Archives of Physical Medicine and Rehabilitation
Volume 87, Issue 7 , Pages 967-973, July 2006

The Effect of Environment and Task on Gait Parameters After Stroke: A Randomized Comparison of Measurement Conditions

  • Susan E. Lord, MSc

      Affiliations

    • Department of Medicine (Rehabilitation), Wellington School of Medicine and Health Sciences, University of Otago, Wellington South, New Zealand
    • Corresponding Author InformationReprint requests to Susan E. Lord, MSc, Dept of Medicine (Rehabilitation), Wellington School of Medicine and Health Sciences, University of Otago, PO Box 7343, Wellington South, New Zealand
    • ,
  • Lynn Rochester, PhD

      Affiliations

    • School of Health, Community and Education Studies, Northumbria University, Newcastle upon Tyne, UK
    • ,
  • Mark Weatherall, FRACP

      Affiliations

    • Department of Medicine (Rehabilitation), Wellington School of Medicine and Health Sciences, University of Otago, Wellington South, New Zealand
    • ,
  • Kathryn M. McPherson, PhD

      Affiliations

    • Division of Rehabilitation and Occupation Studies, Auckland University of Technology, Auckland, New Zealand
    • ,
  • Harry K. McNaughton, PhD

      Affiliations

    • Medical Research Institute of New Zealand, Wellington, New Zealand.

Article Outline

Abstract 

Lord SE, Rochester L, Weatherall M, McPherson KM, McNaughton HK. The effect of environment and task on gait parameters after stroke: a randomized comparison of measurement conditions.

Objectives

To assess the effect of environment and a secondary task on gait parameters in community ambulant stroke survivors and to assess the contribution of clinical symptoms to gait performance.

Design

A 2×3 randomized factorial design with 2 main factors: task (no task, motor task, cognitive task) and environment (clinic, suburban street, shopping mall).

Setting

Subjects were assessed in 1 of 3 settings: 2 in the community (a suburban street and shopping mall) and 1 clinical environment.

Participants

Twenty-seven people with stroke (mean age, 61±11.6y; mean time since stroke onset, 45.8±34.2mo), living at home, were recruited from community stroke groups and from a local rehabilitation unit. Selection criteria included the following: ability to give informed consent, unilateral first ever or recurrent stroke at least 6 months previously, walking independently in the community, a gait speed between 24 and 50m/min, Mini-Mental State Examination score of 24 or higher, and no severe comorbidity.

Interventions

Not applicable.

Main Outcome Measures

Gait speed (in m/min), cadence, and step length were assessed by using an accelerometer with adjustable thresholds. Clinical measures hypothesized to influence gait parameters in community environments were also assessed including fatigue, anxiety and depression, and attentional deficit.

Results

Twenty-seven people with a mean baseline gait speed of 42.2±5.9m/min were randomly allocated to 1 of 9 conditions in which the setting and distraction were manipulated. Analysis of variance showed a significant main effect for environment (P=.046) but not for task (P=.37). The interaction between task and environment was not significant (P=.73). Adjusting for baseline gait speed, people walked on average 8.8m/min faster in the clinic (95% confidence interval, 0.3−17.3m/min) than in the mall. Scores for fatigue, anxiety and depression, and attentional deficit were higher than normative values but did not influence gait performance.

Conclusions

This study suggests that people with chronic stroke cope well with the challenges of varied environments and can maintain their gait speed while performing a secondary task. Despite moderate levels of gait impairment, gait automaticity may be restored over time to a functional level.

Key Words:  Cerebrovascular accident , Gait , Rehabilitation , Task performance and analysis

 

REINTEGRATION INTO COMMUNITY life contributes to improved quality of life after stroke and attainment of community mobility is an important part of that process. However, full, independent community mobility is not always achievable and instead variable levels of community mobility are attained.1, 2, 3 Stroke survivors have reported dissatisfaction with their ability to ambulate outdoors and to access their communities.4

The skills required to achieve community mobility include the ability to walk at a given speed for a minimum requisite distance.5 The role of attention in gait has been the focus of recent research, with evidence that cortical input demands vary with the difficulty of the task, the environment in which it is performed, and the type of task being performed. By using a dual-task paradigm, decrements in performance have been observed for both healthy and balance-impaired older adults and for people with specific types of neurologic disorder.6, 7 For example, decrements in gait performance and postural control have been identified when participants have carried out a secondary task such as talking,8, 9 carrying a tray or tumbler,10, 11 stepping over virtual and real obstacles,12, 13 responding verbally to auditory tones,14, 15 and performing cognitive tasks.16

A number of studies have reported changes in gait performance in people with stroke during dual-task testing in a clinical environment. Bowen et al17 noted a significant decrease in gait speed (4m/min) and a significant increase in double-support time when a cognitive activity was added to a walking test for 11 people with stroke. Haggard and Cockburn18 reported a 7% decrement in stride duration and a simultaneous 4% decrement in cognitive task performance under dual-task conditions for 50 people undergoing neurologic rehabilitation that included 11 people with stroke. Gait decrement improved in 7 of 10 people with stroke who were reassessed 1 to 9 months later, and cognitive decrement improved in 3 people.19 Changes in gait performance under motor and cognitive dual-task conditions may be explained by a loss of central capacity to perform more than 1 task simultaneously, which occurs either because gait is less automatic and more reliant on central cognitive processing and use of attentional resources,18, 20 or because cognitive acuity is diminished, both of which result in “cognitive-motor interference.”19 Research21, 22 suggests that people who have had a stroke experience attention deficits, even late after the stroke event, and the importance of sustained attention in particular to motor recovery and motor performance has been reported. However, the research to date has been conducted in laboratory or clinic and hospital settings, and the real-world relevance of the findings for community mobility can only be estimated. The changes in gait speed and stride duration noted previously do not exceed the bounds of measurement error for people with stroke,23 and the findings to date do not necessarily reflect a meaningful decline in functional performance.

This study set out to investigate the effect of environment and a secondary task on gait parameters for people with stroke to gain a better understanding of the skills required for effective community mobility. We hypothesized that study participants would experience greater difficulty walking in unpredictable, community environments or when performing a secondary task compared with walking in predictable clinical environments without these obstructions.

Back to Article Outline

Methods 

Participants 

Twenty-seven people with chronic stroke were studied. Participants who met the following criteria were recruited: a first-ever or recurrent stroke (World Health Organization Monitoring Trends and Determinants in Cardiovascular Disease [MONICA] definition)24 at least 6 months previously, home-dwelling, walking independently outdoors without the need for close supervision; baseline 10-m timed walks between 24 and 50m/min, Mini-Mental State Examination score of 24 or higher,25 and no severe comorbidity. The rationale for a restriction in gait speed was based on earlier work that categorized levels of community mobility according to gait speed.1 In addition, we wished to engage participants who would find the testing procedure challenging rather than select those whose gait speed approached referent values. The participants came from a convenience sample of people who were recruited through community stroke groups, from newspaper and local advertising, and from physical therapists at 1 regional hospital. All participants gave written consent to enter the study, which was approved by the relevant regional ethics committee. Figure 1 describes the flow of participants through the trial.

Study Design 

The study was a randomized factorial design, with 2 factors, task and environment, each at 3 levels. The 3 levels of task were no task, cognitive task, and motor task, and the 3 levels of environment were clinic, suburban street, and shopping mall. Participants were randomized to 1 of the different combinations of the 3 levels of each main experimental factor, task and environment (see fig 1). A factorial design was chosen for this exploratory study because we wanted to investigate the interaction effect between environment and task, which is not possible with a repeated-measures design. For example, a secondary task may be achieved in a predictable environment but not in a more challenging environment Also, there may have been a learning effect associated with the planned secondary cognitive and motor tasks even with randomization of testing order.

Experimental Protocol 

Participants were screened in their homes and invited to participate in the study if they met the inclusion criteria. Baseline testing on standardized measures was performed in the clinic environment, and participants were randomly assigned to 1 of 9 testing conditions. Two researchers were required to implement the protocol. Each subject was then asked to perform a 6-minute walk test (6MWT) for the condition in which they were randomized. For both motor and cognitive tasks, the researcher used a stop watch to ensure accurate timing of tasks and also made a note of any inaccurate responses during the cognitive task. To measure distance walked during 6 minutes, the researcher walked several meters behind the participant wheeling an odomoter (a measuring wheel). The researcher requested the participant to walk at his/her comfortable pace and to concentrate on both walking and the secondary task rather than focus on one. All tests were performed at a similar time of the day, between 10:00 am and 3:00 pm.

Testing Conditions 

For the single-task condition, participants were asked to walk at a comfortable walking pace for 6 minutes without talking in 1 of 3 selected environments: (1) the clinic environment was a quiet, wide hospital corridor, which necessitated a turn after approximately 150m; (2) the suburban street had a footpath with a slight camber on it, there was a small incline in the street and participants were asked to cross the street after walking on grass for 15m as well as negotiate a curb up and down; and (3) the shopping mall was a large, busy city mall with good ambient conditions, wide walkways, and moderate shopping crowds. The secondary tasks included the following. One was a motor task in which participants were asked to step over a wooden block that was placed approximately 2m in front of the participant on the anticipated route at 30-second intervals. The block measured 45cm long, 6.5cm high, and 6.5cm wide and had a nonstick cover wrapped around the central 30cm to ensure good contact with the ground. The second one was a cognitive task in which participants were asked to respond to numbers spoken by the researcher every 10 seconds by stating “yes” if the number was even and “no” for odd numbers. This test was chosen because of the ease of standardizing it for a nonclinical or laboratory environment. Although the goal of the study was not to measure change in cognitive performance (attention), scores on the cognitive task were recorded to show the accuracy with which the cognitive task was performed. A yes-no response test requires continuous attention over the 6-minute period, and the appropriateness of response was easy to verify. Although a wide range of cognitive tasks has been used in dual-task studies, none have been developed with the specific aim of testing relevant to the community excursions undertaken in this study. Although the task was not strictly functional, it imposed a constant cognitive load as may occur when attending to a conversation and responding to questions.

Equipment 

Cadence and step length were measured by using the StepWatch Activity Monitor (SAM),a a stride counter that uses a microprocessor-linked custom accelerometer with adjustable thresholds for motion and cadence parameters. Data were then downloaded through a docking station and stored. Reliability and validity of the SAM has been established for people with stroke by using 1-minute and 6-minute walks inside,26 and reliability was further established before this study by testing 10 other people with stroke on 2 occasions a week apart in 2 outside environments: a suburban street and a supermarket. The test-retest reliability for this equipment in both environments was tested and found to be excellent, with intraclass correlation coefficients of r equal to .98.

Outcome Measures 

The primary outcome measure was gait speed (measured in m/min) obtained during the 6MWT for all experimental conditions. The 6MWT was selected because it measures endurance, which is a prerequisite for successful community ambulation,20 rather than the 10-m walk test (10MWT), which measures speed over a short distance and has been shown to overestimate walking27 capacity. Secondary measures were cadence and step length, which were included so that we could investigate more closely the gait patterns that emerged.

Baseline Measures 

We hypothesized that 4 key clinical symptoms of stroke, balance, attention, fatigue, and anxiety and depression, may be associated with gait performance. Balance was measured by using the Berg Balance Scale (BBS).28 Attention was measured by using 3 subtests of the Test of Everyday Attention (TEA): (1) elevator counting, which has been used previously to measure sustained attention in stroke29; (2) telephone search while counting−dual-task decrement also for sustained attention, which has been shown to predict functional capacity late after stroke22; and (3) visual elevator (number correct), which loads onto attentional switching, has been tested before in chronic stroke by using response time tasks,21 and which may be relevant to community mobility. Fatigue was measured by using 2 of the 5 fatigue domains from the Multidimensional Fatigue Inventory (MFI)30 that we hypothesized were most likely to be affected; the physical domain and the reduced activity domain and depression and anxiety were measured by using the Hospital Anxiety and Depression Scale (HADS).31 Baseline gait speed, which was measured by using the 10MWT at comfortable pace32 was also included as a possible covariate.

Data Analysis 

Univariate statistics were used to describe the data. The prespecified analysis was an analysis of covariance (ANCOVA) to examine the relation between gait speed (with baseline gait speed as a covariate) and environment and task and their interaction. Analysis of variance (ANOVA) was used to examine the relation between the secondary measures: step frequency and step length and environment and task. The Tukey-Kramer multiple comparison procedure was used for multiple comparisons between the different levels of environment and task for gait speed. Normality assumptions were met for the use of these methods.

Simple plots, product-moment correlation coefficients, and linear regression were used to examine the possible relation between baseline gait speed, balance, attention, fatigue and depression, and gait speed. Justification of the sample size was based on our earlier work,1 which showed a standard deviation (SD) of 6.2 for community walkers within a 30 to 50m/min range. Nine participants at each level of task or environment had an 80% power to detect a difference of 9.7m/min with a type I error rate of .05. We were also interested in estimation of variability of gait speed under experimental conditions to guide future research with similar subjects. In our study with 27 subjects, with baseline gait speed as a covariate, we had 21 degrees of freedom to estimate the root mean square error (RMSE), which concurs with Mead’s33 recommendation that there should be 15 to 20 for this calculation. SPSSb was used to analyze the results.

Back to Article Outline

Results 

Participant Characteristics 

Twenty men and 7 women with stroke participated in the study; 14 presented with a left-sided stroke, 12 with a right-sided stroke, and 1 with a subarachnoid hemorrhage. The demographic and clinical details for the sample are given in table 1. Scores for depression and anxiety and fatigue were higher than population norms, as might be expected. The BBS scores were low given the level of mobility and were only just higher than the threshold of 45, which discriminates fallers from nonfallers.34 Participants also had lower attention scores than age-matched population norms at the 50th percentile, although the median scores were normal for elevator counting and visual elevator subtests. Subtest scores were as follows: (1) for elevator counting, 15 (55.5%) scores were normative, 9 (33.3%) possibly abnormal, and 3 (11.1%) abnormal; (2) for visual elevator, 16 (69.5%) of the 23 who completed reached the 50th percentile or above; and (3) for telephone search while counting, 7 (29.1%) of the 24 who completed reached the 50th percentile or above.

Table 1. Baseline Characteristics of Stroke Participants (N=27 unless otherwise stated)⁎
Baseline CharacteristicsValues
Age (y)61.0±11.6
BBS score (normative score, 56)46.5±5.7
Time since onset (mo)45.8±34.2
HADS anxiety (total score, 21; normative range, 0–7)5.8±3.2
HADS depression (total score, 21; normative range, 0–7)4.6±2.5
MFI reduced activity (total score, 20)11.8±4.4
MFI physical fatigue (total score, 20)12.4±3.1
TEA elevator counting (normative score, 7.0)7.0(6.0–7.0)
TEA visual elevator (n=23) (age-matched 50th percentile, 9.0)9.0(7.0–10.0)
TEA telephone search while counting (n=24) (age-matched 50th percentile, 1.4)4.0(0.9–14.3)
Baseline gait speed (m/min) (age-matched normative score, 73.0)42.3±6.0

NOTE. Values are mean ± SD or mean (interquartile range).

⁎ The total possible score is indicated by each test along with normative values where appropriate. Median scores have been given for data not normally distributed. TEA figures are raw scores.

Gait Parameters 

Table 2 shows the descriptive statistics for the 3 environments and 3 tasks. The mean gait speed ± SD for all conditions was 41.0±10.4m/min, with less variance shown in gait parameters in the clinic compared with other environments. The interaction between environment and task was not significant (F=.49, P=.73). There was a moderately strong and significant relation between the 10MWT (baseline gait speed) and gait speed (r=0.7, P<.01), and baseline gait speed was an important predictor of gait speed. After adjusting for baseline gait speed, the environment influenced gait speed but neither motor nor cognitive task did (table 3).

Table 2. Descriptive Statistics of Gait Parameters by Testing Condition
EnvironmentMean ± SDConditionMean ± SD
Speed (m/min) Speed (m/min)
All (N=27)41.0±10.4All (N=27)41.0±10.4
Clinic47.0±6.5None42.5±10.6
Street39.9±13.3Cognitive43.9±10.2
Mall36.1±8.0Motor36.7±10.1
Step frequency Step frequency
(steps/min) (steps/min)
All (N=27)83.4±11.7All (N=27)83.4±11.7
Clinic88.1±6.5None83.1±10.2
Street82.1±13.6Cognitive85.6±12.6
Mall80.1±13.4Motor81.6±13.2
Step length (cm) Step length (cm)
All (N=27)0.49±0.09All (N=27)0.49±0.09
Clinic0.53±0.06None0.51±0.08
Street0.49±0.09Cognitive0.51±0.08
Mall0.44±0.09Motor0.45±0.09

NOTE. Number of subjects per environment and per task (n=9).

Table 3. Summary Table for ANOVA for Gait Speed
SourcedfSum of SquaresMean SquareF RatioP
Baseline speed1992.4992.48620.3<.001
Environment2348.2174.13.6.046
Task2100.150.01.0.37
Error211014.948.3
Total262817.8

NOTE. RMSE=6.95; R2=.64.

Table 4 shows the multiple comparisons between different levels of environment and task. The comparisons between different tasks are shown for illustrative purposes only as the overall test for this factor was not important in the ANCOVA. The main difference was a decrease in walking speed of 8.5m/min (20.7% of gait speed) in the shopping mall compared with the clinic environment. Step frequency and step length were not significantly altered as a result of walking in different environments or doing different tasks (see table 2).

Table 4. Data for Multiple Comparison Tests: Differences in Gait Speed for Environment and Task Conditions by Using the Tukey-Kramer Multiple-Comparison Procedure
EnvironmentTask
ComparisonEstimate (95% CI)ComparisonEstimate (95% CI)
Clinic minus mall8.5(−.09to17.1)Cognitive minus motor4.3(−4.2to12.9)
Clinic minus street2.5(−6.5to11.2)Cognitive minus none−0.4(−9.0to8.1)
Street minus mall6.1(−2.4to14.8)Motor minus none−4.3(−12.9to4.2)

Abbreviation: CI, confidence interval.

Of the 9 participants asked to perform the cognitive test, 4 (44.4%) answered 34 questions correctly, 2 (22.2%) 32, 1 (11.1%) person 27, and 2 (22.2%) 15 (both of these subjects also had the lowest TEA scores).

Relation Between Variables and Walking Speed 

Simple bivariable plots of balance, attention, fatigue, and anxiety and depression suggested no relation with gait speed. Although the participants were all chronic stroke survivors, time since onset was explored further as a potentially important variable because it was hypothesized that the skills required for community ambulation may be acquired over time with exposure to different community environments, especially in a cohort who presented with gait speeds low enough to indicate that community ambulation was marginal.1, 2 Examination of the plots suggests that people with stroke walk faster as time goes on and with less variable gait speeds (fig 2). Participants with a time since onset of less than 36 months (n=17) presented with gait speeds ranging from 30.3 to 50m/min when compared with those after 36 months (n=10) whose gait speeds ranged from 42.3 to 49.6m/min. There was a moderate and significant relation between time since onset and baseline gait speed (r=0.5, P<.01) confirmed by regression analysis (F1,25=8.5, P<.01) with 25% of the variance (R2) in gait speed explained by time since onset. However, the correlation between time since onset and overall gait speed was not significant and was very low (r=−.01, P=0.9), and caution must be taken when interpreting this result. The addition of covariates to the ANCOVA model did not improve the fit of the model and did not reduce the amount of variability.

Back to Article Outline

Discussion 

The main finding from this exploratory study was that the environment, in particular the shopping mall, resulted in a slower gait speed for people with stroke while performing a secondary task did not. There was no evidence of an interaction between task and environment, although, as is frequently the case with testing for the importance of interaction effects, the study lacked statistical power to detect a small but potentially important interaction effect. The 8.5m/min difference in gait speed between the clinic and the shopping mall needs to be examined more critically because it only just meets the threshold for a clinically relevant difference,32 which this study was sufficiently powered to detect. Also, a slower walking speed is more likely to be observed in a shopping mall compared with other locations simply because of the unpredictability of constraints imposed by the environment itself. Although none of the conditions significantly affected the step length and step frequency, all gait parameters in the clinic were more stable than in more complex environments, which has implications for transference of skill given that gait rehabilitation is often confined to clinical settings and also for safety because of the comparative unpredictability of community environments, with altered ambient conditions, obstacles, perturbations, and so on. We failed to show a significant difference in gait parameters during task performance, although the motor task was slower by 5.8m/min (point estimate). This suggests that participants were able to divide their attention between walking and the secondary task without any impact on walking speed. Although this study did not conform to a classic dual-task research paradigm and address the methodologic concerns of a dual-task study,35 the experimental protocol was designed to reflect impaired gait performance under secondary task conditions should one exist. However, there may have been a cognitive or attentional cost associated with the maintenance of gait performance under these conditions, which a formal dual-task research design may have been able to discern.

A criticism of the motor task is that it represents obstacle negotiation rather than a secondary motor task. Therefore, any changes in gait performance, compared with walking only, may be because of motor impairment and biomechanic constraints rather than cognitive status or cognitive-motor interference. However, there are inherent difficulties in selecting a motor task for people with stroke because of their often poor-functioning hemiparetic upper limb, which limits the utility of tasks such as carrying or picking up an object.

For the cognitive task, 6 of the 9 participants had no problem in answering the questions put to them during their walks, whereas 3 found this more difficult. Because we did not test them on the cognitive test before their walk, we cannot make any comment about a change in this aspect of their performance. The lack of change in gait speed is surprising and noteworthy given the growing body of evidence that reports a decrement in performance under secondary task conditions for people with compromised motor and/or cognitive systems. There are several potential explanations that may account for this finding. The first is that the participants in this study had experienced their stroke, on average, almost 4 years ago. They were all independent community ambulators who, although their gait speeds were far from normative, were regularly walking in a range of environments. The participants in the study by Bowen et al17 were, on average, only 48 days poststroke, and those followed up in the study by Haggard and Cockburn18 were on average 11 months poststroke. Because motor relearning occurs over time and gait skills are reacquired, the demand on central attentional resources is decreased.36 It is possible, therefore, that over time the cohort in this study had improved their gait and ability to manage distractions, through a process of restitution and compensation, to the point in which their gait had achieved a level of automaticity that approximates normal even in the presence of abnormal gait patterns. The association between time since onset and reduced variability and increase in gait speed supports this view; however, this is only tentative and restoration of gait automaticity over time needs to be explored further.

The second possible reason for these findings is that the tasks were not difficult enough for a group of people at this level of functioning. There is evidence that there is a critical level of task complexity above which performance will decrease, and this task may not have reached the critical level for these subjects.37 A decrement in gait performance may have been more evident if the obstacle had been higher or wider or out of their approaching view or if the environment had included more uneven surfaces such as gravel or a steeper incline. Also, a more challenging cognitive task, such as those used in earlier studies (eg, the Stroop test)38 or autobiographical recall10 might have produced a change in gait parameters. However, the safety concerns of the participants who were at possible risk of falling were paramount, especially in view of the exploratory nature of this study. Also, the participants’ attentional resources may have been stimulated because of the testing procedure, particularly because the secondary task was delivered at set rather than random, unpredictable time intervals. Third, it is important to note that the participants in this study were relatively young and their gait performance unimpeded by the secondary affects of age-related impairments.

We have not shown an influence of fatigue, anxiety and depression, and attention on gait speed despite earlier studies indicating that these factors may be important to people poststroke. However, the study was not designed with sufficient statistical power to detect small but important associations that may in fact exist between these covariables and gait speed. Depression is common after stroke, and a strong link has been established between poststroke depression, stroke severity, and functional performance.39 Fatigue is also reported as a common complaint late after stroke, with or without associated depression.40 Although fatigue and depression were present in the stroke participants to a greater extent that in normal populations, their effect on performance, if one existed, was undetected in this study. There may be a threshold of symptom severity that must be passed before the effects of depression and fatigue impact noticeably on gait function, and this small cohort did not reach that threshold. Similarly, the TEA scores were sufficiently high not to show an impact on gait performance, although the scores obtained on the elevator counting and the telephone search while counting supported earlier findings on a similar cohort.29

Back to Article Outline

Conclusions 

This group of community ambulant chronic stroke survivors maintained their gait speed even under challenging environmental conditions and while performing a secondary task. The reasons for this are not totally clear but may be because of a restoration over time of gait automaticity, despite the presence of ongoing gait impairment. This is encouraging for people with stroke and for those involved in their rehabilitation and raises the question of the appropriateness of timing of interventions to enhance this aspect of gait performance for community conditions.

Suppliers

Back to Article Outline

Acknowledgment 

We thank Kimberley Donovan for her assistance in recruitment and assessment of study participants.

Back to Article Outline

References 

  1. Lord S , McPherson K , McNaughton H , Rochester L , Weatherall M . Community ambulation after stroke (how important and obtainable is it and what measures appear predictive?) . Arch Phys Med Rehabil . 2004;85:234–239
  2. Hill K , Ellis P , Bernhardt J , Maggs P , Hull S . Balance and mobility outcomes for stroke patients (a comprehensive audit) . Aust J Physiother . 1997;43:173–180
  3. Perry J , Garrett M , Gronley JK , Mulroy SJ . Classification of walking handicap in the stroke population . Stroke . 1995;26:982–989
  4. Pound P , Gompertz P , Ebrahim S . A patient-centred study of the consequences of stroke . Clin Rehabil . 1998;12:338–347
  5. Patla A , Shumway-Cook A . Dimensions of mobility (defining the complexity and difficulty associated with community mobility) . J Aging Phys Act . 1999;7:7–19
  6. Woollacott M , Shumway-Cook A . Attention and the control of posture and gait (a review of an emerging area of research) . Gait Posture . 2002;16:1–14
  7. Paul S , Ada L , Canning C . Automaticity of walking—implications for physiotherapy practice . Phys Ther Rev . 2005;10:15–23
  8. Camicioli R , Howieson D , Lehman S , Kaye J . Talking while walking (the effect of a dual task in aging and Alzheimer’s disease) . Neurology . 1997;48:955–958
  9. Lundin-Olsson L , Nyberg L , Gustafson Y . “Stops walking when talking” as a predictor of falls in elderly people . Lancet . 1997;349:617
  10. Rochester L , Hetherington V , Jones D , et al.   Attending to the task (interference effects of functional tasks on walking in Parkinson’s disease and the roles of cognition, depression, fatigue, and balance) . Arch Phys Med Rehabil . 2004;85:1578–1585
  11. Lundin-Olsson L , Nyberg L , Gustafson Y . Attention, frailty, and falls (the effects of a manual task on basic mobility) . J Am Geriatr Soc . 1998;46:758–761
  12. Chen HC , Ashton-Miller JA , Alexander NB , Schultz AB . Effects of age and available response time on ability to step over an obstacle . J Gerontol . 1994;49:M227–M233
  13. Weerdesteyn V , Schillings AM , van Galen GP , Duysens J . Distraction affects the performance of obstacle avoidance during walking . J Mot Behav . 2003;35:53–63
  14. Lajoie Y , Teasdale N , Bard C , Fleury M . Upright standing and gait (are there changes in attentional requirements related to normal aging?) . Exp Aging Res . 1996;22:185–198
  15. Brauer S , Woollacott M , Shumway-Cook A . The influence of a concurrent cognitive task on the compensatory stepping response to a perturbation in balance-impaired and healthy elders . Gait Posture . 2002;15:83–93
  16. Maylor EA , Wing AM . Age differences in postural stability are increased by additional cognitive demands . J Gerontol B Psychol Sci Soc Sci . 1996;51:P143–P154
  17. Bowen A , Wenman R , Mickelborough J , Foster J , Hill E , Tallis R . Dual-task effects of talking while walking on velocity and balance following a stroke . Age Ageing . 2001;30:319–323
  18. Haggard P , Cockburn J . Concurrent performance of cognitive and motor tasks in neurological rehabilitation . Neuropsychol Rehabil . 1998;8:155–170
  19. Cockburn J , Haggard P , Cock J , Fordham C . Changing patterns of cognitive-motor interference (CMI) over time during recovery from stroke . Clin Rehabil . 2003;17:167–173
  20. Patla A . Mobility in complex environments (implications for clinical assessment and rehabilitation) . Neurol Rep . 2001;25(3):82–90
  21. McDowd J , Filion D , Pohl P , Richards L , Stiers W . Attentional abilities and functional outcomes following stroke . J Gerontol B Psychol Sci Soc Sci . 2003;58:P45–P53
  22. Robertson I , Ridgeway V , Greenfield E , Parr A . Motor recovery after stroke depends on intact sustained attention (a 2-year follow-up study) . Neuropsychology . 1997;11:290–295
  23. Evans M , Goldie P , Hill K . Systematic and random error in repeated measurements of temporal and distance parameters of gait after stroke . Arch Phys Med Rehabil . 1997;78:725–729
  24. Asplund K , Tuomilehto J , Stegmayr B , Wester P , Tunstall-Pedoe H . Diagnostic criteria and quality control of the registration of stroke events in the MONICA project . Acta Med Scand Suppl . 1998;728:26–39
  25. Bleeker M , Bolla-Wilson K , Kawas C , Agnew J . Age-specific norms for the Mini-Mental State Exam . Neurology . 1988;38:1565–1568
  26. Macko R , Haeuber E , Shaughnessy M , et al.   Microprocessor-based ambulatory activity monitoring in stroke patients . Med Sci Sports Exerc . 2000;34:394–399
  27. Dean C , Richards C , Malouin F . Walking speed over 10 metres overestimates locomotor capacity after stroke . Clin Rehabil . 2001;15:415–421
  28. Berg K , Wood-Dauphinee S , Williams JI . The Balance Scale (reliability assessment with elderly residents and patients with an acute stroke) . Scand J Rehabil Med . 1995;27:27–36
  29. Hyndman D , Ashburn A . People with stroke living in the community (attention deficits, balance, ADL ability and falls) . Disabil Rehabil . 2003;25:817–822
  30. Smets E , Garssen B , Bonke B , de Haes JC . The Multidimensional Fatigue Inventory (MFI) psychometric properties of an instrument to assess fatigue . J Psychosom Res . 1995;39:315–325
  31. Zigmund A , Snaith R . The hospital anxiety and depression scale . Acta Psychiatr Scand . 1983;67:361–370
  32. Tyson S , DeSouza L . The measurement of balance and walking post-stroke. Part 2: functional performance tests . Phys Ther . 2002;7:187–191
  33. Mead R . The design of experiments . Cambridge: Cambridge Univ Pr; 1988;
  34. Thorbahn L , Newton R . Use of the Berg Balance Scale to predict falls in the elderly . Phys Ther . 1996;76:576–583
  35. Abernethy B . Dual-task methodology and motor skills research (some applications and methodological constraints) . J Hum Mov Stud . 1988;14:101–132
  36. Geurts AC , Mulder TW , Nienhuis B , Rijken RA . Dual-task assessment of reorganization of postural control in persons with lower limb amputation . Arch Phys Med Rehabil . 1991;72:1059–1064
  37. Bond J , Morris M . Goal-directed secondary motor tasks (their effects on gait in subjects with Parkinson disease) . Arch Phys Med Rehabil . 2000;81:110–116
  38. Persad C , Giordani B , Chen HC , et al.   Neuropsychological predictors of complex obstacle avoidance in health older adults . J Gerontol B Psychol Sci Soc Sci . 1995;50:P272–P277
  39. Bogousslavsky J . Emotions, mood, and behaviour after stroke . Stroke . 2003;34:1046–1050
  40. Staub F , Bogousslavsky J . Fatigue after stroke (a major but neglected issue) . Cerebrovasc Dis . 2001;12:75–81
  • a Cyma Inc, 8515 35th Ave NE, Ste C, Seattle, WA 98115.
  • b Version 12.0.1; SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.

 Supported by the Wellington Medical Research Foundation (grant no. 2003/78).

 No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated.

PII: S0003-9993(06)00281-4

doi:10.1016/j.apmr.2006.03.003

Archives of Physical Medicine and Rehabilitation
Volume 87, Issue 7 , Pages 967-973, July 2006

Article Tools

* Image Usage & Resolution