Over 50,000 Canadians suffer from stroke each year, making it the number one cause of neurological disability in Canada today (22) and a leading cause of disability in the community (18). Ninety percent of stroke survivors have some functional disability, with mobility being the major impairment (20). Although some individuals with stroke will have received some rehabilitation during the acute and subacute phase, rarely does rehabilitation extend beyond 1 yr postinjury due to the belief that functional recovery has plateaued by this time (41). Impairments resulting from stroke, such as muscle weakness, pain, spasticity, and poor balance, in addition to the lack of accessible and appropriate community-based exercise programs, can lead to reduced tolerance to activity, further sedentary lifestyle, and additional declines in function and disability status (31).
Activities that promote mobility and fitness are imperative for the prevention of further pathological events (e.g., falls resulting in fracture, recurrent strokes, or cardiac events). Stroke is one of the top risk factors for incurring fractures as a result of a fall in older adults; Kanis et al. (23) analyzed 16.3 million hospitalizations due to fractures and reported a sevenfold hip fracture risk for individuals with stroke. In fact, the incidence of falls has been reported to be as high as 73% of individuals with stroke falling within 6 months after hospital discharge to home with an average of 3.4 falls per person during this 6-month time period (17). In addition, cardiovascular disease is the leading prospective cause of death in chronic stroke. Inactivity and low cardiovascular fitness, a major occurrence in persons with stroke, is one of the modifiable risk factors associated with cardiovascular disease.
In the past, intensive training in persons with stroke has been controversial due to the belief that strenuous activity would increase spasticity and reinforce abnormal movement (5). However, recent evaluation of intensive exercise programs has not found any evidence of an increase in spasticity (38).
Intensive treadmill protocols (29,33,36) are a recent addition to stroke rehabilitation and have resulted in improvements in gait and aerobic capacity; however, Smith et al. (37) found no significant improvements in reactive balance using an endurance treadmill protocol and suggested that functional or task-specific training may be needed to improve balance. Duncan et al. (13) also reported no significant improvements for the Berg Balance Score using a randomized controlled home-based individual exercise program (strengthening and walking program). Functional balance may be difficult to improve due to the varied tasks and movements under which balance is required. The one exception was a noncontrolled pilot study by Weiss et al. (42) that reported a 12% improvement in the Berg Balance Score for seven individuals with stroke by using a one-to-one high-intensity strengthening program. However, Kim et al. (25) recently undertook a double-blind randomized controlled trial of strength training in chronic stroke and found no carry-over into functional tasks and these authors emphasized the need for functional task-based practice.
Intensive rehabilitation programs for individuals with stroke have traditionally involved a one-to-one client-therapist ratio due to the close supervision required when challenging balance in these individuals, in addition to the necessary monitoring when taxing their cardiovascular function. However, given the current limited rehabilitation resources, it would be ideal to develop safe and effective community-based group exercise programs that are accessible to larger numbers of individuals. There is a clear and impressive void in the current literature that evaluates community-based group exercise programs for individuals with stroke, and only three studies have examined such programs. Rimmer et al. (36) undertook an intensive 12-wk community-based group training program (7 staff to 18 clients) that resulted in improvements in peak V̇O2, strength, and back flexibility but did not measure or train balance. A recent controlled pilot study that evaluated an 8-wk circuit training program found improvements in walking speed and 6-min walk distance, in addition to weight-bearing ability through the affected limb for the five experimental subjects (supervised by two physical therapists) compared with the four control subjects (11). Teixeira-Salmela et al. (38) found improvements in gait and stair climbing speed, in addition to muscle strength from a 10-wk muscle strengthening and physical conditioning program for 13 individuals with stroke. No studies to date have assessed the effect of a community-based group exercise program on both balance and functional capacity in individuals with stroke, and in addition, the retention of these effects has never been evaluated.
The purpose of this study was to evaluate a community-based group exercise intervention on both balance and functional capacity, two functions that are severely compromised in persons with stroke and can lead to devastating secondary complications We evaluated the effects of an 8-wk group exercise intervention on balance, walking ability, and functional capacity and the retention of these effects 1 month postintervention. Lastly, the psychosocial effects of exercise are infrequently evaluated in stroke, despite the well-documented high incidence of clinical depression in this population (4,9) and the knowledge that exercise can have substantial benefits to one’s well-being (10,24). Therefore, we also evaluated the effect of the exercise intervention on measures of health-related quality of life.
A single-group, repeated measures design was used for this study. Subjects underwent a baseline assessment (T1) followed by a second baseline assessment (T2) 4 wk later to determine the stability of the outcome values. After this second baseline assessment, subjects participated in an 8-wk intensive group exercise program focused on balance, strength, and functional capacity. After the 8-wk intervention, subjects underwent the first postintervention assessment (T3). A second postintervention assessment was undertaken 4 wk after the first postintervention test (T4) to determine retention effects.
Subjects were recruited from the community (e.g., advertisements placed in the local newspapers and community centers). Ethics approval was acquired from the local university and hospital review boards. In accordance with the university and hospital policy and to protect the rights of the subjects, informed written consent was received from all subjects before participating in this study. Inclusion criteria were: a minimum of 1 yr poststroke, present hemiparesis secondary to one cerebrovascular accident, positive answers on a modified Par-Q Test, study participation approval by their family physician who had been provided a letter that described the program and exclusion criteria, able to provide informed consent, follow one- and two-step commands, have an activity tolerance of 60 min with rest intervals, and able to walk 10 m with rest intervals with or without assistive devices. Exclusion criteria were: musculoskeletal or neurological conditions in addition to the stroke; uncontrolled hypertension, congestive heart failure, or atrial fibrillation; and scored less than 24 on the Folstein Mini-Mental Test (16).
For each of the four assessment times (two baselines and two postinterventions), the following outcomes were measured: 1) Berg Balance Test (2), 2) 12-Minute Walk Test distance (30), 3) self-paced gait speed, 4) fast-paced (as fast as you can walk safely) gait speed, 5) self-paced stair climbing speed, 6) fast-paced (as fast as you can climb safely) stair climbing speed, and 7) 11-item Reintegration to Normal Living Index (RNL Index). All testing was undertaken by evaluators who did not know the subjects and were not involved in the training intervention. For all measures and test sessions, both the evaluators and subjects did not have information regarding the subject’s evaluation from the previous test session.
The Berg Balance Test consists of 14 tasks that challenge balance while sitting, standing, or stepping (minimum score = 0 and maximum score = 56, with higher scores indicating better balance performance) and has been shown to be a valid and reliable measure and related to fall risk (2). The 12-Minute Walk Test (12MWT) is a reliable measure of functional capacity in persons with stroke (14) as defined by the extent in which a person can increase exercise intensities and maintain those increased levels (8). Gait speed has been shown to be a reliable measure and related to the stage of recovery, lower-extremity strength, and clinical tests of motor function in stroke (35). Gait speed was calculated from the mean of three trials using the distance and time measured by infrared emitting diodes (Optotrak, Northern Digital) attached to the lateral malleoli and toe for the middle 4-m section (i.e., representative window of constant gait speed) of a 10-m walkway. The mean of two trials for the time to climb up four stairs was assessed and has been shown to be reliable in healthy adults (32).
The RNL Index is a self-report measure that has been tested in stroke survivors that asks individuals to rate their satisfaction on 11 items regarding their physical lives (e.g., “I move around my community as I feel is necessary”), emotional lives (e.g., “I feel that I can deal with life events as they happen”), and social lives (e.g., “I participate in social activities with my family, friends, and/or business acquaintances as is necessary or desirable to me”) (45). The score is normalized to 100 with a greater score representing a higher level of satisfaction. Excellent internal consistency (Cronbach’s alpha > 0.90) has been reported for the RNL for 80 individuals whose health problems included heart disease, neurological conditions, musculoskeletal conditions, and tumors (45). In addition, the RNL has been shown to be responsive to changes in the clinical status of patients with stroke (3).
The Canadian Occupational Performance Measure (COPM) was used only at the second baseline assessment and the first postintervention assessment. The COPM is a valid and reliable individualized, outcome instrument designed to detect change in self-perception of occupational performance over time (27). Subjects are asked to identify their own problems in the areas of self-care (e.g., difficulty bathing), productivity (e.g., no longer able to work outside the home), and leisure (e.g., difficulty reading). Subjects then rate their 1) perceived performance and 2) perceived satisfaction on a 10-point scale for each of the five most important problems. Ratings for the five problems are summed and divided by five for a Performance Score and a Satisfaction Score (each with a minimum score of 1 and maximum score of 10 with higher scores indicating better perceived performance or higher levels of satisfaction). As there was only one baseline measure of the COPM, the stability of this measure before the intervention was not assessed in this study. However, others have reported good test-retest reliability (intraclass correlation coefficients > 0.80) for this particular outcome measure in individuals in outpatient rehabilitation settings (28).
In this study, we used only equipment generally available in community centers (e.g., chairs, ankle weights, and stackable risers/steppers). The only safety modification was a nonslip pad to the bottom of each chair leg. The functional exercise program was an 8-wk, 1-h program three times per week and comprised light aerobic warm-up (e.g., marching on the spot, arm swinging), stretching (e.g., calf stretches, hamstring stretches), functional lower-extremity strengthening (e.g., repetitive rise from a low chair, repetitive rise on toes), functional lower-extremity strengthening with light weights (e.g., marching with 0.5-kg ankle weights to improve hip flexor muscles), balance and weight-shifting (e.g., lunges in different directions), aerobic stepping with steppers, a walking circuit (Fig. 1), and cool-down/stretching. Exercises in this program were designed with grade increments to meet the capabilities of each subject. In addition, the exercises were generally similar to tasks that were familiar to the subjects (e.g., rising from a chair) and a part of everyday mobility. Three classes were undertaken each day with 9–10 subjects in each class and three trainers (one physical therapist and two kinesiologists) per class.
Subjects were familiarized with the 16-point Borg Rating of Perceived Exertion (RPE) (7) and the rating scale was visible from all areas of the room. Subjects were asked throughout the exercise program to determine their RPE. The trainers encouraged subjects to work at a level between 11 and 13 (fairly light to somewhat hard) and to ensure that they did not exceed a level 15 (hard). Subjects took home a feedback sheet at the end of each class that asked subjects to indicate whether muscle soreness and general fatigue interfered with any of the subject’s daily activities after the exercise session. They returned it the following session so that the trainers could ensure that subjects were not overly sore or fatigued from the previous session.
A repeated measures ANOVA was used to compare the outcome measures at the four assessment times (baseline 1, baseline 2, postintervention 1, postintervention 2) and was followed by Tukey’s post hoc test when a significant time effect was found (SPSS 9.0, SPSS Inc.). This ANOVA was performed for each of the outcome measures: 1) Berg Balance Test, 2) 12-Minute Walk Test distance, 3) self-paced gait speed, 4) fast-paced gait speed, 5) self-paced stair speed, 6) fast-paced stair speed, and 7) RNL Index. As the COPM was only undertaken at two time points, paired t-tests were undertaken to assess the change in the COPM Performance Score, and COPM Satisfaction Score. Correlational analyses were undertaken to relate the change in the outcome measures with the baseline values to determine whether the baseline level of function related to improvement. A significant level of P < 0.05 (two-tailed) was selected for all statistical tests.
Subject characteristics and participation.
A total of 25 subjects completed the training program and postintervention tests (see Table 1 for subject characteristics) with subjects attending a mean 94% of the classes. Five subjects completed the first preassessment but did not undertake the training program for the following reasons: three did not feel that the program would be beneficial for them, and two had changes to their schedule and left the country on holiday. One subject completed 3 wk of the training program and then underwent surgery and did not return. Comorbidities included hypertension (15 subjects), obesity with body mass index greater than 30 (five subjects), diabetes (three subjects), hyperlipidemia (five subjects), clinical depression (four subjects), and sedentary lifestyle (exercise < three times per week) (25 subjects).
Given the lack of significant difference between the two baseline measures (see section below: “Effect of test time on the outcome measures”), the T2 measures are presented here to provide an indication of baseline values. There was a large range in motor function with baseline measures of self-paced gait speed, 12MWT, and Berg Balance Test ranging between 0.28 and 1.14 m·s−1, 148 and 691 m, and 31 and 51 points, respectively. Mean self-paced gait speed was less than 50% of healthy older adult values (44), and the 12MWT distances were less than 60% of values reported in chronic respiratory subjects (30). On the other hand, baseline RNL Scores were relatively high (all mean scores were above 80 out of a possible 100). For the COPM, subjects most frequently identified problems involving functional mobility (e.g., walking tolerance, transfer ability, stairs) and personal care (e.g., shaving, dressing).
Effect of test time on the outcome measures.
There was a significant test time effect (P < 0.05) for the following six variables 1) Berg Balance Test, 2) 12MWT distance, 3) self-paced gait speed, 4) fast-paced gait speed, 5) self-paced stair speed, and 6) fast-paced stair speed (Table 2). Tukey’s post hoc analyses on these six variables found no difference between T1 (baseline 1) and T2 (baseline 2) demonstrating that the baseline measures were stable over 1 month. For all six of these variables, there was a significant difference between T2 (baseline 2) and T3 (postintervention 1), demonstrating a training effect for these variables. Lastly, there was no difference between T3 (postintervention 1) and T4 (postintervention 2) demonstrating a retention effect for these six measures. The RNL Index stayed within one point between T1 (baseline 1) and T2 (baseline 2) and then increased 5 points between T2 (baseline 2) and T3 (postintervention 1); however, this increase was not significant. There was a significant intervention effect for the COPM for both the Performance and Satisfaction Score.
Change in outcome measures and relation to baseline measures.
Those with the poorer balance demonstrated the greatest percent improvement with the Berg Balance Scale (r = −0.72, P < 0.001). In fact, those below the median Berg Score had a mean change of 4 points, whereas those above only changed 1 point. Gait and stair climbing improvement ranged from 5 to 15%. Significant correlations were found (r = −0.4 to −0.6, P < 0.05) between the walking measures and percent improvement, suggesting that those with lower baseline function improved the most. Lastly, strong correlations were found (r = −0.76, P < 0.001) between the baseline RNL and the improvement in RNL, suggesting that those with poorer life satisfaction improved the most in this domain.
The program accommodated community-dwelling individuals with a wide range of function with gait speed as slow as 0.28 m·s−1 and a Berg Score as low as 31. In addition, subjects with a number of comorbidities were able to tolerate the program.
Adherence to the program was excellent and may have been facilitated by the collegiality that developed among participants and the tremendous encouragement and support they provided to one another. Others have reported similar observations regarding the social benefits of a group community exercise program in stroke (36,38).
There is some evidence that has shown that perceptions of excessive exertional demands or the lack of confidence in the ability to perform the exercise are barriers to sustained physical activity (26). One subject commented that he really enjoyed the first 4 wk of the program but that he did not enjoy the latter 4 wk of the program as much due to the increasing exercise intensity. A moderate exercise intensity such as was followed in this study might help to promote life-long exercise adherence, as opposed to more challenging programs.
The trainers felt that an increase beyond the 3:1 client-staff ratio could only be safely undertaken if there was more focus on strengthening and less on the challenging agility and balance exercises which required close supervision. Alternatively, a reduction in the number of subjects with low motor function would maintain the safety of the program and reduce the need for so many staff. Four subjects who required close supervision had either the lowest Berg Balance Scores (30–32) and/or the slowest gait speed (0.27–0.28 m·s−1). Thus, one might consider each subject’s balance or gait performance to distribute the number of subjects with low function across groups, or to increase the staff-client ratio for groups with lower function.
Physical outcome measures.
In our intervention, the training (e.g., strengthening, balance) was undertaken within the context of everyday functional tasks. We believe that such an exercise program was not only meaningful to the subjects, but the task-specificity of the training contributed to the carry-over to those activities that are important for mobility. Despite the “chronic” nature of the participants’ injury, improvements were found for all physical measures in this group of subjects. As some subjects were as much as 12 yr poststroke, this study demonstrates that improvements in balance and functional capacity can be acquired through practice well after the initial stroke episode. A limitation of the 12MWT outcome measure should be noted. The 12MWT distance reflects the ability to undertake everyday activities and may be a good indicator of functional capacity required for normal daily activities; however, it is not necessarily a measure of cardiovascular function (14,15). The 12MWT is influenced by cardiorespiratory and cardiovascular fitness, in addition to factors such as motivation, neuromuscular function, and peripheral muscle strength (15).
Importantly, the improvements in the physical outcome measures were retained at least 1 month after the cessation of the program. The greater percent improvement for individuals with lower function is an important observation as these individuals are at a higher risk for falls and cardiac disease. In addition, even small changes for these very impaired individuals have the potential to result in clinically significant effects. Accessible evidence-based community-based exercise programs are urgently needed to encourage persons to engage in regular physical activity poststroke. Recently, the World Health Organization issued a warning that sedentary lifestyle is one of the more serious yet insufficiently addressed public health problems of our time and attributes 2 million deaths per year to physical inactivity.
Adherence to exercise programs has the potential to improve activity tolerance, reduce the risk of falls and cardiac events, and maintain or improve functional mobility and physical independence.
A limitation of the study design was the absence of an independent control group. However, the baseline measures were found to be stable over a full 1-month period, and the intervention produced a consistent response for six different physical function measures (i.e., no difference between the two baseline measures, a significant improvement from the intervention and a plateau after the intervention indicating retention of these abilities). The Berg and 12MWT appeared to have some learning or familiarization effect after the first test session; however, these baseline changes were not significant, and in addition, the changes from the intervention were up and above any changes observed during the two baseline measures.
Could the improvements in motor function be simply a result of time or functional recovery? Motor recovery typically shows the greatest improvements between 3 and 6 wk postinjury, and a clear plateau is reached by 90 d (12,19). Our subjects were a minimum of 1 yr poststroke and were therefore at the “chronic” stage, and one would expect little changes with standard rehabilitation efforts at this point in time. In fact, a post hoc analyses of our data found no relationship between improvements in function and length of injury (e.g., correlation of Berg improvement and length of injury was r = 0.008, P = 0.97). Our study represents one of the few studies that has demonstrated the ability to improve balance function in chronic stroke. Only one other study in the literature has demonstrated balance improvements (42), whereas others found no change even with intensive treatments like aerobic treadmill training or muscle conditioning (13,37). A number of reasons may attribute to the difficulties in improving balance function in a chronic stroke population: 1) previous interventions may have addressed specific structures that relate to balance (e.g., muscle strength); however, the interventions may not have sufficiently challenged balance in the context of the large variety of tasks and movements under which balance is required; 2) impaired balance is predominantly a result of the nature and severity of the neurological lesion and may not be improved to a large extent; and 3) there is an underlying decline in mobility with time given the older age of this group and perhaps even maintenance (but not necessarily improvement) in balance function is an accomplishment. In fact, Wade et al. (40) reported deterioration in mobility over a 3-month period in individuals with chronic stroke.
Psychosocial outcome measures.
The RNL Index is a measure of satisfaction with physical and social performance, including coping skills and comfort in personal relationships (45). The absence of an intervention effect found with this measure may be a consequence of the relatively high RNL Scores found in this group, suggesting that several of these community-living individuals had adapted to their situation and were relatively satisfied. In fact, four subjects had complete satisfaction with their lives (score of 100) at the baseline assessment. In this subject group, the mean time since the onset of stroke was 4 yr, and it is reasonable to expect that some subjects have adapted psychologically and physically to their impairments. Similar results were reported by Bastien and her colleagues (1) among community-dwelling individuals with stroke. A high level of emotional well-being was also recently noted by Hackett et al. (21), who reported that some individuals with chronic stroke appear to adjust psychologically to their illness despite significant and ongoing physical disability. On the other hand, the strong relationship between the baseline RNL Index and the improvements in the RNL Index suggests that exercise had the greatest effects on health-related quality of life for those individuals who commenced the program with low satisfaction of their physical, social, and emotional lives.
In addition, the exercise intervention did have an impact upon perceptions of performance and satisfaction as measured by the COPM. The COPM is an individualized outcome measure that detects change in the performance of activities and satisfaction of that performance that are relevant and important to the person. It was not surprising that functional mobility was one of the most common identified problems as improved walking has been previously reported to be the most frequently stated goal by individuals with stroke (6). Thus, the noted improvements in gait performance may have been particularly important to these subjects.
The results of this study suggest that the effect of an exercise program that focuses upon and improves strength, endurance, and balance is meaningful to the subjects and also has a demonstrable impact upon the performance and satisfaction of this performance relevant to daily living activities. This relationship between physical activity and improved health-related quality of life is in agreement with literature examining other chronic diseases (34). The effects on health-related quality of life measures (e.g., RNL, COPM) can be considered as important as the effects on physical function. There is a growing recognition that health-related quality of life can be an outcome in itself and this sentiment is reflected by the Healthy People 2010 Report (national health objectives for the United States), whose top priority now includes improving quality of life, in addition to increasing life expectancy (39). Recently, the World Health Organization (43) also provided greater focus on “qualify of life” with its revised International Classification of Functioning, Disability and Health (ICF), which emphasizes how people live with their health conditions and the achievement of a productive, fulfilling life. Multiple dimensions exist for the term “health-related quality of life” (34), and the measures used in this study focused on the satisfaction of physical function, emotional lives, and social lives and are representative of the “participation” dimension of the ICF (43).
The improvements found in this study for the health-related quality of life measures have important implications for potentially countering the high incidence of clinical depression documented in people with stroke (4,9). In addition, a perception of improved well-being can only augment adherence to community-based exercise programs
Community-based programs for stroke survivors.
This is an urgent need to define safe and effective community-based exercise programs to enhance mobility and fitness of individuals with chronic stroke who generally are older, less fit, and are prone to falls. Thus, we designed a group exercise program that could be easily implemented into the community as it did not require costly one-on-one training ratios, specialized settings, and expensive equipment. The challenge was to achieve a level of exercise intensity that would result in positive effects yet ensure safety to avoid an exercise-related fall or cardiovascular event. Rimmer et al. (36) used a screening symptom-limited graded exercise test to obtain peak V̇O2 measures to grade the cardiovascular component of the exercise program. This screening stress test is advantageous in identifying those individuals who may have a negative response to exercise and can also determine an individualized exercise intensity that can be safely undertaken with the client. However, a screening stress test may also eliminate a large number of potential clients; the stress test is not an assessment that is readily available to the community, and lower functioning candidates may be limited in part by peripheral factors in addition to central fatigue.
In this study, screening was undertaken using a modified Par-Q Test, in addition to confirmation that the specified exclusion criteria were met. However, the lack of a symptom-limited exercise test prohibited a high-intensity cardiovascular component. Instead, a moderately high exercise intensity was undertaken and monitored with the use of ratings of perceived exertion. The results of this study showed that a short-term community-based exercise intervention can improve mobility, functional capacity, and balance and retain these improvements. In addition, an exercise intervention can have a demonstrable impact upon the performance of activities and abilities that are considered meaningful to the participants and may in turn promote exercise adherence. Implementation of such community-based programs has potential for improving activity tolerance and reducing the risk of cardiac events and falls resulting in fractures, which are secondary complications which occur frequently in stroke.
Financial support: Grant in aid from the Heart and Stroke Foundation of BC and Yukon and salary support award to J. Eng from the Canadian Institute of Health Research and the Michael Smith Foundation for Health Research.
1. Bastien, M., N. Koerner-Bitensky, S. Lalonde, N. Lebrun, and D. Matte. A health and leisure program for community dwelling individuals with stroke: a pilot study. Can. J. Rehabil. 12: 7–14, 1998.
2. Berg, K., S. Wood-Dauphinee, J. I. William, and D. Gayton. Measuring balance in the elderly: preliminary development of an instrument. Physiother Can. 41: 304–311, 1989.
3. Bethoux, F., P. Calmels, and V. Gautheron. Changes in the quality of life of hemiplegic stroke patients with time. Am. J. Phys. Med. Rehabil. 78: 19–23, 1999.
4. Black-Schaffer, R. M., A. E. Kirsteins, and R. L. Harvey. Stroke rehabilitation. 2. Co-morbidities and complications. Arch Phys. Med. Rehabil. 80: S8–S16, 1999.
5. Bobath, B. Adult Hemiplegia: Evaluation and Treatment. London: Heinemann Medical Books, 1978, pp. 136.
6. Bohannon, R. W., A. W. Andrews, and M. B. Smith. Rehabilitation goals of patients with hemiplegia. Int. J. Rehabil. Res. 11: 181–183, 1988.
7. Borg, G. Perceived exertion as an indicator of somatic stress. Scand. J. Rehabil. Med. 2: 92–98, 1970.
8. Brooks, G. A., T. D. Fahey, and T. P. White. Exercise Physiology: Human Bioenergetics and Its Applications. Mountain View, CA: Mayfield Publishing, 1996, pp. 565.
9. Carod-Artal, J., J. A. Egido, J. L. Gonzalaz, and E. V. Seijas. Quality of life among stroke survivors evaluated 1 year after stroke. Stroke 31: 2995–3000, 2000.
10. Chen, J., and W. J. Miller. Health effects of physical activity. Health Rep. 11: 21–30, 1999.
11. Dean, C. M., C. L. Richards, and F. Malouin. Task-related circuit training improves performance of locomotor tasks in chronic stroke: a randomized, controlled pilot trial. Arch Phys. Med. Rehabil. 81: 409–417, 2000.
12. Duncan, P. W., L. B. Goldstein, R. D. Horner, P. B. Landsman, G. P. Samsa, and D. B. Matchar. Similar motor recovery of upper and lower extremities after stroke. Stroke 25: 1181–1188, 1994.
13. Duncan, P., C. L. Richards, D. Wallace, et al. A randomized, controlled pilot study of a home-based exercise program for individuals with mild and moderate stroke. Stroke 29: 2055–2060, 1998.
14. Eng, J. J., K. S. Chu, A. S. Dawson, C. M. Kim, and K. Hepburn. Functional walk tests in individuals with stroke: Relationship to perceived exertion and myocardial exertion. Stroke 33: 756–761, 2002.
15. Fitts, S. S., and M. R. Guthrie. Six-minute walk by people with chronic renal failure: assessment of effort by perceived exertion. Am. J. Phys. Med. Rehabil. 80: 837–858, 1995.
16. Folstein, M. F., S. E. Folstein, and P. R. Mchugh. “Mini-mental state”: a practical method for grading the cognitive state of patients for the clinician. J. Psychiatr. Res. 12: 189–198, 1975.
17. Forster, A., and J. Young. Incidence and consequences of falls due to stroke: a systematic inquiry. Br. Med. J. 311: 83–86, 1995.
18. Fried, L. P., W. H. Ettinger, B. Lind, A. B. Newman, and J. Gardin. Functional disability in older adults: a physiological approach: cardiovascular health study research group. J. Clin. Epidemiol. 47: 747–760, 1994.
19. Gray, C. S., J. M. French, D. Bates, N. E. Cartlidge, O. F. James, and G. Venables. Motor recovery following acute stroke. Age Ageing 19: 179–184, 1990.
20. Gresham, G. E., T. E. Fiztpatrick, P. A. Wolf, P. M. Mcnamara, W. B. Kannel, and T. R. Dawber. Residual disability in survivors of stroke: the Framingham Study. N. Engl. J. Med. 293: 954–956, 1975.
21. Hackett, M. L., J. R. Duncan, C. S. Anderson, J. B. Broad, and R. Bonita. Health-related quality of life among long-term survivors of stroke: results from the Auckland Stroke Study, 1991–1992. Stroke 31: 440–447, 2000.
22. Heart and Stroke Foundation of Canada. Heart Disease and Stroke in BC for 2000. Vancouver, BC: Heart and Stroke Foundation of Canada and Ministry of Health Canada, 2000, p. 51.
23. Kanis, J., A. Oden, and O. Johnell. Acute and long-term increase in fracture risk after hospitalization for stroke. Stroke 32: 702–706, 2001.
24. Kaplan, M. S., J. T. Newson, B. H. Mcfarland, and L. Lu. Demographic and psychosocial correlates of physical activity in late life. Am. J. Prev. Med. 21: 306–312, 2001.
25. Kim, C. M., J. J. Eng, D. L. Macintyre, and A. S. Dawson. Effects of isokinetic strength training on walking in persons with stroke: a double-blind controlled pilot study. J. Stroke Cerebrovasc. Dis. 10: 265–273, 2001.
26. King, A., S. Blair, D. Bild, et al. Determinants of physical activity and interventions in adults. Med. Sci. Sports Exerc. 24: S221–S236, 1992.
27. Law, M., H. Polatajko, N. Pollock, M. A. Mccoll, A. Carswell, and S. Baptiste. Pilot testing of the Canadian Occupational Performance Measure: clinical and measurement issues. Can. J. Occup. Ther. 61: 191–197, 1994.
28. Law, M., S. Baptiste, A. Carswell, M. A. Mccoll, H. Polatajko, and N. Pollock. Canadian Occupational Performance Measure. Ottawa, ON: CAOT Publications, 1998, p. 26.
29. Macko, R. F., C. A. Desouza, L. D. Tretter, et al. Treadmill aerobic exercise training reduces the energy expenditure and cardiovascular demands of hemiparetic gait in chronic stroke patients. Stroke 28: 326–330, 1997.
30. Mcgavin, C. R., M. Artvinli, H. Naoe, and G. J. McHardy. Dyspnoea, disability and distance walked: a comparison of estimates of exercise performance in respiratory disease. Br. Med. J. 22: 241–243, 1978.
31. Mol, V., and D. Baker. Activity intolerance in the geriatric stroke patient. Rehabil. Nurs. 16: 337–343, 1991.
32. Olney, S. J., N. D. Elkin, P. J. Lowe, and D. O. Symington. An ambulation profile for clinical gait evaluation. Physiother. Can. 31: 85–90, 1979.
33. Potempa, K., M. Lopez, L. T. Braun, J. P. Szidon, L. Fogg, and T. Tincknell. Physiological outcomes of aerobic exercise training in hemiparetic stroke patients. Stroke 26: 101–105, 1995.
34. Rejeski, W. J., L. R. Brawley, and S. A. Shumaker. Physical activity and health-related quality of life. Exerc. Sport Sci. Rev. 24: 71–108, 1996.
35. Richards, C. L., F. Malouin, F. Dumas, and D. Tardiff. Gait velocity as an outcome measure of locomotor recovery after stroke. In: Gait Analysis: Theory and Application, R. L. Craik and C. A. Oatis (Eds.). St. Louis: Mosby, 1995, pp. 355–364.
36. Rimmer, J. H., B. Riley, T. Creviston, and T. Nicola. Exercise training in a predominantly African-American group of stroke survivors. Med. Sci. Sports Exerc. 32: 1990–1996, 2000.
37. Smith, G. V., L. W. Forrester, K. H. C. Silver, and R. F. Macko. Effects of treadmill training on translational balance perturbation responses in chronic hemiparetic stroke patients. J. Stroke Cerebrovasc. Dis. 9: 238–245, 2000.
38. Texeira-Salmela, L. F., S. J. Olney, S. Nadeau, and B. Brouwer. Muscle strengthening and physical conditioning to reduce impairment and disability in chronic stroke survivors. Arch. Phys. Med. Rehabil. 80: 1211–1218, 1999.
39. U.S. Department of Health and Human Services. Healthy People 2010: With Understanding and Improving Health and Objectives for Improving Health. Washington, DC: U.S. Government Printing Office, 2000, p. 2.
40. Wade, D. T., F. M. Collen, G. F. Robb, and C. P. Warlow. Physiotherapy intervention late after stroke and mobility. Br. Med. J. 304: 609–613, 1992.
41. Wade, D. T., and R. L. Hewer. Functional abilities after stroke: measurement, natural history and prognosis. J. Neurol. Neurosurg. Psychiatry 50: 177–182, 1987.
42. Weiss, A., T. Suzuki, J. Bean, and R. A. Fielding. High intensity strength training improves strength and functional performance one year after stroke. Am. J. Phys. Med. Rehabil. 79: 369–376, 2000.
43. World Health Organization. International Classification of Functioning, Disability and Health: ICF.
Geneva: World Health Organization, 2001, pp. 5, 18–20.
44. Winter, D. A., A. E. Patla, J. S. Frank, and S. E. Walt. Biomechanical walking pattern changes in the fit and healthy elderly. Phys. Ther. 70: 340–347, 1990.
45. Wood-Dauphinee, S. L., A. Opzoomer, J. I. Williams, B. Marchand, and W. O. Spitzer. Assessment of global function: the Reintegration to Normal Living Index. Arch. Phys. Med. Rehabil. 69: 583–590, 1988.