SELECTING AN APPROPRIATE manual wheelchair and

654 ORIGINAL ARTICLE Effect of Backrest Height on Wheelchair Propulsion Biomechanics for Level and Uphill Conditions Yu-Sheng Yang, PhD, Alicia M. Koontz, PhD, Shan-Ju Yeh, BS, Jyh-Jong Chang, PhD ABSTRACT. Yang Y-S, Koontz AM, Yeh S-J, Chang J-J. Effect of backrest height on wheelchair propulsion biomechanics for level and uphill conditions. Arch Phys Med Rehabil 2012;93: 654-9. Objective: To evaluate the effect of backrest height on wheelchair propulsion kinematics and kinetics. Design: An intervention study with repeated measures. Setting: University laboratory. Participants: Convenience sample included manual wheelchair users (N 36; 26 men and 10 women) with spinal cord injuries ranging from T8 to L2. Intervention: Participants propelled on a motor-driven treadmill for 2 conditions (level and slope of 3 ) at a constant speed of 0.9m/s while using in turn a sling backrest fixed at 40.6cm (16in) high (high backrest) and a lower height set at 50% trunk length (low backrest). Main Outcome Measures: Cadence, stroke angle, peak shoulder extension angle, shoulder flexion/extension range of motion, and mechanical effective force. Results: Pushing with the low backrest height enabled greater range of shoulder motion (P.01), increased stroke angle (P.01), push time (P.01), and reduced cadence (P.01) regardless of whether the treadmill was level or sloped. Conclusions: A lower cadence can be achieved when pushing with a lower backrest, which decreases the risk of developing upper-limb overuse related injuries. However, postural support, comfort, and other activities of daily living must also be considered when selecting a backrest height for active, longterm wheelchair users. The improvements found when using the low backrest were found regardless of slope type. Pushing uphill demanded significantly higher resultant and tangential force, torque, mechanical effective force, and cadence. Key Words: Exercise test; Rehabilitation; Spinal cord injuries; Wheelchairs. 2012 by the American Congress of Rehabilitation Medicine From the Department of Occupational Therapy, College of Health Science, Kaohsiung Medical University, Kaohsiung City (Yang, Chang); Department of Rehabilitation Medicine, Chung-Ho Memorial Hospital, Kaohsiung City (Yeh), Taiwan; Human Engineering Research Laboratories, Veterans Affairs Pittsburgh HealthCare System, Pittsburgh (Koontz); and Department of Rehabilitation Science and Technology, University of Pittsburgh, Pittsburgh (Koontz), PA. Supported by the Kaohsiung Medical University Research Foundation (grant no. Q097043) No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Reprint requests to Jyh-Jong Chang, PhD, Dept of Occupational Therapy, Kaohsiung Medical University, 100, Shi-chuan 1st Rd, Kaohsiung, Taiwan, e-mail: jjchang@kmu.edu.tw. In-press corrected proof published online on Feb 13, 2012, at www.archives-pmr.org. 0003-9993/12/9304-00757$36.00/0 doi:10.1016/j.apmr.2011.10.023 SELECTING AN APPROPRIATE manual wheelchair and seating system is an important decision for individuals with spinal cord injury (SCI). As these individuals spend an average of 9.2 h/d in their wheelchairs, 1 careful consideration should be given to ensure that the wheelchair/seating system is optimized for comfort, support, and function. Numerous studies have found that the setup of the wheelchair can affect the user s comfort, posture, stability, and ability to propel the wheelchair. 2-7 A properly adjusted wheelchair is more comfortable to sit in, makes maneuvering more efficient, and reduces musculoskeletal stress on the upper limbs. There are several critical seat dimensions to be considered during wheelchair selection: for example, seat vertical and fore-aft position, seat angle, seat depth, backrest angle, and backrest height. Most of the research has focused on studying the effects of vertical and fore-aft seat positioning with respect to the rear wheels. A low seat and a seat that is positioned posterior of the rear axle reduces forces, increases access to the pushrim (resulting in larger stroke angles), decreases cadence, and increases efficiency. 4,5,7-10 However, this same seat position also makes the wheelchair less stable, so additional training may be necessary to learn how to maneuver the wheelchair safely in this configuration. In the clinic, inclining the seat plane (also referred to as seat angle or seat dump) while keeping the backrest upright or performing a system tilt (eg, tilting back the seat plane and backrest together through an angle) commonly helps users with poor trunk control to fit more securely in the wheelchair and enables a higher level of comfort and upper-limb function. 11 Research has shown that a system tilt of 10 significantly increased the percentage of force that was applied in the tangential direction relative to the resultant force (ie, mechanical effective force [MEF]). On the other hand, reclining the backrest angle from 95 to 105 showed no significant effect on propulsion effectiveness. 2 Furthermore, neither tilting the seat angle nor increasing the backrest angle revealed any significant changes in shoulder joint moments during propulsion. 12 Because high forces and increased cadence have been linked to wrist and shoulder injuries, 13,14 a wheelchair configuration (seat position, seat inclination, backrest height) that increases propulsion efficiency (eg, MEF) and reduces force and cadence is likely to minimize injury risk. Little attention has been given to understanding the effects of back supports on propulsion biomechanics. From our clinical experience we found that many new lightweight manual wheelchairs come equipped with sling backrest upholstery at a standard height of 40.6cm (16in). Often the backrest, as well as other aspects of the wheelchair setup, remains in the default dimensions if the patient does not undergo a comprehensive MEF MWU SCI List of Abbreviations mechanical effective force manual wheelchair user spinal cord injury

BACKREST HEIGHT ON WHEELCHAIR PROPULSION, Yang 655 wheelchair evaluation with practitioners who are skilled in performing wheelchair/seating evaluations. The backrest height should be determined based on the wheelchair user s level of trunk impairment, balance control, and activity level, and it is important to recognize the tradeoff between support and function. 15 For example, high backrests provide more support to the spine and head but can impede the range of motion of the upper arms during activities of daily living. Thus, active manual wheelchair users (MWUs) who need more trunk support and who also self-propel are challenged with determining a height that provides the best balance between support and function. To our knowledge, there are no other studies that have investigated the effects of backrest heights on wheelchair propulsion biomechanics. Therefore, the purpose of this study was to investigate the influence of backrest height on propulsion during 2 different slope conditions. It was hypothesized that a low backrest would help MWUs with low paraplegia to increase range of shoulder movement, enabling greater access to the pushrim (resulting in larger stroke angles and lower cadence) and higher MEF during propulsion. We further hypothesized that the differences would hold true regardless of whether the subject was traversing a level or a sloped surface. METHODS Participants Participants were included in the study if they met the following criteria: (1) had SCI between the levels of T8 and L4, American Spinal Injury Association Impairment Scale grade A or B; (2) used a manual wheelchair as a primary mode of mobility; and (3) were between the ages of 20 and 65 years. Participants were excluded if they (1) had previous history of upper-extremity pain interfering with wheelchair propulsion or (2) had a heart or lung condition that is worsened by pushing a wheelchair. All participants signed informed consent in accordance with the procedures approved by the Institutional Review Board of the Kaohsiung Medical University Hospital prior to participation in the study. Experimental Setup The experimental design used a wheelchair accessible treadmill (Model KM-8520) a with a tread length of 1.8m, a tread width of 1.06m, and an adjustable incline from 0 to 10 (fig 1). An infrared light system was installed in the rear of the treadmill frame to trigger an emergency stop if a wheelchair user was unable to keep pace at the desired speed. Two lightweight manual wheelchairs (Karma KM-8520) a were used for the testing. The wheelchairs were identical on all accounts with the exception of seat width in order to accommodate different body sizes. One of the wheelchairs had a seat width of 40.6cm (16in) and the other had a seat width of 45.7cm (18in). The wheelchair seat width for each subject was determined by taking the widest point between the subjects hips when sitting comfortably and adding about 2.54cm (1in). The test wheelchair with the closest seat width to this measure was selected. Eighty-one percent of the participants were matched to the test wheelchair with the 40.6-cm (16-in) seat width, whereas the other participants were matched to the chair with the 45.7-cm (18-in) seat width. Each test wheelchair was set up with 1 of 2 different backrest heights: (1) high backrest as a fixed 40.6cm (16in) and (2) low backrest as 50% of the participant s trunk length. Trunk length was defined as the length from the top of the seat base to the acromion process while sitting in the wheelchair without a cushion, which was not used during the propulsion Fig 1. The experimental setup used in this study included a research treadmill, a SmartWheel, and a motion analysis system. Abbreviation: 3-D, 3-dimensional. trials. This backrest height algorithm was based on clinical experience of custom fitting wheelchairs to patients who have been seen in our wheelchair seating clinic. The backrest frame of each test wheelchair was modified so that it could be adjusted from a minimum height of 20cm to a maximum height of 40.6cm, measured from the intersection of the seat post and backrest post to the center top edge portion of the sling. No other wheelchair setup parameter (eg, seat height, seat to rear axle position, seat angle, and backrest angle) was altered in this study except the backrest height. Propulsion kinetics were obtained using a SmartWheel, b a 3-dimensional force- and torque-sensing pushrim, on both sides of the test wheelchair to measure forces and moments (see fig 1). The SmartWheel has demonstrated excellent accuracy 16,17 and has been used in many wheelchair propulsion studies. 2,8,12,18-20 A 6-camera Qualisys motion analysis system c was used to collect 3-dimensional coordinates of reflective markers (see fig 1) placed bilaterally on bony landmarks of the participant s upper extremities: acromion process, lateral epicondyle, olecranon, radial styloid, ulnar styloid, the head of the third metacarpal bone, and spinous processes of C7 and T2. In addition, 3 markers on the sternum were attached to a rigid body and were used to define a coordinate system that accounted for the movements of the shoulder. Procedure Each participant completed four 30-second propulsion trials (2 different slope conditions 2 different backrest heights) on the motor-driven treadmill at a constant speed of 0.9m/s. As most wheelchair users have never pushed on a treadmill, participants were asked to push the test wheelchair at least 2 minutes before data collection to help them acclimate to pushing on the treadmill. Trial order was randomized based on a computerized random function generator using Matlab software. d Participants rested at least 5 minutes between propulsion trials. Data Reduction Propulsion kinetic data were collected with a sample frequency of 240Hz and filtered with an 8th order Butterworth low-pass filter, zero lag, and 20Hz cutoff frequency. 21 After-

656 BACKREST HEIGHT ON WHEELCHAIR PROPULSION, Yang ward, kinetic data were linearly interpolated for synchronization with the kinematic data sampled at 120Hz. For each stroke, the SmartWheel provided forces (Fx, Fy, and Fz) and moments (Mx, My, and Mz) in 3 global reference planes. The origin of the 3 global reference planes was at the center of the Smart- Wheel. The positive x axis was oriented in the anterior direction, the positive y axis was oriented in the superior direction, the positive z axis for the right wheel was directed medially, and the positive z axis for the left wheel was directed laterally. The left wheel Fz and Mz were mirrored to the right wheel to align the data. The resultant force (Ftotal), which is the total force applied to the pushrim, was calculated by vector addition of Fx, Fy, and Fz. The tangential force, Ft, is the only component of Ftotal that directly contributes to rotation of the wheel. Ft was estimated by the moment around the hub, Mz, which was measured directly by the SmartWheel Ft Mz r, where r was the radius of the handrim. Mz comprises the amount of force, Ft, that contributes to forward motion, and a wrist moment, which was assumed to be negligible. 17 Propulsion MEF represented the ratio of tangential force and total resultant force and was calculated as 22 F2 t MEF F 2 total. The start and end of each push phase was determined visually by the presence and absence, respectively, of forces detected by SmartWheel. Mean tangential force, mean resultant force, mean torque, and the mean MEF were calculated for each push phase in each trial. Stroke cadence was defined as the number of strokes that occurred per second. Push time was defined as the time spent on the pushrim during the push phase. From the propulsion kinematic data, the start angle, end angle, and total stroke angle propelled during each stroke were determined. The start and end angles were defined as the angle between the line from the hand marker (on the third metacarpal) through the wheel axle, relative to the horizontal plane, at the start and the end of the push phase. The stroke angle was defined as the differential of the start and end angles. 8,23 Shoulder joint angles relative to the trunk were calculated using a local coordinate system approach described by Cooper et al. 24 To determine 3-dimensional glenohumeral motion, both humerus and trunk (sternum) local coordinate systems were first defined. Shoulder anatomical flexion/extension, abduction/ adduction, and horizontal flexion/extension angles were calculated by rotating the humerus into the trunk coordinate system. The peak extension angle and the range of shoulder flexion/ extension motion (differential of the maximal shoulder extension angle and maximal shoulder flexion angle) during the push phase of each stroke were selected for analysis. Statistical Analyses Propulsion kinetic/kinematic variables for 10 consecutive strokes in the middle of each trial were averaged. Because previous studies have found high correlations between rightand left-sided propulsion kinetics/kinematics, 13,18 the rightand left-sided data were averaged, resulting in 1 mean value for each dependent variable for each individual. A 2-way analysis of variance with repeated measures, with backrest height (high backrest and low backrest) and slope (0 and 3 ) as main factors, was carried out to determine whether a backrest height or a slope condition had a significant effect on the kinematic and kinetic variables. Statistical analyses were performed using the SPSS for Windows 11.0 software package. e The level of significance was set to.05. RESULTS Participant Characteristics Twenty-six men and 10 women with SCIs ranging from T8 to L2 participated in this study. Their mean age and years postinjury were 39.1 10.5 and 11.8 8.4 years, respectively. The minimal and maximal duration of wheelchair use among participants was 2.7 and 32.1 years, respectively. The average trunk height and the height from seat to the inferior angle of the scapula were 55.9 5.9 and 44.4 5.6cm, respectively. Main Effect of Backrest When participants propelled with the low backrest (27.6 3.2cm), push times were longer (P.01), cadence was lower (P.01), and stroke angles were larger (P.01) (see table 1). The larger stroke angles resulted from hand contact farther back on the pushrim (start angle; P.07) and hand release farther forward on the pushrim (end angle; P.01). Significantly larger shoulder extension angles (P.02) were found at the beginning of the push phase, as well as greater range of shoulder flexion/extension motion (P.01). No effect of backrest height on propulsion kinetics was found. Main Effect of Slope On the slope of 3, cadence increased (P.01) and the start and end angles were smaller (P.01), meaning that both hand contact and hand release shifted forward on the pushrim (see table 1). Participants also had greater range of shoulder flexion/ extension motion (P.01). Furthermore, participants demonstrated significantly larger resultant force (P.01), tangential force (P.01), propulsion torque (P.01), and MEF (P.01) when pushing uphill compared with the level condition. Interaction Effects As shown in table 1, no interaction effects were found. Therefore, we can conclude that the differences found between the 2 backrest heights were independent of the type of terrain traversed (sloped or flat). DISCUSSION Consistent with our hypothesis, lowering the height of the backrest enabled the arm to move more freely and as a result the shoulder passed through a greater range of motion. We also found increased shoulder extension and hand contact farther back on the pushrim, thereby contributing in part to the significantly larger stroke angles that were found. Interestingly, the larger stroke angles were also a result of hand release occurring farther forward on the rim even though other aspects of wheelchair setup were held constant and the trunk likely had little to no contact with the backrest at this point in the propulsion cycle. At the time of hand release at the end of the push phase the users flexed the trunk forward, especially when faced with sloped conditions, in order to keep the wheelchair stable. 25 It is possible that the recovery phase stroke pattern changed when using the low backrest because the arm was able to move more freely during this phase of the cycle. Variations in arm kinematics during recovery have been shown to influence propulsion variables such as stroke frequency and stroke angles during the push phase. 18,20 Our study found that the participants contacted the pushrim less frequently when using the low

BACKREST HEIGHT ON WHEELCHAIR PROPULSION, Yang 657 Table 1: Effects of Propulsion Variables While Using the High and Low Backrests Under 2 Different Slope Conditions Propulsion Variable Backrest Height Level Surface (0 Deg) Uphill Condition (3 Deg) Main Effect of Backrest Main Effect of Slope Interaction Effect Push time (s) HIGH 0.42 0.09 0.43 0.09.01*.36.95 LOW 0.44 0.10 0.45 0.10 Cadence (stroke/s) HIGH 1.24 0.32 1.43 0.30.01.01*.86 LOW 1.16 0.29 1.36 0.28 Start angle ( ) HIGH 97.43 10.28 92.67 10.62.07.01*.41 LOW 100.33 10.96 94.57 11.04 End angle ( ) HIGH 39.76 10.99 31.30 8.95.01*.01*.59 LOW 35.78 8.81 28.71 10.58 Stroke angle ( ) HIGH 57.67 14.44 61.37 16.88.01*.16.47 LOW 64.55 12.19 66.03 14.14 Shoulder extension angle ( ) HIGH 42.36 7.54 41.58 8.67.02.39.94 LOW 45.07 7.74 44.41 7.59 Shoulder ROM ( ) HIGH 48.53 11.92 57.96 11.79.01*.01*.96 LOW 52.38 10.30 61.69 10.62 Resultant force (N) HIGH 32.70 10.74 69.08 16.08.50.01*.62 LOW 32.37 10.41 67.42 14.63 Tangential force (N) HIGH 20.05 6.59 47.52 10.33.46.01*.93 LOW 19.35 6.08 46.64 11.00 Moment (Nm) HIGH 6.11 2.01 14.48 3.15.46.01*.93 LOW 5.90 1.85 14.22 3.35 MEF (%) HIGH 47.95 13.57 54.85 12.29.41.01*.21 LOW 44.99 12.44 55.22 14.52 NOTE. Data presented as mean SD. Abbreviations: HIGH, high backrest as a fixed 40.6cm (16in); LOW, low backrest as 50% of the participant s trunk length; ROM, range of motion. *P.01; P.05. backrest than when using the high backrest, to travel at the same speed. Because propulsion cadence has been linked to the risk of median nerve injury, 13 and wheelchair propulsion is a highly repetitive task with strokes occurring approximately once per second, it is critical that the backrest height be a major consideration in the wheelchair selection and configuration process, particularly for active, full-time MWUs. In contrast to our hypothesis, there was no main effect of backrest height on the MEF variable. We thought that freeing the arms would have facilitated a higher tangentially directed force, but instead subjects maintained the same MEF despite having positive alterations in their stroke and cadence. These results are consistent with wheelchair propulsion training studies, which have found that push angle and cadence are aspects of techniques that are more amenable to improvements in experienced wheelchair users. 26,27 MEF is strongly associated with wheelchair geometry and the mechanical constraints of arm positioning when the hand is in contact with the pushrim. Wheelchair users with SCI trained to push with a higher MEF experience a higher metabolic cost and report fatigue and difficulty in leaning forward and holding a trunk posture that enables directing force more tangentially along the pushrim. 26,27 Thus, it is very possible that the subjects in our study, being experienced MWUs, have optimized their MEF, and to increase MEF further would have required guided feedback and probable discomfort and energy losses. Moreover, the test wheelchairs in this study were equipped with sling backrests. Studies have shown that the use of a sling backrest in a wheelchair can have a negative impact on posture and can be less supportive than a rigid back. 3,6,28 In a recent study we investigated differences between a rigid backrest and the standard sling backrest on wheelchair propulsion variables in 26 MWUs with paraplegia. Under similar propulsion conditions as this study, the rigid backrest kept the trunk more upright, reduced nontangential propulsion forces, and increased MEF. 29 Consequently, there may be added benefits of pushing a wheelchair with a low rigid backrest instead of one with a low sling backrest. Although this finding was not directly tied to our hypothesis, we found no differences in any kinetic variables between the 2 backrest heights. However, because approximately the same amount of resultant force was used for both heights and because more time was spent on the pushrim with the low backrest to go the same speed as the high backrest, force would have been distributed over a greater period of time, implying a lower force impulse. Because cadence was also reduced with the low backrest, the amount of force absorbed by the joints over the course of daily propulsion would also be less. These are potentially added benefits for selecting a low backrest height. Therefore, a simple modification of lowering backrest height in wheelchair setup could help decrease the risk of developing upper-limb overuse related injuries. Further study is warranted to evaluate the impact of backrest height on force parameters not analyzed in this study but linked to injuries, such as force impulse and peak rate of force loading at the hand and upper-limb joints. Consistent with other studies, change in slope significantly affected propulsion kinematic and kinetic variables. Our results showed that the average propulsion resultant force and torque were over 2.09 and 2.38 times higher when pushing up the 3 slope in comparison with the level surface. This finding was consistent with previous work, but larger than the values reported by Richter et al. 30 This discrepancy may be attributed to different velocities of propulsion. Richter 30 tested 26 subjects with paraplegia and found that peak forces were twice as high when pushing up a 6 hill at a self-selected speed of 0.43m/s. In our study, participants pushed at 0.9m/s on the sloped condition. Nevertheless, Richter s 30 results combined with ours suggest that greater slopes demand a much higher force output.

658 BACKREST HEIGHT ON WHEELCHAIR PROPULSION, Yang Our findings on propulsion kinematic variables related to the slope conditions were also consistent with a previous study based on data collected at low slope (0 4 ) conditions. 25 When pushing a wheelchair uphill, MWUs must push against gravity to avoid rolling back down the hill. They may spend the same amount of time pushing, but must be quick in returning the arm during the recovery phase. As a result, an increase in propulsion cadence was found (P.01) when pushing uphill. The significant decrease in the start and end angles (P.01) indicated that the stroke angle shifted forward, likely from the increased trunk lean (eg, an anterior displacement of the center of mass) that is required to keep the wheelchair from tipping when traversing the slope. 25 Study Limitations Our sample included only MWUs with low paraplegia. Using similar low backrest heights for high paraplegia or tetraplegia may not be clinically appropriate. Back height selection must solve the dilemma of ensuring adequate postural support while also enabling the highest degree of upper-limb mobility for activities of daily living. This trade-off relationship becomes more significant as the level of SCI increases. Furthermore, we have found from our clinical experience that 50% of the participant s trunk length is low enough to free the upper extremities for greater movement but high enough to gain some support from the backrest. This setting might not be the optimal indicator for backrest height, but it could be a good starting point for future study. Further research is needed to determine a more precise setting for the backrest to improve wheelchair maneuverability, posture, and comfort. CONCLUSIONS Using a backrest height lower than 40.6cm (16in) afforded MWUs more freedom of arm movement, increased stroke angles, and decreased cadence. As a result, this simple modification in wheelchair setup could help decrease the risk of developing upper-limb overuse related injuries. The improvements found when using the low backrest were regardless of slope type. Consistent with findings in prior studies, pushing uphill demanded significantly higher resultant and tangential force, torque, MEF, and cadence. Ideally the backrest height should provide adequate postural support while affording as much freedom of arm movement as possible. Future studies should be directed on rigid backrests, as they come in various sizes and shapes and provide added benefits related to propulsion effectiveness and posture. References 1. Yang YS, Chang GL, Hsu MJ, Chang JJ. Remote monitoring of sitting behaviors for community-dwelling manual wheelchair users with spinal cord injury. Spinal Cord 2009;47:67-71. 2. Aissaoui R, Arabi H, Lacoste M, Zalzal V, Dansereau J. Biomechanics of manual wheelchair propulsion in elderly: system tilt and back recline angles. Am J Phys Med Rehabil 2002;81:94-100. 3. Harms M. Effect of wheelchair design on posture and comfort of users. Physiotherapy 1990;76:266-71. 4. Kotajarvi BR, Sabick MB, An KN, Zhao KD, Kaufman KR, Basford JR. The effect of seat position on wheelchair propulsion biomechanics. J Rehabil Res Dev 2004;41:403-14. 5. Masse LC, Lamontagne M, O Riain MD. Biomechanical analysis of wheelchair propulsion for various seating positions. J Rehabil Res Dev 1992;29:12-28. 6. Parent F, Dansereau J, Lacoste M, Aissaoui R. Evaluation of the new flexible contour backrest for wheelchairs. J Rehabil Res Dev 2000;37:325-33. 7. van der Woude LH, Veeger DJ, Rozendal RH, Sargeant TJ. Seat height in handrim wheelchair propulsion. J Rehabil Res Dev 1989;26:31-50. 8. Boninger ML, Baldwin M, Cooper RA, Koontz A, Chan L. Manual wheelchair pushrim biomechanics and axle position. Arch Phys Med Rehabil 2000;81:608-13. 9. Gutierrez DD, Mulroy SJ, Newsam CJ, Gronley JK, Perry J. Effect of fore-aft seat position on shoulder demands during wheelchair propulsion: part 2. An electromyographic analysis. J Spinal Cord Med 2005;28:222-9. 10. Mulroy SJ, Newsam CJ, Gutierrez DD, et al. Effect of fore-aft seat position on shoulder demands during wheelchair propulsion: part 1. A kinetic analysis. J Spinal Cord Med 2005;28:214-21. 11. Cooper RA. Wheelchair selection and configuration. New York: Demos; 1998. 12. Desroches G, Aissaoui R, Bourbonnais D. Effect of system tilt and seat-to-backrest angles on load sustained by shoulder during wheelchair propulsion. J Rehabil Res Dev 2006;43:871-82. 13. Boninger ML, Cooper RA, Baldwin MA, Shimada SD, Koontz A. Wheelchair pushrim kinetics: body weight and median nerve function. Arch Phys Med Rehabil 1999;80:910-5. 14. Collinger JL, Boninger ML, Koontz AM, et al. Shoulder biomechanics during the push phase of wheelchair propulsion: a multisite study of persons with paraplegia. Arch Phys Med Rehabil 2008;89:667-76. 15. Koontz AM, Karmarkar A, Spaeth DM, Schmeler MR, Cooper RA. Prescription of wheelchairs and seating systems. In: Braddom RL, editor. Physical medicine and rehabilitation. 4th ed. Philadelphia: Saunders/Elsevier; 2010. p 373-402. 16. Asato KT, Cooper RA, Robertson RN, Ster JF. SMARTWheels: development and testing of a system for measuring manual wheelchair propulsion dynamics. IEEE Trans Biomed Eng 1993;40: 1320-4. 17. Cooper RA, Robertson RN, VanSickle DP, Boninger ML, Shimada SD. Methods for determining three-dimensional wheelchair pushrim forces and moments: a technical note. J Rehabil Res Dev 1997;34:162-70. 18. Boninger ML, Souza AL, Cooper RA, Fitzgerald SG, Koontz AM, Fay BT. Propulsion patterns and pushrim biomechanics in manual wheelchair propulsion. Arch Phys Med Rehabil 2002;83:718-23. 19. Koontz AM, Cooper RA, Boninger ML, Souza AL, Fay BT. Shoulder kinematics and kinetics during two speeds of wheelchair propulsion. J Rehabil Res Dev 2002;39:635-49. 20. Koontz AM, Roche BM, Collinger JL, Cooper RA, Boninger ML. Manual wheelchair propulsion patterns on natural surfaces during start-up propulsion. Arch Phys Med Rehabil 2009;90:1916-23. 21. Cooper RA, DiGiovine CP, Boninger ML, Shimada SD, Robertson RN. Frequency analysis of 3-dimensional pushrim forces and moments for manual wheelchair propulsion. Automedica 1998; 16:355-65. 22. Boninger ML, Cooper RA, Robertson RN, Shimada SD. Threedimensional pushrim forces during two speeds of wheelchair propulsion. Am J Phys Med Rehabil 1997;76:420-6. 23. Vanlandewijck Y, Theisen D, Daly D. Wheelchair propulsion biomechanics: implications for wheelchair sports. Sports Med 2001;31:339-67. 24. Cooper RA, Boninger ML, Shimada SD, Lawrence BM. Glenohumeral joint kinematics and kinetics for three coordinate system representations during wheelchair propulsion. Am J Phys Med Rehabil 1999;78:435-46. 25. Chow JW, Millikan TA, Carlton LG, Chae WS, Lim YT, Morse MI. Kinematic and electromyographic analysis of wheelchair propulsion on ramps of different slopes for young men with paraplegia. Arch Phys Med Rehabil 2009;90:271-8. 26. de Groot S, de Bruin M, Noomen SP, van der Woude LH. Mechanical efficiency and propulsion technique after 7 weeks of

BACKREST HEIGHT ON WHEELCHAIR PROPULSION, Yang 659 low-intensity wheelchair training. Clin Biomech (Bristol, Avon) 2008;23:434-41. 27. Kotajarvi BR, Basford JR, An KN, Morrow DA, Kaufman KR. The effect of visual biofeedback on the propulsion effectiveness of experienced wheelchair users. Arch Phys Med Rehabil 2006;87: 510-5. 28. Alm M, Gutierrez E, Hultling C, Saraste H. Clinical evaluation of seating in persons with complete thoracic spinal cord injury. Spinal Cord 2003;41:563-71. 29. Yang YS, Lin SA, Chang JJ. The biomechanical analysis of effect of rigid backrest on wheelchair propulsion among people with spinal cord injury. Taiwan J Occup Ther Assoc 2010; 28:1-13. 30. Richter WM, Rodriguez R, Woods KR, Axelson PW. Stroke pattern and handrim biomechanics for level and uphill wheelchair propulsion at self-selected speeds. Arch Phys Med Rehabil 2007; 88:81-7. Suppliers a. Karma Medical Products Co, NO.2363, Sec 2, University Rd, Chia-Yi 621, Taiwan. b. Three Rivers Holdings, LLC, 1826 W Broadway Rd, Ste 43, Mesa, AZ 85202. c. Qualisys AB, Packhusgatan 6, S-411 13 Gothenburg, Sweden. d. The MathWorks Inc, 3 Apple Hill Dr, Natick, MA 01760-2098. e. SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606-6307.