(Stroke. 1997;28:722-728.)© 1997 American Heart Association, Inc.

Articles

Task-Related Training Improves Performance of Seated Reaching Tasks After Stroke

A Randomized Controlled Trial

Catherine M. Dean, MA Roberta B. Shepherd, EdD

From the Faculty of Health Sciences, The University of Sydney (Australia).

Correspondence to Catherine Dean, School of Physiotherapy, Faculty of Health Sciences, The University of Sydney, East St, PO Box 170, Lidcombe, NSW, 2141, Australia. E-mail c.dean@cchs.usyd.edu.au.

   Abstract

 Background and Purpose After stroke, the ability to balance

in sitting is critical to independence. Although impairments in sitting balance are common, little is known about the effectiveness of rehabilitation strategies designed to improve it. The purpose of this randomized placebo-controlled study was to evaluate the effect of a 2-week task-related training program aimed at increasing distance reached and the contribution of the affected

lower leg to support and balance.

Methods Twenty subjects at least 1 year after stroke were randomized into an experimental or control group. The experimental group participated in a standardized training program involving practice of reaching beyond arm's length. The control group received sham training involving completion of cognitive-manipulative tasks within arm's length. Performance of reaching in sitting was measured before and after training using electromyography, videotaping, and two force plates. Variables tested were movement time, distance reached, vertical ground reaction forces through the feet, and muscle activity. Subjects were also tested on sit-to-stand, walking, and cognitive tasks. Nineteen subjects completed the study.

Results After training, experimental subjects were able to reach faster and further, increase load through the affected foot, and increase activation of affected leg muscles compared with the control group (P<.01). The experimental group also improved in

TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences

sit-to-stand. The control group did not improve in reaching or sit-to-stand. Neither group improved in walking.

Conclusions This study provides strong evidence of the efficacy of task-related motor training in improving the ability to balance during seated reaching activities after stroke.

Key Words: exercise • hemiplegia • motor activity • rehabilitation • stroke outcome

   Introduction

 The ability to balance while reaching for a variety of objects

both within and beyond arm's length is critical to independent

living. Balance impairments in sitting are common after stroke,1 2 and sitting balance has been shown to be a useful prognostic indicator of outcome for this population.2 3 4 It has been reported that more than 70% of stroke patients admitted to rehabilitation are unable to reach sideways to the floor while sitting.5 The disability in reaching tasks arises not only from the impairments

caused by the neural lesion, such as weakness and loss of coordinated movement, but also from the tendency to adapt behavior to avoid the threat to balance.6

Restoration of sitting balance is one of the goals of rehabilitation; however, the effect of sitting balance training with hemiplegic patients has not been specifically investigated. The most commonly used method of neurological rehabilitation is Bobath therapy.7 8 This therapy does not use the training of self-initiated activities in sitting as a means of improving balance control. Instead, emphasis is on the patient responding to physiotherapist-induced movement.9 10 11 In addition, Bobath therapy emphasizes the role of the trunk and not the lower limbs.9 10 11 Recent evidence suggests that such a therapeutic approach may not be well founded. For example, the leg muscles are more impaired than the trunk muscles after stroke, possibly due to the greater bilateral innervation of trunk muscles,12 and there is evidence that the muscles

of the lower limbs are critical to moving the body mass over the base of support. In contrast to the Bobath approach, task-related training of volitional actions in sitting has for several years been proposed as a method of improving balance control in sitting.13 14 15

Performance of seated reaching tasks requires the coordinated motion of the trunk and upper limbs.16 17 18 Healthy subjects are able to reach significantly further when the feet are in contact with the ground compared with when they are not.19 However, the

role of lower limbs is not only to provide a larger base of support; according to recent studies, the lower limbs also play an active role in balance.16 20 21 22 23 Several factors have been identified that influence the contribution of the lower limbs to balance in sitting. Distance and direction of reach, seat height, and extent of thigh support on the

TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences

seat have all been shown to affect the magnitude of the load borne through the feet and in some cases also the activity in leg muscles.16 20 21 22 23 24

The purpose of this investigation, a randomized placebo-controlled trial of task-related training after stroke, was to examine the effect of a training program designed to improve the ability to balance in sitting after stroke. Individuals at least 1 year after stroke who were discharged from all formal rehabilitation services were recruited as subjects. These criteria were set to ensure that the amount and type of training could be controlled. In addition, the possibility that spontaneous recovery would confound

any training effects was eliminated. Specifically, the aims of training were to increase the distance reached and the active contribution of the lower limbs to support and balance. A standardized training program was devised based on research into the performance of seated reaching tasks and on a published program of motor training after stroke.13

   Subjects and Methods

 Twenty stroke subjects were recruited for this study through written advertisements sent to all Stroke Recovery Clubs in metropolitan Sydney. To participate, subjects had to meet the following criteria: (1) diagnosis of stroke resulting in hemiplegia at least 12 months

ago; (2) discharge from all rehabilitation services; (3) ability to understand instructions; (4) ability to give informed consent; (5) no orthopedic problem that would interfere with the ability to perform seated reaching tasks; and (6) ability to sit unsupported for a period of 20 minutes. All subjects lived either at home or in a retirement village and were able to walk short distances within the home. No subject had hemianopsia or any obvious cognitive or perceptual problems as evaluated with the Mini-Mental State Examination25 and the letter cancellation test.26 Individual subject characteristics are outlined in Table 1 . All procedures were performed in accordance with the approval of the University of Sydney Ethics Committee.

View this table:[in this window]

[in a new window] 

Table 1. Subject Characteristics

DesignA pretest and posttest group design was used. Subjects were randomly assigned to either the experimental or control group. Randomization was blocked to ensure equal numbers in the groups. The procedure involved random sampling without

TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences

replacement; subjects drew a card from a box that was originally filled with 10 control and 10 experimental cards. One subject (subject 17) from the control group dropped out of the study because of an acute neurological episode that required hospitalization. There were no significant differences between the groups in terms of age, time since stroke, or walking velocity (age, P=.717; time since stroke, P=.864; walking velocity, P=.248).

Training ProgramsSubjects in both groups participated in standardized training programs. Each program consisted of 10 sessions spread over a 2-week period. The training programs were carried out by the first author (C.D.) in the subject's home. The training for the experimental group was designed to improve sitting balance and involved emphasis on appropriate loading of the affected leg while practicing reaching tasks using the unaffected hand to grasp objects located beyond arm's length. The reaching tasks were performed under systematically varied conditions. Distance and direction were varied by changing the location of the object. Seat height, movement speed, object weight, and extent of thigh support on the seat were also varied. The training was advanced by increasing the number of repetitions and complexity of the tasks over the 2-week period.

The control group had sham training that incorporated the performance of cognitive-manipulative tasks while seated at a table. Sham training was performed so that subjects would consider themselves involved in a training program and to eliminate any effect due to placebo. Subjects were seated in a chair with arm and back supports

and the forearms resting on a table. They performed manipulative tasks using the unaffected hand over small distances (less than 50% of arm length). Training was advanced over sessions by increasing the repetitions and cognitive difficulty of the tasks. The subjects in the control group performed an equal number of reaching movements as the subjects assigned to the experimental group; however, the nature of the tasks ensured that only a minimum balance perturbation occurred.

Subjects in both groups were given feedback about their performance, all tasks were timed, and subjects were encouraged to perform the tasks faster. Each session took an average of 30 minutes to complete.

Testing ProtocolAt pretests and posttests conducted in the University laboratory, subjects were measured on two seated reaching tests. In addition, sit-to-stand, walking, and cognitive-manipulative tasks were examined. Walking speed and cognitive-manipulative tasks were evaluated by an assessor blinded to the subject's group allocation. Biomechanical data collection and analysis for the seated reaching tasks and sit-to-stand were computerized, which minimized experimenter bias because group allocation was not evident to the operator.

Reaching TestsSubjects were videotaped from the side as they reached with the unaffected hand to pick up and drink water from a glass under three reach direction conditions: forward, 45° toward the unaffected side (ipsilateral), and 45° across the body toward the affected side (across) (Fig 1 ).

View larger version (20K):[in this window]

[in a new window] 

Figure 1. Experimental setup. Inset shows diagram of reach direction conditions.

Subjects sat on a height-adjustable stool with each foot resting on adjacent force plates (Advanced Medical Technology Inc, model OR6-5-1) that sampled ground reaction forces (GRF) at 100 Hz for 10 seconds. Vertical GRF data through each plate were extracted and normalized for body weight.

Subjects started with the unaffected hand resting on a pressure-sensitive switch with a sensitivity of 5 g that was placed on the side of the stool and connected to a light. Deactivation of the light signaled the start of hand movement. The starting position was standardized. Seat height was adjusted to 100% lower leg length, measured from the lateral knee joint line to the floor with the subject standing barefoot. A steel plate (50x30 cm) mounted on a steel pole formed a height-adjustable table. Table height was adjusted to 75% of shoulder height, defined as the distance from the shoulder

marker to the floor with the subject sitting in the standardized position. Reaching distance was standardized to 140% of arm length. Subjects reached for a glass filled with water that was placed over a pressure-sensitive table switch. Lifting the glass activated a light, which defined the end of hand movement.

Adhesive surface electromyographic (EMG) electrodes (Meditrace, 11 mm, self-adhesive silver/silver chloride) were placed over the following muscles: anterior deltoid of the unaffected arm and bilaterally over the lateral vastus, anterior tibial, and

soleus. Electrode placement was in accordance with the location described by Basmajian and Blumenstein.27 EMG signals were acquired through isolated preamplifiers and AC amplifiers (Neomedix) with a band-pass filter from 5 Hz to 1 kHz and incorporating a 50-Hz notch filter. The EMG signal was sampled at 1001 Hz

using an analog-to-digital converter (Data Translation DT2814). To detect significant muscle onsets, EMG analysis involved using a computer program based on an algorithm developed by Di Fabio.28 The EMG signal was full-wave-rectified and a mean baseline level of activity measured before the reaching activity was calculated.

The program detected when the EMG activity exceeded the mean±3 SD for a period of at least 25 milliseconds. All onsets were checked with visual inspection.

Subjects were instructed to reach for the glass and drink from it. No instructions about movement speed were given. Five practice trials were given before data collection in each condition, and four trials in each of the three conditions were analyzed. The order

of conditions was randomized across subjects. For each subject, the order of testing was identical for both pretests and posttests.

Maximum distance that subjects could reach was tested by having subjects reach in each direction as far as possible. The best attempt from three was measured to the nearest 0.5 cm using a steel pole with millimeter increments.

Additional TestsSubjects were asked to stand up with their weight evenly distributed through both feet from a seat height of 115% of lower leg length. GRF data were collected for three trials. The time taken to walk 10 m was measured with a stopwatch. Cognitive function was measured on a variety of tasks, and the time taken to complete these tasks was recorded. The tasks assessed were letter cancellation,26 number of words found in a word puzzle within 3 minutes, and mathematics questions.

Statistical AnalysisBoth between- and within-group comparisons were made. For the between-group

comparisons, the change scores from pretest to posttest on all dependent variables were calculated. All variables except the EMG activity were compared using two-tailed independent sample t tests. When the data did not meet the assumptions of normality and equal variance, nonparametric tests were also used. Since they produced the same results as the parametric tests, t test results are presented for all variables. EMG activity was evaluated using the Mann-Whitney U test. Within-group analysis used two-tailed paired t tests for pretest and posttest scores on all variables except EMG, which was analyzed using the Wilcoxon ranked-sign test. Significance was set at P<.01.

   Results

 Reaching TestsSubjects in the experimental group performed significantly better than the control group on the reaching tasks after 2 weeks of training (Fig 2 , Table 2 ). Subjects in the experimental group improved from pretesting to posttesting on all reaching variables tested. Control group subjects did not, showing a decline in some scores by posttest.

View larger version (13K):[in this window]

Figure 2. Mean±SEM change in scores between pretests and posttests on the reaching tests for experimental (open bars) and control (shaded bars) groups. Left, Maximum distance reached; middle, hand movement time (note positive

TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences

[in a new window] 

values indicate a decrease in hand movement time); right, peak vertical GRF through affected foot for the three reach direction conditions. Horizontal bars indicate significant between-group differences and asterisks significant within-group differences (P<.01). GRF indicates ground reaction forces; BW, body weight.

View this table:[in this window]

[in a new window]

 

Table 2. Mean±SEM Pretest and Posttest Scores for Reaching Test Variables

After training, the experimental group significantly increased the maximum reaching distance in all directions. In contrast, the control group did not improve. The difference between the groups was significant for all directions (P<.001).

Experimental subjects performed the reaching tasks in a shorter time than control subjects. Differences between the groups were significant in the two diagonal reach conditions (ipsilateral, P=.008; across, P=.001), and there was a trend toward a reduction in movement time for the experimental group compared with the control group in the forward condition (P=.048).

Experimental subjects showed a significant increase in peak vertical GRF through the affected foot in the forward and across conditions after training. Mean peak vertical GRF through the affected foot increased to 25.5% and 33.3% body weight for the

forward and across conditions, respectively. In contrast, the control group showed no improvement in any direction and a slight decrease in peak vertical GRF in the forward condition. The difference between the groups was significant for the forward

and across conditions (P<.001). Note that peak vertical GRF through the affected foot in the ipsilateral direction would not be expected to increase because, in this condition, the unaffected lower limb would be making the major contribution to support and balance.

The improvements in pattern and amplitude of loading through the affected foot and the reduction in hand movement time by the experimental group from pretest to posttest are illustrated in Fig 3 , where representative trials are presented. Pretraining

trials illustrate two deficits that are typical of stroke: (1) slowness of movement (in the across condition, the subject took almost 5 seconds to reach the glass) and (2) lack of loading of the affected leg (regardless of reach direction, the subject did not load the affected foot). Note that posttraining graphs are similar to the patterns of force displayed by a healthy subject. Fig 4 shows a representative trial from a subject in the control group, illustrating the lack of improvement in force production and movement time at posttest.

View larger version (23K):[in this window]

[in a new window] 

Figure 3. Vertical ground reaction forces (GRF) through each foot during the three reach conditions. Top and middle, Representative trials from an experimental subject pretest and posttest (heavy line, affected foot; light line, unaffected foot). Bottom, A trial from a healthy 62-year-old subject reaching with his right hand (heavy line, left foot; light line, right foot). Dotted vertical lines indicate onset and end of hand movement; BW, body weight.

View larger version (16K):[in this window]

[in a new window] 

Figure 4. Representative trials from a subject in the control group. Format and abbreviations are as for Fig 3 .

Subjects in the experimental group were able to activate muscles in the affected leg (lateral vastus, anterior tibial, and soleus) in more trials at posttest than subjects in the control group (Table 3 ). This increased consistency of muscle activation was particularly noticeable in the forward and across conditions, the conditions in which the affected leg was bearing weight. There was a significant difference between the groups in anterior tibial and soleus muscle activation in the forward condition (anterior tibial, P=.005; soleus, P=.006). There were no other significant between- or within-group differences.

View this table:[in this window]

[in a new window] 

Table 3. Percentage of Trials Each Muscle Was Activated

Other Functional TestsFor the experimental group, there was a significant increase in the peak vertical GRF produced through the affected foot when subjects stood up from sitting at posttest (P=.002). Overall, this increase was from 43% to 51% body weight. In contrast, there was no significant difference from pretest to posttest for the control group (P=.247).

There were no significant differences between or within groups on time taken to walk 10 m. The control group performed better than the experimental group on the three cognitive tasks (letter cancellation, word puzzle, and mathematics questions). However, the difference between the groups was only significant on the letter cancellation task (P=.002).

   Discussion

 The major finding of this study was that stroke patients who

were specifically trained to improve their sitting balance were able to reach further and faster, increase the load taken through the affected foot when this was required by the task, and increase the consistency in activation of task-related muscles in the affected leg. In contrast, subjects in the control group showed no improvement in these variables, with some variables showing a decline in score. The results demonstrate that the 2-week program that involved training in a variety of tasks was effective in improving performance.

On completion of training, the experimental subjects had not only increased the speed of reaching but were also able to reach at a speed comparable with that of healthy, similarly aged subjects.23 Two particular aspects of the training program may have contributed to the increase in movement speed. First, the trainer specifically instructed subjects to increase their speed during practice and provided feedback about movement time. Second, successful accomplishment of one of the tasks trained required subjects to complete it within the shortest period possible.

The ability to reach further and faster appears to be directly linked to the improved ability to use the affected leg for support and balance. The beneficial effect of training on affected lower limb function shown in this study is reflected in the significant

increase in the load taken through the affected foot when subjects reached forward and toward that side. Notably, the values attained at posttest approached values reported for healthy subjects reaching in these directions.21 This improvement probably reflects

certain factors incorporated into the training program and illustrates the effectiveness of basing rehabilitation strategies on critical biomechanical features. Several factors were systematically manipulated. Distance to be reached was increased over the training period because the load taken through the feet increases as reaching distance

increases.16 The extent of thigh support on the seat was reduced within each training session because there is an inverse relationship between extent of support and magnitude of the load taken through the feet.21 22 Reaching direction was varied because direction of reach determines the relative contribution from each lower limb to support and balance.20 21

The findings illustrate the role of the lower limbs in balancing the body in sitting during reaching tasks. Peak vertical GRF through the feet occurs normally around the end of the reach. At this time, force through the feet acts to brake the forward

TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences

momentum of the body mass and to prevent a fall forward. Additional force enables the return of the upper body to the upright position. After training, stroke subjects had improved the timing of peak vertical GRF to normal and were able to activate anterior tibial and soleus muscles in the affected leg more consistently. Activation of these muscles, which link the shank to the foot, assists in balancing the body mass during distant reaches both in sitting20 and standing.29 Therefore, both GRF and EMG findings support the view that the lower limbs play an active role in support and balance during seated reaching tasks and that stroke patients can be trained to use their affected lower limb more actively when reaching beyond arm's length.

Overall, the findings of this study support the concept of specificity of training, which has been discussed in relation to able-bodied subjects30 and proposed as a means of rehabilitating the movement disabled.13 14 15 31 Experimental subjects improved on the actions that they practiced. However, they also improved on an action (sit-to-stand)

that was not itself included in the training. The improvement in force production in sit-to-stand but not in walking speed suggests that generalization to a biomechanically similar action occurred. Rapid movement of the trunk forward and loading of the feet, both of which were practiced as part of reaching to a distant object, are also critical biomechanical components of the early phase of sit-to-stand.32 In contrast, the performance of control subjects on reaching tasks, which they had not practiced, actually declined, while performance of the cognitive-manipulative tasks that they had practiced improved, in one task significantly.

The results have implications for rehabilitation, demonstrating that stroke patients can improve their performance on seated reaching tasks to near normal with a short task-related training program that takes into account normative biomechanics and lower limb function. The findings directly challenge the assumption that improvement in function after stroke is due only to spontaneous recovery. In addition, the outcome is consistent with the increasing evidence that stroke patients can improve their performance of specific tasks if those tasks are included in training and practiced.33 34 35 It is reasonable to assume that such training would also be effective in many patients if implemented early after stroke, when the potential for neuroplasticity may be greatest.36

In conclusion, this study investigated the efficacy of a motor training program designed to improve sitting balance control in stroke patients after discharge from rehabilitation. The results demonstrate that the training program, which was designed

from existing scientific knowledge about human movement and how persons acquire skill in movement, was effective. Training involved intensive practice of reaching to objects beyond arm's length with systematic variation of speed, type of task, seat

height, extent of thigh support, and distance and direction reached. Subjects significantly increased the contribution from the affected lower limb to support and balance and were able to reach faster and further. Furthermore, they also significantly increased the load taken through the affected foot when standing up from sitting. The study highlights the value of designing scientifically based rehabilitation programs.

   Acknowledgments

 This project was partially funded by the School of Physiotherapy, The University of Sydney, and by financial support from an Australian Postgraduate Award and the Cumberland Burniston Foundation (C. Dean). The assistance of Darran Dawson, Anne Moseley, Merren Richardson, and Wantana Vattanasilp during data collection and Dr Roger Adams during data analysis is gratefully acknowledged. We also acknowledge the helpful comments made by Associate Professor Janet Carr and Dr Jean-Michel Gracies.

Received December 5, 1996; revision received January 23, 1997; accepted January 23, 1997.

   References

 

1. Wade DT, Wood VA. Langton-Hewer R. Recovery after stroke: the first 3 months. J Neurol Neurosurg Psychiatry. 1985;48:7-13. [Abstract]

2. Morgan P. The relationship between sitting balance and mobility outcome in stroke. Aust J Physiother. 1994;40:91-96.

3. Loewen SC, Anderson BA. Predictors of stroke outcome using objective measurement scales. Stroke. 1990;21:78-81. [Abstract]

4. Sandin KJ, Smith BS. The measure of balance in sitting in stroke rehabilitation prognosis. Stroke. 1990;21:82-86. [Abstract]

5. Dean CM, Mackey FH. Motor Assessment Scale scores as a measure of rehabilitation outcome following stroke. Aust J Physiother. 1992;38:31-35.

6. Shepherd RB. Adaptive motor behavior in response to perturbations of balance. Physiotherapy Theory and Practice. 1992;8:137-143.

7. Nilsson L, Nordholm L. Physical therapy in stroke rehabilitation: bases for Swedish physiotherapists' choice of treatment. Physiotherapy Theory and Practice. 1992;8:49-55.

8. Carr JH, Mungovan SF, Shepherd RB, Dean CM, Nordholm LA. Physiotherapy in stroke rehabilitation: bases for Australian physiotherapists' choice of treatment. Physiotherapy Theory and Practice. 1994;10:201-209.

9. Bobath B. Adult Hemiplegia: Evaluation and Treatment. 3rd ed. London, UK: William Heinemann Medical Books; 1990.

TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences

10. Ryerson SD. Hemiplegia. In: Umphred DA, ed. Neurological Rehabilitation. 3rd ed. St Louis, Mo: CV Mosby Co; 1995:681-721.

11. Davies PM. Steps to Follow. New York, NY: Springer-Verlag; 1985.

12. Kuypers HGJM. Anatomy of the descending pathways. In: Brooks VB, Brookhart JM, eds. Handbook of Physiology: The Nervous System II. New York, NY: American Physiology Society; 1981:597-666.

13. Carr JH, Shepherd RB. A Motor Relearning Programme for Stroke. 2nd ed. Oxford, UK: William Heinemann Medical Books; 1987.

14. Carr JH, Shepherd RB. A motor learning model for stroke rehabilitation. Physiotherapy. 1989;89:372-380.

15. Carr JH, Shepherd RB. Reflections on physiotherapy and the emerging science of movement rehabilitation. Aust J Physiother. 1994;40th Jubilee:39-47.

16. Dean CM, Shepherd R, Adams R. Intersegmental coordination during reaching in seated subjects. In: Proceedings of the 12th International Congress of World Confederation of Physical Therapy; June 25-30, 1995; Washington, DC. Abstract, p 720.

17. Kaminski TR, Bock C, Gentile AM. The coordination between the trunk and arm motion during pointing movements. Exp Brain Res. 1995;106:457-466. [Medline] [Order article via Infotrieve]

18. Son K, Miller JAA, Schultz AB. The mechanical role of the trunk and lower extremities in a seated weight-moving task in the sagittal plane. J Biomech Eng. 1988;110:97-103. [Medline] [Order article via Infotrieve]

19. Chari VR, Kirby RL. Lower-limb influence on sitting balance while reaching forward. Arch Phys Med Rehabil. 1986;67:730-733. [Medline] [Order article via Infotrieve]

20. Crosbie J, Shepherd R, Squire T. Postural and voluntary movement during reaching in sitting: the role of the lower limbs. J Hum Movement Stud. 1995;28:103-112.

21. Dean CM, Shepherd RB, Adams R. The effect of reach direction and extent of thigh support on the forces through the feet during seated reaching tasks. In: Proceedings of the First Australasian Biomechanics Conference; February 1-2, 1996; Sydney, Australia. Abstract, p 18.

22. Lino F, Bouisset S. Is velocity of pointing movement performed in a sitting posture increased by upper body instability? In: Proceedings of the XIVth International Society of Biomechanics Conference; July 4-8, 1993; Paris, France. Abstract, p 802.

23. Teyssedre C, Zattara M, Lino F. Does postural muscular activity associated with a pointing task depend on handedness? In: Proceedings of the XIVth International Society of Biomechanics Conference; July 4-8, 1993; Paris, France. Abstract, p 1332.

24. Arborelius UP, Wretenberg P, Lindberg F. The effects of arm rests and high seats on lower-limb joint load and muscular activity during sitting and rising. Ergonomics. 1992;35:1377-1391. [Medline] [Order article via Infotrieve]

25. Bleecker M, Bolla-Wilson K, Kawas C, Agnew J. Age-specific norms for the Mini-Mental State Examination. Neurology. 1988;38:1565-1568. [Abstract/ Free   Full   Text]

26. Wilson B, Cockburn J, Halligan PW. Development of a behavioural test of visuospatial neglect. Arch Phys Med Rehabil. 1987;68:98-102. [Medline] [Order article via Infotrieve]

27. Basmajian JV, Blumenstein R. Electrode Placement in EMG Biofeedback. Baltimore, Md: Williams & Wilkins; 1980.

28. Di Fabio RP. Reliability of computerized surface electromyography for determining the onsets of muscle activity. Phys Ther. 1987;6:43-48.

29. Cordo PJ, Nashner LM. Properties of postural adjustment associated with rapid arm movement. J Neurophysiol. 1982;47:287-302. [Abstract/ Free   Full   Text]

30. Rutherford OM. Muscular coordination and strength training implications for injury rehabilitation. Sports Med. 1988;5:196-202. [Medline] [Order article via Infotrieve]

31. Ada L, Canning CG, Carr JH, Kilbreath SL, Shepherd RB. Task-specific training of reaching and manipulation. In: Bennett KMB, Castellio U, eds. Insights into the Reach to Grasp Movement. New York, NY: North Holland; 1994:239-265.

32. Shepherd RB, Gentile AM. Sit to stand: functional relationship between upper body and lower limb segments. Hum Movement Sci. 1994;13:817-840.

33. Hesse S, Bertelt C, Jahnke MT, Schaffrin A, Baake P, Malezic M, Mauritz KH. Treadmill training with partial body weight support compared with physiotherapy in nonambulatory hemiparetic patients. Stroke. 1995;26:976-981. [Abstract/ Free   Full   Text]

34. Richards CL, Malouin F, Wood-Dauphinee S, Williams JI, Bouchard JP, Brunet D. Task-specific physical therapy for optimization of gait recovery in acute stroke patients. Arch Phys Med Rehabil. 1993;74:612-620. [Medline] [Order article via Infotrieve]

35. Yekutiel M, Guttman E. A controlled trial of retraining of the sensory function of the hand in stroke patients. J Neurol Neurosurg Psychiatry. 1993;56:241-244. [Abstract]

36. Lee RG, van Donkelaar P. Mechanisms underlying functional recovery following stroke. Can J Neurol Sci. 1995;22:257-263.[Medline] [Order article via Infotrieve]