knee position in both groups, as well as with a 60° knee position in the adult group (table 3). The ICC was lower for plantarflexor strength in the 60° knee position (.46) in the children’s group. The intersession reliability of measurements in the SMC was also calculated separately for women and men (Table 3, Table 4). SMC intersession reliability was somewhat lower when sex was considered separately. Table 5 presents the ICC between 3 repetitions during session 3 in the SMC. There was an excellent intrasubject, intrasession reliability coefficient (.85−.97) in the 2 testing positions in both the children and adult groups.
Table 3. ICC2,1 Between Session 1 and Session 3 in the SMC
| Group | n | Session 1−3 SMC |
|---|---|---|
| Adult group | ||
| Knee 60° all | 23 | .87 |
| Women | 13 | .87 |
| Men | 10 | .75 |
| Knee 30° all | 23 | .84 |
| Women | 13 | .80 |
| Men | 10 | .70 |
| All children | ||
| Knee 60° | 15 | .46 |
| Knee 30° | 14 | .86 |
Table 4. Standard Error of Measurement and CV Between Measured Plantarflexor Strength During Session 1 and Session 3 in the SMC
| Group | n | Session 1−3 SMC | |
|---|---|---|---|
| SE (Nm) | CV (%) | ||
| Adult group | |||
| Knee 60° all | 23 | 14.7 | 12 |
| Women | 13 | 11.4 | 12 |
| Men | 10 | 18.0 | 13 |
| Knee 30° all | 23 | 17.3 | 13 |
| Women | 13 | 15.6 | 14 |
| Men | 10 | 19.4 | 12 |
| All children | |||
| Knee 60° | 15 | 9.7 | 32 |
| Knee 30° | 14 | 5.2 | 14 |
Table 5. ICC1,1 Between 3 Repetitions in Plantarflexor Strength During Session 3 in the SMC
| Group | n | Session 3 SMC |
|---|---|---|
| Adult group | ||
| Knee 60° all | 23 | .96 |
| Women | 13 | .97 |
| Men | 10 | .91 |
| Knee 30° all | 23 | .97 |
| Women | 13 | .96 |
| Men | 10 | .96 |
| All children | ||
| Knee 60° | 15 | .87 |
| Knee 30° | 14 | .85 |
There were no systematic variations in the SMC test sessions 1 and 3 in 3 of the 4 observations (children in a 30° knee position, children in a 60° knee position, adults in a 60° knee position), but there was a systematic variation in the adult group with a 30° knee position; in the third session in the SMC, slightly higher strength was measured (figs 4A−4D).
Fig 4.
Bland-Altman plots29 showing the differences between session 1 in the SMC and session 3 (SMC) for plantarflexor peak torque plotted against their mean for each subject, to check for any systematic variations between the 2 test sessions. (A) Adult group (n=23); (B) children group (n=15); (C) adult group (n=23); and (D) children group (n=14).
Instrument Agreement
The ICC values between the SMC and the isokinetic dynamometer indicate excellent agreement (ICC=.82) for plantarflexor strength measurement with a 60° knee position in the adult group, but only fair to good agreement (ICC=.55) in the children’s group. The ICC values calculated separately for women and men indicated good to excellent agreement (.65 and .76, respectively).
There was no systematic variation between the SMC and the isokinetic dynamometer in the adult group (P=.37), but in the children’s group, strength was measured approximately 10Nm lower in the SMC (P<.05) (figs 5A, 5B).
Fig 5.
Bland-Altman plots29 showing the differences between session 2 (isokinetic dynamometer) and session 3 in the SMC for plantarflexor peak torque plotted against their mean for each subject, to check for any systematic variations between the 2 different strength measuring devices. (A) Adult group (n=23) and (B) children group (n=15).
Discussion
We have shown that the SMC reliably measured isometric strength in the plantarflexors. The examiner (MÖ) perceived it to be easier to operate than the isokinetic dynamometer; also, its size is such as to make it feasible for use in, for example, a clinical gait laboratory. When testing the children with the isokinetic dynamometer, the examiner had difficulty finding a stable position with no joint motion and found the SMC easier to adjust for smaller children. Because subject positioning and technical settings were easier with the SMC, its testing time took approximately 15 fewer minutes than with the dynamometer.
Strength Measurements
Subjects who had difficulty in isolating the muscle groups performed a repetition trial. The test order may influence reproducibility and therefore, the test order and the examiner were the same for each subject each time. Verbal encouragement during the test may influence the ability to produce maximum strength.32 The encouraging words spoken during the tests were standardized and strictly followed. Seger and Thorstensson31 have shown that there is a significant difference in muscle strength between sexes in adults, but no such difference has been found in prepubertal children. As such, the reliability was calculated only for each sex individually in the adult group.
Reliability Statistics
Statistical methods for assessing reliability and agreement between repeated quantitative measurements and between different apparatuses have been discussed often in the literature.28, 29 Several statistical tests are appropriate for analysis of reliability, but no individual test alone provides sufficient information. Earlier studies have recommended that several tests be used complementarily.33
The ICC analysis was specifically designed to evaluate reliability and provide a reliability index with which to indicate the measurement of error. The single index ICC is calculated using variance estimates obtained through the partitioning of total variance into between- and within-subject components. It is dependent on the heterogeneity of the group and will be high if the variance of measurements between subjects is higher than that within subjects. There is no standard acceptable level of reliability using the ICC. It will range from 0 to 1, with values closer to 1 representing a higher reliability.6 Fleiss30 recommends that an ICC greater than .75 be interpreted to represent excellent reliability and ICCs from 0.4 to .75 to be considered to be fair to good reliability, but whether a value is truly excellent can depend on the specific application. According to these recommendations, our results indicate excellent reliability in plantarflexor strength measurement with a 30° knee position in both groups and a 60° knee position in the adult group in the SMC, and a fair to good reliability with a 60° knee position in the children’s group. Because the ICC gives no indication of the size of disagreement between measurements, either a standard error of measurement or a Bland-Altman 95% limit of agreement test, or both,33 should be used to complement it.
A Bland-Altman 95% limit of agreement test assesses agreement between 2 different methods of clinical measurements and involves calculating the mean for each method and using it in a series of agreement tests. If there is a systematic variation, for example, when one of the measurements is larger than the other, the values are not distributed equally around the zero line, and a Bland-Altman plot is useful. From the graph, systematic variation and trends can be detected.29 The interpretation of the results depends on the clinical circumstances, and it is not possible to use statistics to define acceptable agreement.20
Reliability and Agreement Results
The excellent reliability for repeated measurements of plantarflexor strength in the adult group with 60° and 30° knee positions and in the children’s group with a 30° knee position in the SMC is promising. There was no apparent systematic variation during the 2 test sessions in the SMC in the children’s group, although some systematic variation was observed in the adult group. A learning effect wherein the adults became more accustomed to the protocol by the second and third session may explain this.
The results also indicate that there is a systematic variation between the 2 measuring devices in children but not in adults. This variation may be attributable to several factors, for example, greater sources of testing errors in the isokinetic dynamometer and differences in testing environments. The examiner experienced difficulty in stabilizing subjects in the isokinetic dynamometer; its modules for stabilizing the foot and the sitting surface are often too large for the children.
To apply our results in a clinical setting, the targeted population must be considered. For example, to conclude that strength has substantially improved, the magnitude of the patients’ original strength and functional improvement must be known. A reasonable goal for strength improvement must be set. Small differences in strength should be interpreted with caution because of the wide range in the within-subject torque.
Studies of lower-extremity strength in adults have reported a wide range of peak isometric strength values. Studies of plantarflexor isometric strength have reported values ranging from less than 100 to approximately 200Nm.15, 23, 24, 34 The differences in testing positions in all the studies must be considered when different results are compared. In this study, we found peak values in plantarflexor isometric strength ranging from 38 to 230Nm (adult group) and 11 to 71Nm (children’s group) in the SMC. Broström et al25 reported values of 40 to 114Nm in able-bodied children measured with an isokinetic dynamometer. The results in the children’s group were lower than those reported earlier.25 The flexed knee testing position might explain this, because Broström used a fully extended knee position in that study.25
The lack of systematic differences observed in the adult group between the isokinetic dynamometer and the SMC is interesting because plantarflexors are the most difficult muscle group to measure because of their strength and the difficulties in standardizing the test procedure and stabilizing the joint.
Reliable and valid measurements of strength in a clinical setting are of great importance. The manual muscle test has the advantage of being easy to use without any special equipment. It does, however, have some disadvantages, particularly in that an examiner’s strength must subjectively assess the amount of a subject’s resistance as being a 4 or a 5.2 Also, the MMT only provides an ordinal grade of muscle strength, which cannot be used in correlations or analyses requiring interval data.35 The commonly used hand-held dynamometer has been proven to be reliable in test-retest situations,4, 5 but its limitations, such as difficulties in stabilizing the patient, accurately assessing the joint position, and using the examiner’s strength as the estimated resistance, cannot be overlooked. Nyström Eek et al7 reported an inability to measure plantarflexor strength with a hand-held dynamometer in able-bodied children older than approximately 9 years. Therefore, the reliability problems associated with both the hand-held dynamometer and the manual muscle test must be considered. Our results indicate that the SMC may be useful in reliably measuring voluntary plantarflexor strength in healthy subjects in a clinical setting.
Further studies are needed to determine the reliability of the SMC for use in populations with different pathologies affecting muscle strength and motion. In addition, by testing strength in each side individually, as well as in both sides together, and by collecting simultaneous electromyographic data, differences in strength resulting from bilateral contractions and co-activation can be evaluated.
Conclusions
The new SMC instrument reliably measured isometric plantarflexor strength in able-bodied children and adults. This is clinically important because there are many demands placed on an instrument for testing plantarflexor strength while controlling the position of the joint. Small statistical differences in strength measured in the SMC, however, should be interpreted with caution and discussed with regard to clinical relevance. We believe that the SMC is a useful device with which to measure plantarflexor muscle strength objectively in children and adults in clinical practice, for example, in a clinical gait laboratory.
Suppliers
Acknowledgments
We thank Mikael Persson and Marie Eriksson, CPO, at Olmed Ortopediska AB for their help in designing and constructing the Strength Measuring Chair.
References
- . Deficits in eccentric versus concentric torque in children with spastic cerebral palsy. Med Sci Sports Exerc. 2001;33:117–122
- . Daniel’s and Worthingham’s muscle testing: techniques of manual examination. 6th ed. Philadelphia: WB Saunders; 1995;
- . Muscle strength development and assessment in children and adolescents. In: Harms-Ringdahl K editors. Muscle strength. Edinburgh: Churchill Livingstone; 1993;p. 93–141
- . Hand-held dynamometry for measuring muscle strength. J Hum Muscle Perform. 1991;1:35–50
- . Interrater reliability of hand-held dynamometry. Phys Ther. 1987;67:931–933
- . Reliability: what is it, and how is it measured?. Phys Ther. 2000;86:94–99
- . Isometric muscle torque in children 5 to 15 years of age: normative data. Arch Phys Med Rehabil. 2006;87:1091–1099
- . A comparison of make and break tests using a hand-held dynamometer and the Kin-Com. J Orthop Sports Phys Ther. 1994;19:28–32
- . Interrater reliability of hand-held dynamometry—effects of rater gender, body-weight, and grip strength. J Orthop Sports Phys Ther. 1992;16:74–81
- . Hand-held dynamometer measurements: tester strength makes a difference. J Orthop Sports Phys Ther. 1991;13:191–198
- . Reliability of isokinetic strength measurements of the knee in children with cerebral palsy. Dev Med Child Neurol. 2000;42:398–402
- . Isokinetic dynamometry (Applications and limitations). Sports Med. 1989;8:101–116
- . Central and peripheral contributions to fatigue in relation to level of activation during repeated maximal voluntary isometric plantar flexions. J Appl Physiol. 2004;96:218–225
- . Tension regulation during lengthening and shortening actions of the human soleus muscle. Eur J Appl Physiol. 2000;81:375–383
- . Measurement and reproducibility of strength and voluntary activation of lower-limb muscles. Muscle Nerve. 2004;29:834–842
- . Isometric strength measurements of knee extensors in women with osteoarthritis of the knee. J Rheumatol. 1996;23:328–331
- . Reliability and validity of the Biodex system 3 pro isokinetic dynamometer velocity, torque and position measurements. Eur J Appl Physiol. 2004;91:22–29
- . Intramachine and intermachine reliability of the Biodex and Cybex II for knee flexion and extension peak, torque and angular work. J Orthop Sports Phys Ther. 1991;13:329–335
- Learning effect of isokinetic measurements in healthy subjects, and reliability and comparability of Biodex and Lido dynamometers. Clin Physiol Funct Imaging. 2005;25:75–82
- . Practical statistics for medical research. London: Chapman & Hall; 1991;
- . Energy generation and absorption at the ankle and knee during fast, natural and slow cadence. 1983;175:147–154
- . The role of the ankle plantarflexors. J Bone Joint Surg Am. 1980;62:354–363
- . Calf muscle strength in humans. Int J Sports Med. 2001;22:186–191
- . Muscle architecture and force-velocity relationships in humans. J Appl Physiol. 1984;57:435–443
- . Plantar- and dorsiflexor strength in prepubertal girls with juvenile idiopathic arthritis. Arch Phys Med Rehabil. 2004;85:1224–1230
- . Lower extremity isometric joint torque in children with juvenile chronic arthritis. Scand J Rheumatol. 2001;30:69–76
- Biodex System 3, Pro Manual (#835/000) applications/operations. Shirley: Biodex Medical Systems; 1998;
- . Intraclass correlations—uses in assessing rater reliability. Psychol Bull. 1979;86:420–428
- . Statistical methods for assessing agreement between 2 methods of clinical measurement. Lancet. 1986;1:307–310
- . The design and analysis of clinical experiments. New York: John Wiley & Sons; 1986;
- . Muscle strength and myoelectric activity in prepubertal and adult males and females. Eur J Appl Physiol Occup Physiol. 1994;69:81–87
- . Verbal encouragement: effects on maximum effort voluntary muscle action. Br J Sports Med. 1996;30:243–245
- . Reliability of assessment tools in rehabilitation: an illustration of appropriate statistical analyses. Clin Rehabil. 1998;12:187–199
- . Plantar flexor strength testing using Cybex isokinetic dynamometer. Phys Ther. 1978;58:847–850
- . An introduction to medical statistics. In: 3rd ed. New York: Oxford Univ Pr; 2000;p. 210–229
Supported by the Sunnerdahl, Norrbacka Foundations, Frimurare Foundation, and Teknikstiftelsen.
No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated.
PII: S0003-9993(07)00424-8
doi:10.1016/j.apmr.2007.05.028
© 2007 American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation. Published by Elsevier Inc. All rights reserved.
Volume 88, Issue 9 , Pages 1164-1170, September 2007
