1Department of Rehabilitation Sciences, The Hong Kong Polytechnic University, Hung Hom, Hong Kong SAR, China
2Research Centre for Chinese Medicine Innovation, The Hong Kong Polytechnic University, Hung Hom, Hong Kong SAR, China
Correspondence: Shamay S. M. Ng Department of Rehabilitation Sciences, QT515, 5/F, The Hong Kong Polytechnic University, 11 Yuk Choi Rd, Hung Hom, Kowloon, Hong Kong SAR, China. Tel: +852-2766-4889 Fax: +852-2330-8656 E-mail: Shamay.Ng@polyu.edu.hk
• Received: December 3, 2024 • Revised: April 9, 2025 • Accepted: April 27, 2025
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
To examine the contribution of perceived upper limb function to participation and activity among community-dwelling people with chronic stroke.
Methods
A cross-sectional study was conducted with eighty-one people with stroke aged ≥50 years. The outcome measures included the Oxford Participation and Activities Questionnaire (Ox-PAQ), Arm Activity Measure (ArmA), Wolf Motor Function Test (WMFT), Timed Up and Go Test (TUG), and Geriatric Depression Scale (GDS).
Results
Correlation analyses revealed that perceived upper limb function, as measured using the ArmA, had the strongest and most significant correlations with the levels of participation and activity, as measured using the Ox-PAQ, among all of the tested variables (rs=0.35–0.59, p<0.01). Multiple linear regression analyses also showed that perceived upper limb function significantly associated the levels of participation and activity, accounting for 2.0% to 9.0% of the variance in the Ox-PAQ scores. The final model, which included TUG time, the GDS score, the WMFT score, and the ArmA score, could explain 54% and 28% of the variance in the routine activities and social engagement subscales of the Ox-PAQ, respectively. The model including the GDS score, the WMFT score, and the ArmA score explains 32% of the variance in the emotional well-being subscale of the Ox-PAQ.
Conclusion
Perceived upper limb function is a crucial determinant of participation and activity among community-dwelling people with chronic stroke. It could thus be a target component of stroke rehabilitation interventions to facilitate participation and activity after stroke.
Stroke is a major cause of disability among adults, with a high global incidence of 12.2 million and a prevalence of 101 million in 2019 [1]. The incidence of stroke is predicted to increase by 48% by 2030 due to population ageing and increased exposure to modifiable risk factors, such as high blood pressure and diabetes [2]. People with stroke may experience restrictions in social participation and limitations in activities of daily living resulting from sensorimotor deficits and psychological complications.
The International Classification of Functioning, Disability and Health (ICF) defines participation as “involvement in a life situation” and activity as “the execution of a task or action by an individual” [3]. At 6 months after stroke, 65% of stroke survivors report restrictions in participation and 54% report limitations in activities of daily living [4]. After stroke, the levels of participation and activity may improve over time, but 5% to 19% of stroke survivors still perceive restrictions in different domains of participation and daily activity, such as family role and autonomy in daily life, 5 years after stroke [5]. As the goal of stroke rehabilitation is to promote independence in daily activities and, ultimately, social participation among people with stroke [6], clinicians and researchers need to identify the major determinants of participation and activity in this cohort.
The Oxford Participation and Activities Questionnaire (Ox-PAQ) is a patient-reported outcome measure of participation and activity and is theoretically grounded in the ICF framework [7]. It has been used to assess the levels of participation and activity in people with motor neuron disease, multiple sclerosis, and Parkinson’s disease [7]. It has also been tested in people with chronic stroke [8]. Although there are a number of validated instruments for assessing participation and activity, such as the Stroke Impact Scale, London Handicap Scale, and Assessment of Life Habits, they fail to cover both the “participation” and “activity” domains [9]. The Ox-PAQ, which is an ICF-model-based measure, may help to provide comprehensive insight into participation and activity among people with stroke in clinical rehabilitation.
In the current stroke rehabilitation, upper limb function can be investigated with observation-based assessment and patient-reported tools. In the observation-based assessment, the actual performance of movement or task using the upper limb is quantified and scored by the therapists and defined as observed upper limb function [10]. On the other hand, perceived upper limb function reflects the perceived performance or ability to perform daily tasks from a patient’s perspective outside the rehabilitation setting and is rated by the respondents in the patient-reported tools [10]. The measures of perceived upper limb function can provide information about the difficulties experienced in daily life, which cannot be directly observed or quantified by others [11]. Prior studies have reported that approximately 20% to 40% of stroke survivors who had perfect to near-perfect observed upper limb function reported perceived residual disability [10-12], suggesting a discrepancy between the observed and perceived upper limb function in people after stroke. It highlighted the importance of assessing perceived upper limb motor function and including patient-reported outcome measures in order to derive a comprehensive understanding of the impairments of motor functions following stroke. For stroke survivors with satisfactory observed upper limb function but fair to poor perceived upper limb function, incorporating interventions aimed at improving perceived motor function is crucial for achieving independence in performing daily tasks and maximizing functional recovery after a stroke [13]. Perceived upper limb function has been shown to be significantly correlated with the health-related quality of life (rs=0.68, p<0.001) [14] in people with stroke. Poor perceived upper limb function may discourage the use of the paretic arm, which may hinder functional recovery [15] and ultimately reduce the health-related quality of life after stroke.
The Arm Activity Measure (ArmA) was developed to assess perceived difficulties in performing daily activity tasks among people with unilateral paresis [16]. It assesses the subjective difficulty in performing active and passive functional aspects of daily activity tasks. Active and passive functions are two separate concepts indicating different functional aspects in people with stroke [17]. Active function is defined as performance in holding and manipulating objects, and passive function is defined as the ability to care for the affected arm [18]. Although several self-reported instrument have been developed to reflect the “real-life” active and passive functions, such as the Leeds Adult Spasticity Impact Scale (LASIS) and ABILHAND, the psychometric testing has not been undergone on the LASIS and the floor effect may exist in people with stroke using ABILHAND due to its complexity of the items [18]. Therefore, considering that the ArmA equally covers both the active and passive functions of the upper limb with good psychometric properties [19], it may help clinicians to comprehensively assess the perceived functional performance of the upper limb after stroke.
Although perceived upper limb function has been identified as a significant predictor of the levels of participation in real-life situations (ꞵ=0.24–0.41, p<0.05) in patients with hereditary motor and sensory neuropathy [20], the role of perceived upper limb function in the levels of participation and activity has not been investigated among people with stroke. Also, despite significant associations between the perceived upper limb function and health-related quality of life identified in prior study [14], whether perceived upper limb function independently contributes to both the participation and activity domains remains unclear. As the participation and activity are two separate concepts, the role of perceived upper limb function in two domains may be different. Thus, this study aimed to quantify the relative contribution of perceived upper limb function to participation and activity among community-dwelling people with chronic stroke. In addition, since previous studies suggested that functional mobility (ꞵ=-0.47, p<0.001) and occurrence of depressive symptoms (ꞵ=-0.39, p<0.001) were significant predictors of social participation [21,22], this study also aimed to elucidate the relationship between perceived upper limb function and the levels of participation and activity after controlling the effect of functional mobility and depressive symptoms in people with stroke.
METHODS
Participants
Community-dwelling people with chronic stroke were recruited from local self-help groups. People with stroke were recruited if they were aged ≥50 years, had had a stroke at least 12 months previously, were able to understand Cantonese, and scored ≥7 on the Abbreviated Mental Test. People with stroke were excluded if they had other neurological diseases, expressive or receptive aphasia, or any other orthopaedic or medical condition that might hinder proper assessment.
The study protocol was approved by the Human Subjects Ethics Committee of The Hong Kong Polytechnic University (approval number: HSEAR20210110002). The study was conducted according to the guidelines of the Declaration of Helsinki. All participants gave their written informed consent prior to their enrolment in the study.
Demographic data
Two types of demographic data were collected before the assessment: (1) background characteristics, namely age, sex, and body mass index, and (2) stroke-related variables, namely post-stroke duration, cause of stroke, and hemiplegic side.
Outcome measures
Levels of participation and activity
The levels of participation and activity were assessed using the Chinese version of the Ox-PAQ [8]. The Ox-PAQ has 23 items that are scored on a 5-point Likert scale and comprises three subscales to quantify the levels of participation and activity, namely routine activities subscale, social engagement subscale, and emotional well-being subscale. Higher scores represent greater difficulties in participation and activities. The Chinese version of the Ox-PAQ has been reported to have excellent content validity (scale-level content validity index=1.00), good internal consistency (Cronbach’s alpha coefficients=0.86–0.94) and excellent test–retest reliability (intraclass correlation coefficients [ICCs]=0.91–0.94) in people with stroke [8].
Perceived upper limb function
Perceived upper limb function was assessed using the Chinese version of the ArmA [19]. The ArmA consists of two subscales, one each to evaluate active and passive functions. The active function subscale contains 13 items to assess the difficulty in performing tasks or activities using the affected arm, such as drinking from a cup or mug, while the passive function subscale comprises eight items to quantify the difficulty in caring for the affected arm, such as cutting fingernails. Higher scores indicate poorer perceived upper limb function. The Chinese version of the ArmA has demonstrated good test–retest reliability (ICCs=0.87–0.93), content validity (scale-level content validity index=0.96), and internal consistency (Cronbach’s alpha coefficients=0.75–0.95) [19].
Upper limb functional performance
The functional performance of the paretic upper limb was evaluated using the functional ability score of the Wolf Motor Function Test (WMFT) [23]. The WMFT includes six joint-segment movements and nine functional movements and is divided into 2 scales which are the functional ability score and time score [23]. The functional ability score in each task is rated on a 6-point scale. A higher score is indicative of better upper limb functional performance. However, the time score of the WMFT was not included in this study due to the potential floor effect and its inability to accurately represent the performance of severely impaired stroke survivors who were not able to complete at least half of the tasks [24]. The functional ability score of the WMFT has demonstrated excellent inter-rater reliability (ICC=0.98) and intra-rater reliability (ICC=0.99) in people with stroke [25].
Functional mobility
Functional mobility was measured using the Timed Up and Go Test (TUG) [26]. The TUG requires participants to stand up from a chair, walk 3 metres forward, turn around, return to the chair, and sit down [26]. The time taken to complete the test is recorded in seconds. A shorter completion time indicates better functional mobility. In this study, the mean completion time of two trials was calculated for analysis. The TUG has been demonstrated to have excellent reliability (ICCs=0.95–0.96) in people with stroke [27].
Depressive symptoms
The Chinese version of the Geriatric Depression Scale (GDS) was used to assess depressive symptoms using 15 dichotomous items [28]. A higher GDS score indicates more severe depressive symptoms. The Chinese version of the GDS has been psychometrically tested in people with stroke and has demonstrated good internal consistency (Cronbach’s alpha coefficient=0.78) [28].
Statistical analysis
The demographic characteristics of the participants and results of all of the outcome measures were summarised using descriptive statistics. The Ox-PAQ scores of participants with different sexes, causes of stroke, and hemiplegic sides were compared using the independent samples t-test. The normality of the distribution of the data was examined using the Kolmogorov–Smirnov test. The relationships between the Ox-PAQ scores and other variables were assessed using Pearson’s correlation coefficients if the data were normally distributed; otherwise, Spearman’s correlation coefficients were calculated. Partial correlation coefficients between the Ox-PAQ scores and other variables were also calculated after controlling for the TUG time and GDS score. Multiple linear regression with a forced entry method was applied to determine the contribution of independent variables in predicting the Ox-PAQ scores. All statistical analyses were performed using the IBM SPSS version 28.0 (IBM Corp.) with the significance level set to 0.05 (two-tailed).
RESULTS
A total of 81 people with chronic stroke (49 males and 32 females) with a mean age of 63.16±6.28 years and an average post-stroke duration of 6.82±4.30 years participated in the study (Table 1). The mean body mass index of the participants was 24.06±3.01 kg/m2. Among the participants, 58 had experienced ischaemic stroke and 23 had experienced haemorrhagic stroke. Thirty-six of them had left hemiplegia and 45 had right hemiplegia. No significant difference in the Ox-PAQ scores was found between male and female, participants with ischaemic stroke and haemorrhagic stroke, and those with left hemiplegia and right hemiplegia (Supplementary Table S1). Their mean TUG test completion time was 18.05±14.96 seconds, and their average GDS score was 4.74±3.97.
No significant correlations were observed between the Ox-PAQ scores and age or post-stroke duration (Fig. 1A-F and Supplementary Table S2). Among the Ox-PAQ subscales, the routine activities subscale score (rs=0.47, p<0.001) and social engagement subscale score (rs=0.22, p=0.046) demonstrated significant medium and small correlations with the TUG time respectively, while the emotional well-being subscale score showed no significant correlation with the TUG time (Fig. 1G-I). The routine activities subscale score, social engagement subscale score, and emotional well-being subscale score demonstrated significant small to medium correlations with the GDS score (rs=0.31, p=0.005; rs=0.22, p=0.049; and rs=0.38, p<0.001, respectively; Fig. 1J-L) and the WMFT score (rs=-0.44, p<0.001; rs=-0.28, p=0.011; and rs=-0.30, p=0.006, respectively; Fig. 1M-O). Furthermore, the routine activities subscale score, social engagement subscale score, and emotional well-being subscale score demonstrated significant medium to large correlations with the ArmA active function (rs=0.53, p<0.001; rs=0.35, p=0.002; and rs=0.35, p=0.001, respectively; Fig. 1P-R) and the ArmA passive function score (rs=0.59, p<0.001; rs=0.52, p<0.001; and rs=0.48, p<0.001, respectively; Fig. 1S-U).
After controlling for the TUG time and GDS score, a significant small partial correlation was identified between the Ox-PAQ routine activities subscale score (rs=-0.26, p=0.021) and the WMFT score, but the correlations between the social engagement subscale and emotional well-being subscale scores and the WMFT score became insignificant (Supplementary Table S3). Furthermore, the routine activities subscale score, social engagement subscale score, and emotional well-being subscale score demonstrated significant small to medium partial correlations with the ArmA active function score (rs=0.44, p<0.001; rs=0.33, p=0.003; and rs=0.27, p=0.017, respectively) and significant medium to large partial correlations with the ArmA passive function score (rs=0.54, p<0.001; rs=0.40, p<0.001; and rs=0.38, p<0.001, respectively).
The functional mobility (TUG time), depressive symptoms (GDS score), and upper limb functional performance (WMFT score) accounted for 34% of the variance in the Ox-PAQ routine activities subscale score (Model 2 in Table 2). The addition of the ArmA active function score accounted for an additional 9% of the variance (Model 3 in Table 2). With the further addition of the ArmA passive function score, the explained variance increased by 7%, and the final model predicted a total of 54% of the variance in the Ox-PAQ routine activities subscale score (F5, 75=17.23, p<0.001) (Model 4 in Table 2). The TUG time, GDS score, ArmA active function score, and ArmA passive function score were significant predictors of the routine activities subscale score. Among the variables, the ArmA passive function score was the best predictor, as reflected by the magnitude of the standardised regression coefficient (ꞵ=0.38) (Model 4 in Table 2) and the highest Spearman’s correlation coefficient (rs=0.59, p<0.001; Fig. 1S, Supplementary Table S2).
Table 3 shows the results of the multiple linear regression analysis to predict the Ox-PAQ social engagement subscale score. The TUG time, GDS score, and WMFT score accounted for 14% of the variance in the social engagement subscale score (Model 2 in Table 3). The addition of the ArmA active function score accounted for an additional 7.0% of the variance (Model 3 in Table 3). After the addition of the ArmA passive function score, a total of 28% of the variance in the Ox-PAQ social engagement subscale score was predicted by the final regression model (F5, 75=5.74, p<0.001) (Model 4 in Table 3). Among the variables, the ArmA passive function score was a significant independent predictor of the Ox-PAQ social engagement subscale score, as reflected by the magnitude of the standardised regression coefficient (ꞵ=0.32) (Model 4 in Table 3) and the highest Spearman’s correlation coefficient (rs=0.52, p<0.001; Fig. 1T, Supplementary Table S2).
The combined multiple linear regression model, which included GDS score and WMFT score predicted 23% of the variance in the Ox-PAQ emotional well-being subscale score (Model 2 in Table 4). The addition of ArmA active and passive function score accounted for an additional 2% and 7% of the variance respectively, and the final model predicted a total of 32% of the variance in the Ox-PAQ emotional well-being subscale score (F4, 76=9.10, p<0.001) (Model 4 in Table 4). The GDS and ArmA passive function score were significant predictors of the Ox-PAQ emotional well-being subscale score.
DISCUSSION
To the best of our knowledge, this is the first study to investigate the contribution of perceived upper limb function to the participation and activity separately among community-dwelling people with chronic stroke. Our results showed that perceived upper limb function measured using the ArmA was a significant determinant of participation and activity levels measured using the Ox-PAQ. The findings of this study improve the current understanding of the role of perceived upper limb function in stroke rehabilitation.
Participation and activity levels after stroke
The community-dwelling people with chronic stroke in this study had mean Ox-PAQ subscale scores ranging from 14.81±20.62 to 16.60±18.81, meaning that their levels of participation and activity were comparable to those of the chronic stroke survivors studied by Ng et al. [8] (mean Ox-PAQ subscale scores=18.06±16.58 to 19.50±20.80). This may be due to the similar characteristics of the participants in the two studies. The participants with stroke in both studies were recruited from local self-help groups and actively engaged in social activities. They also had similar post-stroke duration (6.82±4.30 years in this study and 7.76±4.44 years in the study by Ng et al. [8]). The literature suggests that the level of participation is significantly positively associated with the chronicity of stroke because stroke survivors with longer post-stroke duration may develop compensatory strategies to accommodate constraints in their social participation and daily activities [22]. Thus, the participants with stroke in both studies may have had comparable participation and activity levels.
Perceived upper limb function after stroke
The ArmA scores of our participants with stroke (median scores for passive and active function: 1 and 31, respectively) were lower than those of the patients with upper paresis in the studies by Ashford et al. [16] (median scores for passive and active function: 12 and 48, respectively) and Ramström et al. [29] (median scores for passive and active function: 12 and 46, respectively). This finding indicates that the community-dwelling people with chronic stroke in our study perceived less difficulty in passive and active upper limb functional tasks than the participants in the previous studies. This discrepancy may be attributable to the different characteristics of the study samples. The previous two studies [16,29] recruited people with moderate-to-severe upper limb spasticity (e.g., Modified Ashworth Scale median score=3.0) caused by lesions to the central nervous system. Spasticity is strongly associated with average (χ2=11.00, p<0.001) and peak (χ2=12.53, p<0.001) elbow extension velocity, suggesting a significant effect of spasticity on the voluntary movement of the paretic arm [30]. The presence of spasticity limits the range of motion of the affected arm and, as a result, its performance in functional tasks [31], which may increase patients’ perceived difficulties in upper limb tasks. In contrast, community-dwelling people with stroke may have better upper limb performance due to their habit of regularly engaging in exercise and actively participating in daily activities; therefore, they may have had better passive and active functions, as measured using the ArmA subscales, than did the participants in prior studies.
Perceived upper limb function predicts the levels of participation and activity
The final regression model explained 54% of the variance in the Ox-PAQ routine activities subscale score. The perceived passive and active upper limb functions were significant predictors of the activity level after stroke, accounting for 7% and 9% of the variance, respectively. Bandura and Adams revealed that self-efficacy, defined as one’s belief in their ability to complete a task, could affect decisions about activity and behaviour in actual life [32]. A study revealed that stroke survivors with low perceived upper limb function showed poor functional performance (ꞵ=0.24, p=0.016) [33]. Stroke survivors with a better perception of their affected upper limb function may be more likely to use that limb in their routine activities. In contrast, people with stroke who have low perceived upper limb function may avoid performing activities using the affected limb, even though they may display good observed upper limb function. This can reduce the functional performance and activity level of people with stroke. However, it is surprising to note that the functional upper limb performance measured using the WMFT was not a significant determinant of the Ox-PAQ scores in this study (ꞵ=0.12, p=0.372). This indicates that the perceived upper limb function of people with stroke may make a larger contribution to their activity level than does their actual functional performance. In line with previous findings that the TUG contributes significantly to participation after stroke (ꞵ=-0.47, p<0.001) [21], our study found that functional mobility, as measured using the TUG, also significantly contributed to the activity level after stroke (ꞵ=0.33, p<0.001). This may be because the TUG includes a series of motor tasks, such as sit-to-stand, walking, turning, and stand-to-sit, which are commonly performed in routine activities. In addition, depressive symptom causes low volitional levels and sense of competence of engaging in daily activities in people with stroke [34], which may explain the significant contribution of GDS score to Ox-PAQ routine activities subscale score in this study.
Our results showed that perceived upper limb function significantly predicted the participation level of community-dwelling people with chronic stroke, explaining 28% of the variance in the Ox-PAQ social engagement subscale score. Perceived upper limb passive function was a significant predictor of participation after stroke. Passive function is defined as the ability to care for the affected arm, such as cleaning the palm of the affected hand and positioning the affected arm on a cushion or support in sitting [16]. The performance in maintaining personal hygiene and positioning of the upper limb may affect the appearance of an individual, which may further influence the self-esteem of people with stroke. Chau et al. [35] revealed that low self-esteem had a significant negative effect on the social participation of community-dwelling stroke survivors (ꞵ=0.20, p=0.002), because a decrease in self-esteem may increase social isolation, which can cause considerable constraints in social participation for people with physical disability [36]. This may explain the significant contribution of perceived passive function to participation after stroke. However, in our final model, 72% of the variance in the Ox-PAQ social engagement subscale score remained unexplained. Future research should consider the influence of other physical and psychological factors.
A total of 32% of the variance in the Ox-PAQ emotional well-being subscale score was explained in the final model. In this study, perceived passive upper limb function and depressive symptoms were significant determinants of the emotional well-being of community-dwelling people with stroke. Previous research has demonstrated that the perception of functional recovery is associated with the level of anxiety [37], and that the significant contribution of perceived passive upper limb function to emotional well-being after stroke may be due to emotional distress. Depressive symptoms also negatively influence the emotional well-being of people with stroke because of their associations with higher functional dependency, poor prognosis of functional recovery, and decreased quality of life [38]. Nevertheless, the mechanisms affecting emotional well-being after stroke may be multifactorial. Further studies should investigate the factors affecting the psychological well-being of people with chronic stroke.
Study limitations
Several limitations of this study should be borne in mind. First, approximately or more than half of the variance in the scores of the three Ox-PAQ subscales could not be explained by the variables in the regression models. Therefore, the effects of other elements such as cognitive function, dynamic balance, and walking endurance should be investigated in future studies. Second, the representativeness of the sample in this study may be limited as our participants recruited from local self-help groups were relatively socially active and had good physical functioning. Also, this study recruited stroke survivors over 50 years old only, which may introduce selection bias and limit the generalizability of results. Third, given the cross-sectional study design, no causal relationship between the variables could be established. Further research with a longitudinal design, including stroke survivors of a wider age range and in different clinical setting, is recommended to develop a predictive model of the levels of participation and activity among people with chronic stroke. Lastly, this study did not consider the effect of hand dominance and the location of stroke lesions on perceived upper limb function. As upper limb motor function after stroke could be affected by whether the dominant or non-dominant hand is affected and lesion location, complementary studies are recommended to explore the possible effects of hand dominance and location of stroke lesion on the participation and activity levels among people with chronic stroke.
Clinical implications
Assessment and treatment of perceived upper limb function are crucial to derive an effective intervention in improving the levels of participation and activity in people following stroke. Incorporation of behavioural strategies, such as self-monitoring, problem-solving, and home skill assignment [39], have been suggested to improve perceived upper limb function after stroke. Regular feedback given by clinicians or therapists may also improve the self-efficacy of people with stroke [40], which may further improve their perceived performance. Interventions targeting both perceived and observed upper limb function may improve independence in performing daily tasks, maximize functional recovery, and enhance the participation and activity levels of people with chronic stroke.
Conclusion
Our study revealed that perceived upper limb function is a significant predictor of the levels of participation and activity among community-dwelling people with chronic stroke. The findings underscore the importance of perceived upper limb function in participation and activity after stroke. Therefore, improving perceived upper limb function could be a target component of stroke rehabilitation interventions to facilitate participation and activity among people with chronic stroke.
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was reported.
FUNDING INFORMATION
This work was supported by the research funding of the Research Centre for Chinese Medicine Innovation of The Hong Kong Polytechnic University (Ref. No. P0041139). The funder had no roles in study design, collection, analysis and interpretation of data, writing of the report; and decision to submit the article for publication.
AUTHOR CONTRIBUTION
Conceptualization: Chan NH, Ng SSM. Data curation: Chan NH Formal analysis: Chan NH. Methodology: Chan NH, Ng SSM. Funding acquisition: Ng SSM. Investigation: Ng SSM. Project administration: Chan NH. Software: Chan NH. Supervision: Ng SSM. Writing – original draft: Chan NH. Writing – review & editing: Chan NH, Ng SSM. Approval of final manuscript: all authors.
Scatter plots showing the correlations of (A-C) age, (D-F) post-stroke duration, (G-I) TUG, (J-L) GDS, (M-O) WMFT, (P-R) ArmA (active function), and (S-U) ArmA (passive function) with the Ox-PAQ. TUG, Timed Up and Go Test; GDS, Geriatric Depression Scale; WMFT, Wolf Motor Function Test; ArmA, Arm Activity Measure; Ox-PAQ, Oxford Participation and Activity Questionnaire.
Table 1.
Characteristics of the participants
Characteristic
Value (N=81)
Age (yr)
63.16±6.28
Sex
Male
49 (60.49)
Female
32 (39.51)
Body mass index (kg/m2)
24.06±3.01
Post-stroke duration (yr)
6.82±4.30
Cause of stroke
Ischaemic
58 (71.60)
Haemorrhagic
23 (28.40)
Hemiplegic side
Left
36 (44.44)
Right
45 (55.56)
TUG time (s)
18.05±14.96
GDS score
4.74±3.97
WMFT score
47.10±21.06
ArmA score (passive function)
2.86±3.62
ArmA score (active function)
27.22±15.98
Ox-PAQ score
Routine activities subscale
16.45±16.77
Social engagement subscale
14.81±20.62
Emotional well-being subscale
16.60±18.81
Values are presented as mean±standard deviation or number (%).
TUG, Timed Up and Go Test; GDS, Geriatric Depression Scale; WMFT, Wolf Motor Function Test; ArmA, Arm Activity Measure; Ox-PAQ, Oxford Participation and Activity Questionnaire.
Table 2.
Multiple linear regression analyses using the routine activities subscale of the Ox-PAQ
Ox-PAQ, Oxford Participation and Activity Questionnaire; TUG, Timed Up and Go Test; GDS, Geriatric Depression Scale; WMFT, Wolf Motor Function Test; ArmA, Arm Activity Measure.
*p<0.05.
**p<0.01.
Table 4.
Multiple linear regression analyses using the emotional well-being subscale of the Ox-PAQ
Ox-PAQ, Oxford Participation and Activity Questionnaire; TUG, Timed Up and Go Test; GDS, Geriatric Depression Scale; WMFT, Wolf Motor Function Test; ArmA, Arm Activity Measure.
*p<0.05.
**p<0.01.
REFERENCES
1. GBD 2019 Stroke Collaborators. Global, regional, and national burden of stroke and its risk factors, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Neurol 2021;20:795-820.
2. Fan J, Li X, Yu X, Liu Z, Jiang Y, Fang Y, et al. Global burden, risk factor analysis, and prediction study of ischemic stroke, 1990-2030. Neurology 2023;101:e137-50.
4. Mayo NE, Wood-Dauphinee S, Côté R, Durcan L, Carlton J. Activity, participation, and quality of life 6 months poststroke. Arch Phys Med Rehabil 2002;83:1035-42.
5. Palstam A, Sjödin A, Sunnerhagen KS. Participation and autonomy five years after stroke: a longitudinal observational study. PLoS One 2019;14:e0219513.
6. van der Zee CH, Visser-Meily JM, Lindeman E, Jaap Kappelle L, Post MW. Participation in the chronic phase of stroke. Top Stroke Rehabil 2013;20:52-61.
8. Ng SSM, Ho LYW, Chan NH, Liu TW, So B. Psychometric properties of the Chinese version of the Oxford Participation and Activities Questionnaire in people with stroke. Int J Environ Res Public Health 2022;19:15450.
10. Essers B, Meyer S, De Bruyn N, Van Gils A, Boccuni L, Tedesco Triccas L, et al. Mismatch between observed and perceived upper limb function: an eye-catching phenomenon after stroke. Disabil Rehabil 2019;41:1545-51.
11. van Delden AL, Peper CL, Beek PJ, Kwakkel G. Match and mismatch between objective and subjective improvements in upper limb function after stroke. Disabil Rehabil 2013;35:1961-7.
12. Thrane G, Alt Murphy M, Sunnerhagen KS. Recovery of kinematic arm function in well-performing people with subacute stroke: a longitudinal cohort study. J Neuroeng Rehabil 2018;15:67.
13. Essers B, Coremans M, Veerbeek J, Luft A, Verheyden G. Daily life upper limb activity for patients with match and mismatch between observed function and perceived activity in the chronic phase post stroke. Sensors (Basel) 2021;21:5917.
14. Pan H, Ng SSM, Liu TW, Lam SSL, Chan CCC, Li CSK, et al. Self-perceived upper extremity motor function predicts health-related quality of life in chronic stroke survivors. Disabil Rehabil 2025;47:186-93.
15. Rand D, Eng JJ. Disparity between functional recovery and daily use of the upper and lower extremities during subacute stroke rehabilitation. Neurorehabil Neural Repair 2012;26:76-84.
16. Ashford S, Turner-Stokes L, Siegert R, Slade M. Initial psychometric evaluation of the Arm Activity Measure (ArmA): a measure of activity in the hemiparetic arm. Clin Rehabil 2013;27:728-40.
18. Ashford S, Slade M, Malaprade F, Turner-Stokes L. Evaluation of functional outcome measures for the hemiparetic upper limb: a systematic review. J Rehabil Med 2008;40:787-95.
19. Chan NH, Ng SSM. Psychometric properties of the Chinese version of the Arm Activity Measure in people with chronic stroke. Front Neurol 2023;14:1248589.
20. Videler AJ, Beelen A, van Schaik IN, de Visser M, Nollet F. Limited upper limb functioning has impact on restrictions in participation and autonomy of patients with hereditary motor and sensory neuropathy 1a. J Rehabil Med 2009;41:746-50.
21. de Rooij IJM, Riemens MMR, Punt M, Meijer JG, Visser-Meily JMA, van de Port IGL. To what extent is walking ability associated with participation in people after stroke? J Stroke Cerebrovasc Dis 2021;30:106081.
22. Silva SM, Corrêa JCF, Pereira GS, Corrêa FI. Social participation following a stroke: an assessment in accordance with the international classification of functioning, disability and health. Disabil Rehabil 2019;41:879-86.
23. Wolf SL, Catlin PA, Ellis M, Archer AL, Morgan B, Piacentino A. Assessing Wolf Motor Function Test as outcome measure for research in patients after stroke. Stroke 2001;32:1635-9.
24. Lin JH, Hsu MJ, Sheu CF, Wu TS, Lin RT, Chen CH, et al. Psychometric comparisons of 4 measures for assessing upper-extremity function in people with stroke. Phys Ther 2009;89:840-50.
25. Morris DM, Uswatte G, Crago JE, Cook EW 3rd, Taub E. The reliability of the Wolf Motor Function Test for assessing upper extremity function after stroke. Arch Phys Med Rehabil 2001;82:750-5.
27. Ng SS, Hui-Chan CW. The timed up & go test: its reliability and association with lower-limb impairments and locomotor capacities in people with chronic stroke. Arch Phys Med Rehabil 2005;86:1641-7.
28. Chau J, Martin CR, Thompson DR, Chang AM, Woo J. Factor structure of the Chinese version of the Geriatric Depression Scale. Psychol Health Med 2006;11:48-59.
29. Ramström T, Bunketorp-Käll L, Wangdell J. Arm activity measure (ArmA): psychometric evaluation of the Swedish version. J Patient Rep Outcomes 2021;5:39.
30. Lackritz H, Parmet Y, Frenkel-Toledo S, Baniña MC, Soroker N, Solomon JM, et al. Effect of post-stroke spasticity on voluntary movement of the upper limb. J Neuroeng Rehabil 2021;18:81.
31. Adeel M, Peng CW, Lee IJ, Lin BS. Prediction of spasticity through upper limb active range of motion in stroke survivors: a generalized estimating equation model. Bioengineering (Basel) 2023;10:1273.
33. Chen P, Liu TW, Tse MMY, Lai CKY, Tsoh J, Ng SSM. The predictive role of hand section of Fugl-Meyer Assessment and motor activity log in action research arm test in people with stroke. Front Neurol 2022;13:926130.
36. Kim HM. Effects of self-efficacy, self-esteem, and disability acceptance on the social participation of people with physical disabilities: focusing on COVID-19 pandemic. Brain Behav 2023;13:e2824.
37. Nelsone L, Rafsten L, Abzhandadze T, Sunnerhagen KS. A cohort study on anxiety and perceived recovery 3 and 12 months after mild to moderate stroke. Front Neurol 2023;14:1273864.
38. de Bekker A, Geerlings MI, Uitewaal-Poslawsky IE, de Man-van Ginkel JM. Depression in stroke survivors: ten-year follow-up. Determinants of the natural course of depressive symptoms in stroke survivors in the Netherlands: the SMART-Medea study. J Stroke Cerebrovasc Dis 2022;31:106272.
39. Taub E, Uswatte G, Mark VW, Morris DM, Barman J, Bowman MH, et al. Method for enhancing real-world use of a more affected arm in chronic stroke: transfer package of constraint-induced movement therapy. Stroke 2013;44:1383-8.
Contribution of Perceived Upper Limb Function to the Participation and Activity Levels Among Community-Dwelling People With Chronic Stroke
Fig. 1. Scatter plots showing the correlations of (A-C) age, (D-F) post-stroke duration, (G-I) TUG, (J-L) GDS, (M-O) WMFT, (P-R) ArmA (active function), and (S-U) ArmA (passive function) with the Ox-PAQ. TUG, Timed Up and Go Test; GDS, Geriatric Depression Scale; WMFT, Wolf Motor Function Test; ArmA, Arm Activity Measure; Ox-PAQ, Oxford Participation and Activity Questionnaire.
Graphical abstract
Fig. 1.
Graphical abstract
Contribution of Perceived Upper Limb Function to the Participation and Activity Levels Among Community-Dwelling People With Chronic Stroke
Characteristic
Value (N=81)
Age (yr)
63.16±6.28
Sex
Male
49 (60.49)
Female
32 (39.51)
Body mass index (kg/m2)
24.06±3.01
Post-stroke duration (yr)
6.82±4.30
Cause of stroke
Ischaemic
58 (71.60)
Haemorrhagic
23 (28.40)
Hemiplegic side
Left
36 (44.44)
Right
45 (55.56)
TUG time (s)
18.05±14.96
GDS score
4.74±3.97
WMFT score
47.10±21.06
ArmA score (passive function)
2.86±3.62
ArmA score (active function)
27.22±15.98
Ox-PAQ score
Routine activities subscale
16.45±16.77
Social engagement subscale
14.81±20.62
Emotional well-being subscale
16.60±18.81
Model No.
Independent variables
R2 (R2 adjusted)
R2 change
B (standard error)
ꞵ
p-value
Model 1
0.29 (0.28)
0.29
TUG
0.51 (0.11)
0.46
<0.001**
GDS
1.13 (0.40)
0.27
0.006**
Model 2
0.34 (0.32)
0.05
TUG
0.38 (0.12)
0.34
0.002**
GDS
1.11 (0.39)
0.26
0.006**
WMFT
-0.20 (0.08)
-0.25
0.021*
Model 3
0.43 (0.40)
0.09
TUG
0.40 (0.11)
0.36
0.001**
GDS
0.95 (0.37)
0.22
0.012*
WMFT
0.10 (0.12)
0.12
0.406
ArmA (active function)
0.50 (0.14)
0.47
0.001**
Model 4
0.54 (0.50)
0.07
TUG
0.37 (0.10)
0.33
<0.001**
GDS
0.71 (0.34)
0.17
0.041*
WMFT
0.09 (0.11)
0.12
0.372
ArmA (active function)
0.30 (0.14)
0.28
0.036*
ArmA (passive function)
1.78 (0.44)
0.38
<0.001**
Model No.
Independent variables
R2 (R2 adjusted)
R2 change
B (standard error)
ꞵ
p-value
Model 1
0.11 (0.08)
0.11
TUG
0.39 (0.15)
0.28
0.011*
GDS
0.78 (0.56)
0.15
0.167
Model 2
0.14 (0.10)
0.03
TUG
0.26 (0.17)
0.19
0.129
GDS
0.76 (0.55)
0.15
0.174
WMFT
-0.19 (0.12)
-0.20
0.107
Model 3
0.21 (0.17)
0.07
TUG
0.29 (0.16)
0.21
0.082
GDS
0.58 (0.54)
0.11
0.283
WMFT
0.13 (0.17)
0.13
0.453
ArmA (active function)
0.54 (0.21)
0.42
0.011*
Model 4
0.28 (0.23)
0.07
TUG
0.25 (0.16)
0.18
0.108
GDS
0.34 (0.52)
0.07
0.519
WMFT
0.13 (0.16)
0.13
0.442
ArmA (active function)
0.33 (0.21)
0.26
0.121
ArmA (passive function)
1.81 (0.67)
0.32
0.009**
Model No.
Independent variables
R2 (R2 adjusted)
R2 change
B (standard error)
ꞵ
p-value
Model 1
0.17 (0.16)
0.17
GDS
1.98 (0.49)
0.42
<0.001**
Model 2
0.23 (0.21)
0.06
GDS
1.92 (0.47)
0.41
<0.001**
WMFT
-0.22 (0.09)
-0.24
0.018*
Model 3
0.25 (0.23)
0.02
GDS
1.83 (0.47)
0.39
<0.001**
WMFT
-0.06 (0.14)
-0.06
0.681
ArmA (active function)
0.28 (0.18)
0.23
0.132
Model 4
0.32 (0.29)
0.07
GDS
1.61 (0.46)
0.34
<0.001**
WMFT
-0.05 (0.13)
-0.05
0.721
ArmA (active function)
0.09 (0.19)
0.08
0.617
ArmA (passive function)
1.64 (0.59)
0.32
0.007**
Table 1. Characteristics of the participants
Values are presented as mean±standard deviation or number (%).
TUG, Timed Up and Go Test; GDS, Geriatric Depression Scale; WMFT, Wolf Motor Function Test; ArmA, Arm Activity Measure; Ox-PAQ, Oxford Participation and Activity Questionnaire.
Table 2. Multiple linear regression analyses using the routine activities subscale of the Ox-PAQ
Ox-PAQ, Oxford Participation and Activity Questionnaire; GDS, Geriatric Depression Scale; WMFT, Wolf Motor Function Test; ArmA, Arm Activity Measure.
p<0.05.
p<0.01.
Table 3. Multiple linear regression analyses using the social engagement subscale of the Ox-PAQ
Ox-PAQ, Oxford Participation and Activity Questionnaire; TUG, Timed Up and Go Test; GDS, Geriatric Depression Scale; WMFT, Wolf Motor Function Test; ArmA, Arm Activity Measure.
p<0.05.
p<0.01.
Table 4. Multiple linear regression analyses using the emotional well-being subscale of the Ox-PAQ
Ox-PAQ, Oxford Participation and Activity Questionnaire; TUG, Timed Up and Go Test; GDS, Geriatric Depression Scale; WMFT, Wolf Motor Function Test; ArmA, Arm Activity Measure.
