The amount of active range of motion (AROM) for forearm

Transcription

1 Reliability of 2 Functional Goniometric Methods for Measuring Forearm Pronation And Supination Active Range of Motion Christos Karagiannopoulos, MPT, MEd, ATC 1 Michael Sitler, EdD, ATC 2 Susan Michlovitz, PT, PhD, CHT 3 Study Design: Test-retest reliability study. Objectives: To determine intra- and intertester reliability of the hand-held pencil (HHP) and the plumbline goniometer (PLG) methods for measuring active forearm pronation and supination motions in individuals with and without injuries. Background: The distal forearm method has been considered the gold standard for measuring forearm pronation and supination motion. The HHP and PLG, however, are 2 more functional methods for measuring forearm motions, though limited information on the psychometric properties of these tests is currently available. Methods and Measures: Intra- and intertester reliability of the HHP and PLG methods were determined in 40 subjects of convenience (20 injured and 20 noninjured). Two testers performed 3 repeated measurements for each motion and method on all subjects. Intraclass correlation coefficients (ICC 3,1 for intratester reliability, ICC 2,3 for intertester reliability) and standard error of measurements (SEMs) were determined. Results: The ICCs for the measurements of pronation and supination using the HHP and PLG methods were high (range, ) for individuals with and without injuries, with the reliability for the PLG method being equal or slightly greater than the HHP method for the majority of pronation and supination measurements. Intratester ICCs were higher (SEMs were conversely lower) than intertester ICCs for nearly all measurements. The ICC values were generally the same or higher for individuals with injuries compared to individuals without injuries. Conclusions: The HHP and PLG are highly reliable methods for measuring functional forearm pronation and supination. Because plumbline goniometers are not commercially available and the instrumentation for the HHP method is readily accessible, clinicians should consider the latter as their method of choice for measuring functional forearm pronation and supination. J Orthop Sports Phys Ther 2003;33: Key Words: goniometry, radioulnar joints, upper extremity The amount of active range of motion (AROM) for forearm pronation and supination can be altered by disruption of the radioulnar and radiocarpal joints biomechanics resulting from trauma and inflammatory processes. 8 Reliable methods for assessing pronation and supination AROM are 1 Senior Physical Therapist, Temple University Physical Therapy, Temple University Hospital, Philadelphia, PA. 2 Chair, Department of Kinesiology, Temple University, College of Education, Philadelphia, PA. 3 Professor, School of Physical Therapy, Temple University, College of Health Professions, Philadelphia, PA. Work for this study was done in partial fulfillment of the requirements for the degree Master of Education. The research protocol was approved by the Institutional Review Board of Temple University. Send correspondence to Christos Karagiannopoulos, Temple University Hospital, Outpatient Building, 5th Floor, 3401 North Broad Street, Philadelphia, PA karagianopou@netcarrier.com important for clinical identification of impairments, functional limitations, and monitoring efficacy of interventions during the rehabilitation process. 6 The distal forearm method is the most common measurement technique used and is considered the gold standard for measuring pronation and supination AROM. 5 This method consists of aligning a goniometer s stationary arm parallel to the humerus anterior midline while the moving arm is placed on the volar or dorsal distal forearm. 14 A limitation of the distal forearm method is that it requires measuring pronation and supination AROM in a nonfunctional manner, 5 as it does not simulate many manual activities of daily living that combine forearm rotation and hand use. Current clinical standards stress the use of functional methods for measuring and restoring physical capacity. Assessment of functional pronation (eg, pronation combined with gripping) and supination AROM should include methods that measure concurrent radioulnar joint rotation (proximal and distal) and accessory wrist motion 4,10 at the radiocarpal and midcarpal joints. 5 The hand-held pencil (HHP) and plumbline goniometer (PLG) are the only 2 reported assessment methods that present the potential Journal of Orthopaedic & Sports Physical Therapy 523

2 for measuring functional pronation and supination. The HHP method, which was first described by McRae 13 and later modified by Clarkson and Gilewich, 3 consists of measuring pronation and supination with the goniometer aligned with a pencil held in the hand of the person being measured. The PLG method, which was first described by Flowers et al, 5 consists of measuring pronation and supination with a plumb line attached to a hand-held single-arm goniometer. Limited psychometric testing has been conducted on the 3 aforementioned methods of measuring active and passive range of motion of the forearm in pronation and supination. Intra- and intertester reliability for measurements using the distal forearm method in subjects with injuries range from 0.79 to ,5 Neither intra- nor intertester reliability has been reported for the HHP method. Intratester reliability for the measurement of passive pronation and supination range of motion has been reported (ICC = 0.87 and 0.95) for the PLG method but for subjects with injuries only. 5 The purpose of this study was to determine the intra- and intertester reliability of the HHP and PLG methods for measuring active forearm pronation and supination in individuals with and without injuries. The outcome of this study is intended to improve clinical examination technique by determining the reliability of 2 functional methods for assessing forearm motion. METHODS Research Design Two testers performed 3 measurements of pronation and supination with both methods within a single testing session. The independent variables were measurement method (HHP and PLG), tester (1 and 2), motion (pronation and supination), and group (injured and noninjured). The dependent variable was AROM. Subjects Two independent groups of subjects participated in the study. The first group consisted of a sample of convenience of 20 subjects with recent injuries (10 males, 10 females) who were on average (± SD) 51 ± 14.5 years of age (range, years). This group was used to determine the intra- and intertester reliability of the HHP and PLG methods in a population with injuries. The subjects in this group were patients at Temple University Department of Orthopaedic Surgery and Sports Medicine at Northeastern Hospital and were seen within 1 to 10 weeks following forearm immobilization due to hand, wrist, forearm, or elbow trauma (Table 1). Exclusion criteria were: neurological pathologies resulting in paralysis or paresis of the arm; pain or deformity secondary to rheumatoid disease; hand, wrist, forearm, or elbow musculoskeletal injuries not treated via immobilization; and inability to touch the distal palmar crease with the tips of the second through the fifth digits TABLE 1. Diagnoses and demographics of subjects with injuries. Subject Number Sex Age (y) Diagnosis 1 F 49 R elbow epicondylectomy 2 M 36 L distal humerus fracture 3 M 39 R radial head fracture 4 M 33 L proximal ulnar fracture 5 F 56 L radial head fracture 6 M 44 R lateral epicondyle release 7 F 38 L distal radius fracture 8 M 71 L scapholunate ligament repair 9 F 80 R distal radius fracture 10 F 37 L radial head fracture 11 M 31 R scaphoid fracture 12 M 57 R distal radius fracture 13 M 38 R distal radius fracture 14 F 65 L metacarpal fracture 15 M 60 L scaphoid fracture 16 F 50 R lateral epicondyle fracture 17 M 54 L distal radius fracture 18 F 59 L metacarpal fracture 19 F 77 L distal radius fracture 20 F 46 L distal radius fracture Abbreviations: M, male; F, female; L, left; R, right. 524 J Orthop Sports Phys Ther Volume 33 Number 9 September 2003

3 when making a full fist. The full-fist criterion was required for testing for both measurement methods. The second group consisted of a sample of convenience of 20 subjects without injuries (8 males, 12 females), who were on average (± SD) 41 ± 13 years of age (range, years). This group was used to determine the intra- and intertester reliability of the HHP and PLG methods in a population without injuries. All subjects were employees or patients at the aforementioned medical facility. The exclusion criteria were limited to the neurological pathologies, rheumatoid disease complications, and upperextremity musculoskeletal injuries as identified for the subjects with injuries. The study was approved by the Temple University Institutional Review Board. Subjects read and signed an informed consent prior to participating in the study. Instrumentation Data Recording Forms A demographic data form was used to collect subject age, sex, hand dominance, and presence of inclusion and exclusion criteria information. Hand dominance was determined via the modified Edinburgh Handedness Inventory, 2 a 5-item questionnaire based on writing, throwing, drawing, scissor use, and toothbrush use. Dominance was defined as the hand with the highest number of positive responses. A separate form was used to record the pronation and supination AROM values measured for the HHP and PLG methods. Data were recorded for each subject and each tester on a separate form to minimize tester bias. Hand-Held Universal Goniometer A plastic goniometer (AliMed, Dedham, MA) with a central 360 scale in FIGURE 1. Plumbline goniometer. 5 increments, and two 14.5-cm-long arms were used for the HHP method. One side of the goniometer s central scale was covered with white adhesive paper to minimize tester bias during each measurement. The same goniometer was used for all HHP measurements for both testers. A 5 increment goniometer was selected due to its common use in the clinical setting and the fact that measurement fluctuation of up to ± 5 is usually expected and attributed to instrument error. 12 Plumbline Goniometer A 14.5-cm-long single-arm plastic goniometer with a plumbline attached to the center of its 360 central scale 5 was used for the PLG method (Figure 1). The central scale of the plumbline goniometer had identical degree increments as the central scale of the HHP goniometer. The string of the plumbline was free to move about its axis of rotation at the center of the goniometer s central scale. A 2.54-cm-diameter tubular handle was attached at the goniometer s midpoint at a 90 angle. This angle allowed the plumbline goniometer to be positioned horizontally over the radial aspect of the hand while the subject held the tubular handle. The same plumbline goniometer was used for all PLG measurements for both testers. HHP Goniometric Measurement Method Pronation and supination AROM using the HHP method were measured via a standardized testing protocol as described by Clarkson and Gilewich. 3 Measurements were conducted with each subject seated in an armless chair in a comfortable erect position with hips and knees flexed to 90, feet flat on the floor, ipsilateral arm fully adducted to the side, and elbow flexed to 90. The forearm was unsupported and placed in a neutral position (ie, forearm horizontal to the floor, palmar aspect of the hand facing medially with the thumb directed upwards) (Figure 2). For this method, each subject held a 15-cm-long pencil in a tight full fist. The pencil extended from the radial aspect of the subject s hand. The goniometer was positioned as follows: axis aligned with the head of the third metacarpal, stationary arm perpendicular to the floor, and movement arm parallel to the pencil (Figure 2). The goniometer s central scale was positioned with its covered side facing the tester. When the patient reached maximum active pronation or supination, the tester aligned the goniometer s movement arm parallel to the pencil while the axis and stationary arms remained at their initial positions (Figure 3). Then, the tester turned the goniometer and read the exposed side of its central scale. All pronation and supination AROM measurements were taken with reference to the neutral, or 0, position. Measurements were rounded to the closest 5 increment (eg, 5, 10, 15 ). 11 The goniometer was repositioned to 0 prior to each measurement. Goniometric measurements were repeated 3 times each for pronation and supination. J Orthop Sports Phys Ther Volume 33 Number 9 September

4 Procedures After completion of the demographic data form, subjects were stratified into 1 of 2 groups (noninjured or injured). Order of testing for measurement method (HHP and PLG) and motion (supination and pronation) were randomized by coin flip within each group for each subject. All subjects were measured by 2 testers, and the order of testing for the testers was also randomized. The testers were 2 male physical therapists with 2 and 4 years of professional experience and no prior clinical exposure to the HHP and PLG methods. Testers were formally trained on each method and practiced each method on 10 subjects without injuries prior to the study. For the subjects with injuries, measurements were taken on the injured arm. For subjects without injuries, measurements were taken on the dominant arm. Subjects were positioned as aforementioned and asked to turn their palms towards the floor as far as possible prior to each pronation measurement, and turn their palms towards the ceiling as far as possible prior to each supination measurement. Testers used the same instrument for each method to complete all measurements. Each tester read and recorded his own goniometric measurements in separate recording forms. Privacy of each tester s recording form was ensured during measurements to prevent tester bias. Three repeat measures were completed on each FIGURE 2. Neutral forearm position for the hand-held pencil method. PLG Goniometric Measurement Method Pronation and supination AROM using the PLG method were measured via a standardized testing protocol as described by Flowers et al. 5 Measurements were conducted with each subject seated in the same position as for the HHP method. Each subject held the plumbline goniometer by its tubular handle in a tight full fist (Figure 4). The tester rotated the goniometer handle in the subject s hand to a horizontal position, allowing the plumbline to make full contact with the goniometer s central scale. All pronation and supination AROM measurements were taken with reference to the neutral, or 0, position (Figure 4). The plumbline was placed medial to the hand for the measurement of pronation and lateral to the hand for supination. A measurement was taken when the subject reached maximum active pronation or supination with the plumbline lying flush on the surface of the goniometer s central scale without oscillating (Figure 5). The tester manually stopped any oscillatory movements of the plumbline prior to reading the value. Goniometric measurements were repeated 3 times each for pronation and supination. FIGURE 3. Measurement of active range of motion in supination with the hand-held pencil method. 526 J Orthop Sports Phys Ther Volume 33 Number 9 September 2003

5 TABLE 2. Descriptive statistics for measurement of active range of motion (in degrees) of the forearm with the hand-held pencil and plumbline goniometer techniques. Values are based on the average of trial 2 and 3 made by each tester. HHP PLG Motion Tester Group Mean ± SD Range Mean ± SD Range Pronation 1 IG 76 ± ± NIG 79 ± ± IG 78 ± ± NIG 79 ± ± Supination 1 IG 81 ± ± NIG 103 ± ± IG 80 ± ± NIG 101 ± ± Abbreviations: HHP, hand-held pencil; PLG, plumbline goniometer; IG, injured group (n = 20); NIG, noninjured group (n = 20). subject by both testers within the same test session. The testers visually inspected the subjects body positions prior to and during each trial to ensure proper body and arm alignment. The same type of chair was utilized for all measurements. Data Analysis Intraclass correlation coefficients (ICC) and standard error of measurements (SEM) were used to determine the intra- and intertester reliability of the HHP and PLG methods. ICC 3,1 was used to calculate the intratester reliability for each tester, and ICC 2,3 was used to calculate the intertester reliability. 16 ICC calculations were based on the 20 subjects for each reliability measure. From the 3 trials for each motion, the second and third trials of both testers were used to calculate the intratester reliability for each method. The mean value from the 3 trials of each tester was used in the intertester reliability analysis for each method. The Statistical Package for Social Sciences for Windows 10.1 (SPSS, Inc., Chicago, IL) was used to analyze the data. ICC values were interpreted as follows: equal to or greater than 0.75 were classified as high, 0.40 to 0.75 were classified as moderate, and below 0.40 were classified as poor. 1,9,16 RESULTS Descriptive statistics for pronation and supination AROM using the HHP and PLG methods for each tester are reported in Table 2. These values are based on the average of trial 2 and 3 made by each tester. Subjects With Injuries The intra- and intertester ICC and SEM values for pronation and supination AROM for the HHP and PLG methods for the subjects with injuries are reported in Tables 3 and 4, respectively. ICC values for the 2 methods ranged from 0.92 to SEM values ranged from 1.4 to 3.6. The PLG method had the same or higher ICC values than the HHP method except for intertester pronation, which had the lowest ICC value (highest SEM value). The PLG had the same or lower SEM values than the HHP within testers but higher between testers. Intratester ICCs were higher (SEMs were conversely lower) than intertester ICCs (SEMs were conversely higher) for both goniometric measurement methods except HHP pronation measurement for tester 2. FIGURE 4. Neutral forearm position for the plumbline method. J Orthop Sports Phys Ther Volume 33 Number 9 September

6 values ranged from 1.4 to 3.9. No consistent pattern of differences existed in the ICC and SEM values of the PLG and HHP methods for intra- and intertester measurements. The highest SEM value existed in the HHP intertester supination measurement. Intratester ICCs were higher (SEMs were conversely lower) than intertester ICCs (SEMs were conversely higher) for both goniometric measurement methods except for the HHP pronation and PLG supination measurements for tester 2. Comparison Between Subjects With and Without Injuries Intra- and intertester ICC values were the same or higher for the individuals with injuries compared to individuals without injuries, except for HHP pronation for tester 1. The same was not true for SEM values, which did not display a consistent relationship between the 2 groups. FIGURE 5. Measurement of active range of motion in pronation for the plumbline method. Subjects Without Injuries The intra- and intertester ICC and SEM values for pronation and supination AROM for the HHP and PLG methods for the subjects without injuries are reported in Tables 3 and 4, respectively. ICC values for the 2 methods ranged from 0.86 to SEM DISCUSSION The forearm and wrist constitute a multi-joint system which primary function is to position the hand in space during pronation and supination. A variety of assessment instruments to measure forearm pronation and supination have been used in the clinical setting. 4,7,13,15 The HHP 3 and PLG 5 methods allow for functional measurement of forearm pronation and supination AROM but have not been thoroughly assessed for psychometric properties. Establishing psychometric properties of these methods for individuals with and without injuries provides important information useful for clinical practice and future research. There is no previous reliability study that has presented such a comparison in measuring active range of motion for pronation and supination. For individuals with and without injuries alike, the HHP and PLG methods produced high 16 pronation and supination intra- and intertester ICC values (range, ). A comparison of the 2 methods revealed that the PLG had the same or slightly higher ICC values than the HHP method for all measurements except for intertester pronation TABLE 3. Intratester pronation and supination intraclass correlation coefficients and standard error of measurements (in degrees) of each tester for subjects with and without injuries. Pronation Supination Tester 1 Tester 2 Tester 1 Tester 2 Group Method ICC SEM ICC SEM ICC SEM ICC SEM Injured HHP PLG Noninjured HHP PLG Abbreviations: ICC, intraclass correlation coefficients; SEM, standard error of measurements; HHP, hand-held pencil; PLG, plumbline goniometer. 528 J Orthop Sports Phys Ther Volume 33 Number 9 September 2003

7 TABLE 4. Intertester pronation and supination intraclass correlation coefficients and standard error of measurements (in degrees) for subjects with and without injuries. Pronation Supination Group Method ICC SEM ICC SEM Injured HHP PLG Noninjured HHP PLG Abbreviations: HHP, hand-held pencil; PLG, plumbline goniometer; ICC, intraclass correlation coefficients; SEM, standard error of measurements. in both groups, and for intratester pronation (tester 1) and intratester supination (tester 2) in subjects without injuries. The slightly higher PLG ICCs could be attributed to instrument and methodology differences between the 2 methods. The HHP required tester identification of bony landmarks and goniometer alignment, whereas the PLG method did not. The PLG method consisted of rotation of the hand-held instrument by the subject and tester assessment of joint motion via the constant gravitational force line of the plumbline. Because the gravitational force line is always perpendicular to the ground, the PLG method negated the tester s dual responsibility of aligning and reading the goniometer during a measurement. It would be expected that the aforementioned instrumentation differences would lead to higher ICC values for all conditions. Therefore, the reason for the same or higher intra- and intertester ICC values with the HHP method for forearm pronation and supination in both groups is not clear. In general, the higher the ICC value the lower the SEM value. Such an inverse relationship between ICC and SEM values is expected, as measurements of greater reliability are associated with lower experimental error. This inverse relationship between ICC and SEM values was consistently seen among all subjects except for intra- and intertester supination measurements for the subjects with injuries. Such findings may be attributed to the greater range of motion variability among the subjects with injuries, as indicated by the larger standard deviation values. Intratester ICCs were higher (SEMs were conversely lower) than intertester ICCs (SEMs were conversely higher) for both measurement methods among all subjects, except for the intratester HHP pronation and PLG supination measurements (tester 2) for subjects without injuries. No study to date has reported on the differences between intra- and intertester reliabilities for assessing functional pronation and supination AROM. Armstrong et al 1 reported higher intra- than intertester reliability for active pronation and supination AROM measurements using the distal forearm method. Horger 9 also reported higher intra- than intertester reliability for measurements of range of motion for active wrist flexion, extension, and radial and ulnar deviations when using a universal goniometer. These findings were anticipated because a lower error of measurement is typically associated with repeated measurements conducted by a single rather than multiple testers. 6 The ICC values obtained on the measurements made on the subjects with injuries were the same or higher than the ICC values obtained on the subjects without injuries, except for the measurement of HHP pronation for tester 1. This result was not anticipated as greater measurement variability and error was expected in subjects with injuries. Rothstein et al 17 reported high ICC values (range, ) for elbow flexion and extension AROM in subjects with injuries, but did not include subjects without injuries in their study. No study to date has determined reliability differences between subjects with and without injuries for any upper- or lower-extremity goniometric assessment. Because ROM measurement reliability depends on factors such as patient population and type of injury, 6 comparisons between noninjured and injured populations should be joint or injury specific prior to making generalizations for clinical practice. In contrast to the ICC findings, the SEM values followed an inconsistent relationship between subjects with and without injuries. SEM values for subjects with injuries were higher than for the subjects without injuries for all PLG measurements, except for the measurement of pronation for tester 1 and supination for tester 2. SEM values for subjects without injuries were the same or higher than for the subjects with injuries for all HHP measurements, except for measurements of pronation for tester 1. This result was not anticipated because measurement error would be more difficult to control for subjects with injuries than for subjects without injuries. Accordingly, greater measurement error was expected among the subjects with injuries regardless of measurement method. In our study, potential sources of experimental error included lack of standardization for pain and assessment time postimmobilization, lack of control of subjects clothing, using a goniometer with 5 scale increment, which was not calibrated prior to measurements. Typically, AROM measurements with a 1 scale increment goniometer eliminate the need for estima- J Orthop Sports Phys Ther Volume 33 Number 9 September

8 tion of readings over a 5 interval and are expected to have lower SEM values than measurements with a 5 scale increment goniometer. Interestingly, our study s forearm AROM measurements obtained with a 5 scale increment goniometer resulted in lower SEM values (range, ) than the SEM values reported by Horgen 9 (range, ) for wrist AROM measurements obtained via a 1 scale increment goniometer in subjects with injuries. Although we believe that masking the HHP method goniometer and recording AROM values in separate forms for each tester during measurements provided adequate blinding in this study, a doubleblind study design would have offered lower threat to intratester reliability. Not covering the dial of the plumbline goniometer was not considered a threat to intratester reliability of the PLG method in this study because AROM was controlled by the subject and measurement simply consisted of only reading without aligning the goniometer by the tester. In addition, subjects were not allowed to look at the goniometer and verbal instructions to each subject were standardized to avoid coaching the subject to the previous angle during PLG measurements. The quality of this study s intertester reliability values was protected by having testers blinded to each other s results by using separate recording forms and ensuring the privacy of these forms during measurements. A final limitation of this study was the lack of significant AROM restriction found among the individuals with injuries. Such a limitation may have decreased the anticipated higher measurement variability and experimental error among subjects with injuries. In the current study, pronation values (mean ± SD) obtained with the HHP and PLG methods for subjects without injuries were 79 ± 8 and 78 ± 9, respectively. Glanville and Kreezer 7 and Darcus and Salter 4 reported functional pronation values of 91 ± 26 and 63 ± 10, respectively, for subjects without injuries. The values for forearm supination obtained with the HHP and PLG methods in the current study were 103 ± 10 and 104 ± 12, respectively. These values were similar to those reported by both Glanville and Kreezer 7 (99 ± 11 ) and Darcus and Salter 4 (102 ± 11 ) for subjects without injuries. Variability in functional pronation AROM among these studies is attributed to differences in instrumentation (eg, handle size, grip force), age, and subject sampling (eg, random, subjects of convenience, sex). Current clinical standards require the use of functional methods for measuring and restoring patients physical capacities. In an attempt to improve the clinical examination technique of functional forearm motion, this study s outcome supports the clinical use of both the HHP and PLG methods. Although both methods seem to be reliable, at present, the HHP method is a more accessible and cost-effective instrumentation for today s clinical settings. A future multicenter reliability study that compares today s gold standard distal forearm method to either the HHP or PLG method on a larger sample of subjects with injuries using multiple measurement sessions may offer a greater external validity of results and finally answer, once and for all, the question of which method (nonfunctional versus functional) is the most reliable for measuring forearm pronation and supination. CONCLUSIONS Results of this study demonstrated that the HHP and PLG methods have both high intra- and intertester reliability for measuring forearm pronation and supination AROM in subjects with and without injuries. The slightly higher reliability of the PLG method is offset by the simplicity and availability of the instrumentation needed for the HHP method. Because plumbline goniometers are not commercially available and the instrumentation for the HHP is readily accessible in the clinical setting, the latter should be considered the method of choice by clinicians for measuring functional forearm pronation and supination. ACKNOWLEDGMENTS Thanks to Matthew Spiegler, MPT, who so diligently offered his substantial participation in the data collection process of this study. REFERENCES 1. Armstrong AD, MacDermid JC, Chinchalkar S, Stevens RS, King GJ. Reliability of range-of-motion measurement in the elbow and forearm. J Shoulder Elbow Surg. 1998;7: Bryden MP. Measuring handedness with questionnaires. Neuropsychologia. 1977;15: Clarkson HM, Gilewich GB. Musculoskeletal Assessment: Joint Range of Motion and Manual Muscle Strength. Baltimore, MD: Williams and Wilkins; Darcus HD, Salter N. The amplitude of pronation and supination with the elbow flexed to a right angle. J Anat. 1953;87: Flowers KR, Stephens-Chisar J, LaStayo P, Galante BL. Intrarater reliability of a new method and instrumentation for measuring passive supination and pronation: a preliminary study. J Hand Ther. 2001;14: Gajdosik RL, Bohannon RW. Clinical measurement of range of motion. Review of goniometry emphasizing reliability and validity. Phys Ther. 1987;67: Glanville AD, Kreezer G. The maximum amplitude and velocity of joint movements in normal male human adults. Hum Biol. 1937;9: Graham TJ, Fischer TJ, Hotchkiss RN, Kleinman WB. Disorders of the forearm axis. Hand Clin. 1998;14: Horger MM. The reliability of goniometric measurements of active and passive wrist motions. Am J Occup Ther. 1990;44: J Orthop Sports Phys Ther Volume 33 Number 9 September 2003

9 10. Linscheid RL. Kinematic considerations of the wrist. Clin Orthop. 1986; Low JL. The reliability of joint measurement. Physiotherapy. 1976;62: Mayerson NH, Milano RA. Goniometric measurement reliability in physical medicine. Arch Phys Med Rehabil. 1984;65: McRae R. Clinical Orthopaedic Examination. Edinburgh, UK: Churchill and Livingstone; Norkin CC, White DJ. Measurement of Joint Motion: A Guide to Goniometry. Philadelphia, PA: FA Davis; Patrick J. A study of supination and pronation with especial reference of the treatment of forearm fractures. J Bone Joint Surg Am. 1946;28: Portney LG, Watkins MP. Foundations of Clinical Research: Applications to Practice. Norwalk, CT: Appleton and Lange; Rothstein JM, Miller PJ, Roettger RF. Goniometric reliability in a clinical setting. Elbow and knee measurements. Phys Ther. 1983;63: J Orthop Sports Phys Ther Volume 33 Number 9 September