Characterization of muscle strength using the strength domain of the stroke impact scale: An integrative review

Abstract

BACKGROUND AND OBJECTIVE:

Muscle weakness is among the most common and obvious impairments in older adults and individuals with neurologic disorders. Although impairments in muscle strength are typically characterized using performance measures, the impairments have also been described using patient or observer-report. The objective of this review was to summarize literature describing use of a patient-report instrument, the Strength Domain (SD) of the Stroke Impact Scale (SIS), to grade strength impairments.

METHODS:

Peer-reviewed literature reporting SD scores for the SIS was identified using computerized searches of the CINAHL, PubMed, and Scopus databases followed by hand searches. Potentially relevant articles were then mined for data on the participants tested, the SIS version used, scores documented, and clinimetric properties reported.

RESULTS:

Sixty-five articles were judged appropriate based on inclusion and exclusion criteria. The articles involved more than 7000 residents of 22 countries. All articles focused on individuals with stroke (usually chronic), although one also included community-dwelling adults without stroke. The SIS version used was frequently unreported, but 3.0 was version most often specified. For articles reporting SD scale scores the mean ranged from 19.7 to 85.5. Construct (known groups, convergent, and discriminant) validity of the SD was supported by the literature as was its internal consistency and test-retest reliability. Responsiveness of the SD was evinced by numerous studies showing increases in SD scores with time or accompanying effective interventions. However, only one study indicated responsiveness using an anchor-based statistic.

CONCLUSIONS:

The SD of the SIS is a well-established and mostly clinimetrically sound patient-report measure of paretic limb strength among individuals with stroke. Its use with individuals with weakness accompanying aging or diagnoses other than stroke remains to be substantiated.

Keywords

Muscle strength measurement clinimetrics

1. Introduction

Muscle strength, a muscle power function according to the International Classification of Functioning, Disability and Health (ICF), is well established as an important body function [1]. Numerous options for measuring muscle strength have been described in the literature. Although performance measures are by far the most commonly employed, patient- and observer-report measures have also been used. Bohannon recently reviewed the scores and clinimetric properties of several different patient-report measures of strength [2]. Intentionally, his review did not address what may be the most widely used patient-report measure of strength, that is, the strength domain (SD) of the Stoke Impact Scale (SIS).

The SD of the SIS requires patients to rate their strength over the past week in the: a) arm that was most affected by their stroke, b) hand grip that was most affected by their stroke, c) leg that was most affected by their stroke, and d) foot/ankle that was most affected by their stroke. The strength of each of the 4 aforementioned body regions is rated as: 5 (a lot of strength), 4 (quite a bit of strength), 3 (some strength), 2 (a little strength), and 1 (no strength at all). The total SD score (range 4 to 20) is often converted to a scale score which can range from 0 to 100 [3].

As no critical review of the SD of the SIS exists to guide clinicians in decisions regarding its use, the current review was undertaken to summarize the literature reporting use of the SD of the SIS to describe strength impairments. Of particular interest were the specifics of use, scores reported, and the clinimetric properties of the SD.

2. Methods

Peer-reviewed articles providing information on the SD of the SIS were sought by the author via computerized searches of 3 bibliographic databases (CINAHL, PubMed, and Scopus) as well as hand-searches on September 17, 2020. The search start date was January 1, 1990. The search string employed in all searches was ““Stroke Impact Scale” AND “Strength.”” Only articles including information specific to the SD of the SIS were included. Articles were excluded if they were reviews, case studies, protocols, meeting abstracts, or written in a language other than English (unless adequately translatable by Google Translate). Articles were also excluded if they did not include descriptive statistics for scores of the SD.

Potentially relevant articles addressing the SD of the SIS were mined for information on populations tested, the SIS version used, and findings regarding the SD of the SIS. Findings regarding the SD focused on SD scores and specifics relevant to clinimetrics of the SD (i.e., validity, reliability, and responsiveness).

3. Results

Figure 1 illustrates how the final sample of 65 experimental and non-experimental studies was derived from the 147 non-duplicative potentially relevant articles identified by the searches. The 65 summarized articles included greater than 7000 participants from 22 countries, with the United States, Taiwan, and Sweden contributing the largest number. All articles focused on individuals with stroke (usually chronic), although one also included community-dwelling adults without stroke [38]. The specific SIS version used was often unreported. When stipulated, the most common version was 3.0. Thirteen studies denoted non-English language versions of the SIS [10, 11, 13, 15, 16, 17, 20, 22, 44, 46, 47, 49, 50]. All studies reported SD scores based on patient-report, but 2 studies included SD scores obtained from proxy-report [15, 24] as well. Per design, all included studies reported descriptive statistics for SD scores. However, scores in 2 studies were limited to the “arm” and “hand” items of the SD [9, 19]. Most studies reported SD scale scores- the mean of which ranged from 19.7 to 85.5. One study presented only the standard deviation (19.3) of SD scale scores [43].

Figure 1.

Flow chart summarizing how the ultimate sample of articles was determined.

Clinimetric support for the SD of the SIS was found to be mostly supportive. This support incorporates construct validity, reliability, and responsiveness.

Three subtypes of construct validity were supported by the literature reviewed. The first subtype of validity supported was known groups validity; herein, defined as the ability of the SD scores to distinguish between groups expected to differ in limb strength. Among the articles summarized in this review, significant differences in SD scores were found between hierarchical categories (groups) of the Modified Rankin Scale (disability) [15, 16, 17, 61], Barthel Index (activities of daily living) [13, 15, 17, 67], National Institutes of Health Stroke Severity Scale (NIHSS) [13], Stroke Upper Limb Capacity Scale (upper limb mobility) [10], gait speed [35], and Activity Card Sort (physical activity participation) [63]. The second subtype of validity supported by the literature was convergent validity; herein defined as the degree to which SD scores of the SIS were related to other measures with which they theoretically should be related. SD scores would certainly be expected to be, and were, correlated significantly with other measures of muscle strength. Specifically, SD scores were found to correlate significantly and strongly with NIHSS Arm and Hand Motor Scores ( $r_{s}=$ 0.76 & 0.80) [9] and dynamometer measurements ( $r_{s}=$ 0.70 to 0.85) [9], as well as significantly and moderately with Motricity Index scores ( $r=$ 0.64 to 0.69) [24, 58]. Adding to this evidence were articles showing significant and mostly moderate correlations between SD scores and other measures of physical status. Such measures included SIS domain scores for activities of daily living/ instrumented activities of daily living ( $r_{s}=$ 0.60 to 0.66) [13, 25], mobility ( $r_{s}=$ 0.60 to 0.67) [13, 25], hand function ( $r_{s}=$ 0.70 to 0.73) [13, 25], and the SIS summary score for physical function ( $r_{s}=$ 0.40 to 0.84) [13]. Also included were Barthel Index scores ( $r_{s}=$ 0.37 to 0.54) [13, 15, 16, 20], Duke Health Profile Physical Health scores ( $r_{s}=$ 0.62) [13], NIHSS motor and total scores ( $r_{s}=-$ 0.44 to $-$ 0.65) [3, 16, 18, 20], Physical Domain of World Health Organization Quality of Life Scale scores ( $r=$ 0.40) [25], Short-form 36 Physical Function ( $r=$ 0.36 and $r_{s}=$ 0.55) [16, 20] and Physical Component Summary scores ( $r_{s}=$ 0.40 to 0.50) [16, 20], Lawton Instrumented Activities of Daily Living scores ( $r_{s}=$ 0.43) [16], Modified Rankin Scale scores ( $r_{s}=-$ 0.38 to $-$ 0.53) [16, 54], Berg Balance Scale scores ( $r=$ 0.54) [18], Fugl-Meyer limb and total scores ( $r=$ 0.54 to 0.72) [3, 18], Functional Independence Measure scores ( $r=$ 0.51) [18], 6 minute walk distance ( $r=$ 0.52) [46], and gait speed ( $r=$ 0.64) [35]. Notably, the last 2 measures were not fully supported as evidence of convergent validity; Cheng et al. reported only 2 of 4 correlations ( $r=$ 0.39 to 0.48) between SD scores and 6 minute walk distance to be significant and no correlations ( $r=$ 0.27 and 0.29) between SD scores and gait speed to be significant [19].

The third subtype of validity supported was discriminant validity; that is, the extent to which SD scores of the SIS were unrelated to other measures to which they would not, in theory, be expected to be related. In this review, measures classified as theoretically unrelated were those clearly nonphysical in nature. The measures and their mostly very weak and insignificant correlations with the SD were the Memory ( $r=$ 0.16) and Communication ( $r=$ 0.14) domains of the SIS [25]; the Social Function ( $r_{s}=-$ 0.01), Role Emotional ( $r=$ 0.08), and Mental Health ( $r=$ 0.15) dimensions [20], and the Mental Component Summary of the SF-36 ( $r=$ 0.03 & $r_{s}=$ 0.11) [16, 20]; depression as measured by the Global Depression Scale ( $r_{s}=-$ 0.06) [16] and the Zung Depression Scale ( $r=-$ 0.14) [25]; World Health Organization Quality of Life Psychological ( $r=$ 0.20) and Social Relationships ( $r=$ 0.13) domains [25]; and Mini Mental State Examination scores ( $r_{s}=$ 0.15) [16]. These findings notwithstanding, Carod-Artal et al. reported significant (albeit very low or low) correlations between SIS SD scores and the Social Function ( $r_{s}=$ 0.20), Mental Health ( $r_{s}-$ 0.29), and Role Emotional ( $r_{s}=$ 0.19) dimensions of the SF-36 and Hospital

Table 1

Summary of studies included in review

Study	Participants	Stroke impact scale	Findings
Ahmad et al. (2019) [4]	Malaysians with chronic stroke assigned to/completed virtual reality ( $n=$ 20/18) or standard therapy ( $n=$ 20/18) interventions	Version not specified	Scores: SD scale scores ranged from a mean 57.3 to 68.7 depending on intervention & time.
Responsiveness: SD scale scores $\mathrm{\uparrow}$ SGNF over 8 weeks but $\mathrm{\Delta}$ not SGNF different between interventions.
Allen et al. (2014) [5]	Canadians with chronic stroke enrolled in/completed home rehabilitation intervention ( $n=$ 794/140)	Version not specified	Scores: SD scale scores ranged from a mean 56.1 to 67.1 depending on time.
Responsiveness: SD scale scores $\mathrm{\uparrow}$ SGNF between intake & discharge.
Badke et al. (2011) [6]	Americans with chronic stroke enrolled in/completed biofeedback ( $n=$ 36/24) intervention	Version 2.0	Scores: SD scale scores were a mean 63.8 or 66.8 depending on time.
Responsiveness: SD scale scores $\mathrm{\uparrow}$ over 8 wk, albeit not SGNF.
Barclay-Goddard (2011) [7]	Canadians with chronic stroke enrolled in/followed ( $n=$ 677/ $n=$ 606 @ 1 mo, $n=$ 527 @ 3 mo, $n=$ 478 @ 6 mo, $n=$ 442 @ 12 mo)	Version not specified	Scores: SD scale scores ranged from a mean 64.8 to 73.9 depending on time.
Validity: SD factor loading on self care $=$ 0.80.
Responsiveness: SD scale scores $\mathrm{\uparrow}$ between 1 mo and 12 mo.
Barry et al. (2012) [8]	Americans with chronic stroke assigned to/completied wrist-hand orthosis ( $n=$ 11/10) or manually assisted ( $n=$ 11/9) interventions	Version 3.0	Scores: SD scale scores ranged from a mean 34.8 to 50.8 depending on time & intervention.
Responsiveness: SD scale scores $\mathrm{\uparrow}$ in both groups albeit not SGNF.
Bohannon (2004) [9]	Americans with subacute stroke ( $n=$ 26)	Version not specified (only arm & hand items of SD measured)	Scores: SD item scores for arm & hand ranged from 1 to 4.
Validity: SD arm & hand scores correlated SGNF with NIHSS arm scores ( $r_{s}=-$ 0.76 & $-$ 0.80) & dynamometer scores for grip ( $r_{s}=$ 0.70 & 0.85), elbow flexion ( $r_{s}=$ 0.76 & $r=$ 0.70) & shoulder abduction ( $r_{s}=$ 0.84 to 0.75).
Branco et al. (2017) [10]	Portuguese with chronic stroke ( $n=$ 122)	Portuguese version	Scores: SD scale scores ranged from 12.5 to 100 (mean $=$ 41.2).
Validity: SD scale scores SGNF different between 3 categories of Stroke Upper Limb Capacity Scale scores.
Brandão et al. (2018) [11]	Brazilians with stroke enrolled/completed reliability study ( $n=$ 52/40)	Portuguese Version 2.0	Scores: SD scale scores were a mean 44.4 & 44.8 on 2 days 15 days apart.
Reliability: SD item scores internally consistent ( $\alpha=$ 0.86). Test-retest ICC for SD $=$ 0.86. Inter-tester ICC for SD $=$ 0.92.
Byun et al. (2020) [12]	Americans with chronic stroke ( $n=$ 100)	Version not specified	Scores: SD scale scores ranged from 0 to 100 (mean $=$ 64.3).
Validity: SD scale scores not correlated SGNF with sleep related impairment ( $r=-$ 0.19).
Caël et al. (2015) [13]	French with acute, subacute, or chronic stroke ( $n=$ 288)	French version 2.0	Scores: SD scale scores ranged from a mean 35.1 to 74.1 depending on patient group.
Validity: SD scale scores SGNF different between scoring categories of Barthel Index, Boston Diagnostic Aphasia Examination, professional activity & NIHSS. SD scale scores correlated SGNF with all other SIS domain scores ( $r_{s}=$ 0.33 to 0.73) & Physical Function Score ( $r_{s}=$ 0.84). The SD scale score also correlated with the Hospital Anxiety & Depression Scale (HADS) ( $r_{s}=$ $-$ 0.25 & $-$ 0.37), Barthel Index ( $r_{s}=$ 0.54) & Duke Health Profile domains ( $r_{s}=-$ 0.29 to 0.62).
Reliability: SD item scores internally consistent ( $\alpha=$ 0.94). Test-retest ICC for SD $=$ 0.71.
Responsiveness: SD scores $\mathrm{\uparrow}$ over 3 mo, SGNF more in patients whose mood improved (standardized response mean $=$ 0.67) but not SGNF more in patients whose physical health improved (standardized response mean $=$ 0.43).
Caleffi Segura et al. (2006) [14]	Brazilians with subacute & chronic acute stroke included/completed ( $n=$ 19/18) conventional intervention	Version not specified	Scores: SD scale scores were a mean 34.4 or 63.9 depending on time.
Responsiveness: SD scores $\mathrm{\uparrow}$ SGNF over 3 mo.

Table 1, continued
Study	Participants	Stroke impact scale	Findings
Carod-Artal et al. (2009) [15]	Brazilians with chronic stroke ( $n=$ 180) & proxies ( $n=$ 180)	Brazilian Version 3.0	Scores: SD scale scores ranged from a mean 25.6 to 74.6 depending on reporter (proxy or patient) & NIHSS & BI category.
Validity: SD scale scores for patients SGNF ${>}$ for proxies. SD scale scores SGNF different for MRS & BI categories. SD scale scores correlated SGNF with BI scores ( $r_{s}=$ 0.52).
Reliability: Proxy scored SD item scores internally consistent ( $\alpha=$ 0.82). Stroke vs proxy ICC for SD $=$ 0.61, $K_{w}$ for SD items $=$ 0.39 to 0.60.
Responsiveness: For proxy scored SD scale scores standard error of measurement (SEM) $=$ 10.1, smallest real difference (SRD) $=$ 27.9.
Carod-Artal et al. (2008) [16]	Brazilians with chronic stroke ( $n=$ 174)	Brazilian Version 3.0	Scores: SD scale scores ranged from 0 to 100 (mean $=$ 46.7).
Validity: SD scale scores were SGNF different between MRS categories. SD scale score correlated SGNF with all dimensions of the SF-36 ( $r_{s}=$ 0.19 to 0.55) except Vitality ( $r_{s}=$ 0.14). Correlation with the Physical Function Scale of the SF-36 was highest ( $r_{s}=$ 0.55). The SD scale score correlated SGNF with the SF-36 Physical Component Summary ( $r_{s}=$ 0.50) but not the SF-36 Mental Component Summary ( $r_{s}=$ 0.11). The SD scale score correlated SGNF with the NIHSS ( $r_{s}=-$ 0.56), Barthel Index ( $r_{s}=$ 0.53), Lawton IADL ( $r_{s}=$ 0.43), MRS ( $r_{s}=-$ 0.53) & HADS ( $r_{s}=-$ 0.20 & $-$ 0.24) scores but not with the Global Depression Scale ( $r_{s}=-$ 0.06) or Mini Mental State Examination (MMSE) ( $r_{s}=$ 0.15).
Reliability: SD item scores internally consistent ( $\alpha=$ 0.81). Test-retest ICC for SD $=$ 0.81. Test-retest coefficient ( $K_{W}$ ) for individual SD items $=$ 0.56 to 0.75.
Responsiveness: For SD scale scores SEM $=$ 11.4 & SRD $=$ 31.7.
Carod-Artal et al. (2009) [17]	Brazilians with chronic stroke ( $n=$ 260)	Brazilian Version 3.0	Scores: SD scale scores ranged from 30.3 to 69.7 depending on MRS, BI, & HAD categories.
Validity: SD scale scores SGNF different between categories of MRS, BI & HADS. SD scale scores explained best by MRS, BI &HADS scores ( $R^{2}=$ 0.41).
Chen et al. (2015) [18]	Taiwanese with subacute stroke ( $n=$ 119)	Version 3.0	Scores: Mean SD score $=$ 8.7(?).
Validity: SD score correlated SGNF with time since onset ( $r=-$ 0.21), MMSE ( $r=$ 0.42), NIHSS ( $r=-$ 0.65), Berg Balance Scale ( $r=$ 0.54), Fugl-Meyer upper limb ( $r=$ 0.60) & lower limb ( $r=$ 0.54) assessments & Functional Independence Measure (FIM) ( $r=$ 0.51) scores. The SD score was not SGNF correlated with age ( $r=$ 0.01) or Beck Depression Inventory scores ( $r=-$ 0.16) SD scores explained best by NIHSS and FIM scores and time since onset ( $R^{2}=$ 0.54).
Cheng et al. (2020) [19]	Canadians with acute, subacute, or chronic stroke enrolled/completed test-retest ( $n=$ 21/20)	Version not stated	Scores: SD score (?) was a mean 3.1 for arm & 3.3 for hand.
Validity: SD scores correlated SGNF or close to SGNF with 6 min walk test performance at baseline & retest ( $r=$ 0.39 to 0.48) but not 10 m walk test performance ( $r=$ 0.27 & 0.29).
Choi et al. (2017) [20]	Koreans with subacute or chronic stroke ( $n=$ 70)	Korean Version 3.0	Scores: SD scale scores were a mean 27.0 & 27.8 on repeated days 14 days apart.
Validity: SD scale scores correlated SGNF with the Physical Function ( $r=$ 0.36), Bodily Pain ( $r_{s}=$ 0.32) & General Health ( $r_{s}=$ 0.28) scales & Physical Component Summary ( $r_{s}=$ 0.40) of the SF-36 but not with the Role Physical ( $r_{s}=$ 0.17), Vitality ( $r_{s}=$ 0.150), Social Function ( $r_{s}=-$ 0.01), Role Emotional ( $r=$ 0.08) & Mental Health ( $r_{s}=$ 0.15) scales or the Mental Component Summary ( $r_{s}=$ 0.03) of the SF-36. The SD scores correlated SGNF with the Modified BI ( $r_{s}=$ 0.37) & NIHSS ( $r_{s}=-$ 0.44) but not the MMSE ( $r_{s}=-$ 0.02) or HADS ( $r_{s}=-$ 0.14 & $-$ 0.16).
Reliability: SD item scores internally consistent ( $\alpha=$ 0.83 & 0.86). Test-retest coefficient ( $r_{s}$ ) for SD $=$ 0.57.

Table 1, continued
Study	Participants	Stroke impact scale	Findings
Cook et al. (2020) [21]	Swedes with chronic stroke enrolled & divided into groups based on activity level: inactive ( $n=$ 47) & active ( $n=$ 26)	Version 3.0	Scores: Raw SD scores categorized as problematic if 1 to 3 ( $n=$ 26) & nonproblematic if 4 to 5 ( $n=$ 47).
Validity: Problematic SD scores SGNF more common in inactive patients.
Dettmers et al. (2005) [22]	Germans with chronic stroke enrolled/completed ( $n=$ 12/11) study	German Version	Scores: SD score (?) ranged from a mean 3.3 to 3.6 depending on time.
Responsiveness: Relative to baseline, SD scores $\mathrm{\uparrow}$ SGNF over the course of treatment.
Dobkins et al. (2014) [23]	Americans with chronic stroke ( $n=$ 408)	Version not stated	Scores: SD scale score ranged from a mean 40.3 to 53.0 depending on time & status as responder to intervention.
Validity: SD responder scores SGNF ${>}$ nonresponder scores @ 12 mo.
Responsiveness: SD scores $\mathrm{\uparrow}$ over 12 months, but with no SGNF difference between responders and nonresponders.
Duncan et al. (2002) [24]	Americans with subacute stroke ( $n=$ 287) & proxies ( $n=$ 287)	Version 16	Scores: SD scale scores were a mean 61.9 for patients & 57.2 for proxies.
Validity: SD scale scores SGNF different between Rankin scores for patients & proxies. SD scale scores correlated SGNF with Motricity Index Scores of patients ( $r=$ 0.67) & proxies ( $r=$ 0.69).
Reliability: Stroke vs proxy ICC for SD $=$ 0.62.
Duncan et al. (1999) [3]	Americans with mild/moderate chronic stroke enrolled in/completed ( $n=$ 91/91 @ 1 mo, $n=$ 80 @ 3 mo, $n=$ 69 @ 6 mo) study	Version 2.0	Scores: SD scale scores ranged from a mean 47.5 to 72.2 depending on MRS, time since onset & stroke severity.
Validity: SD scale scores SGNF different between Rankin category scores & correlated SGNF with NIH Motor scores ( $r=-$ 0.59) & total Fugl-Meyer scores ( $r=$ 0.72).
Responsiveness: For patients with moderate strokes SD scores $\mathrm{\uparrow}$ SGNF between 1 & 3 mo & between 1 & 6 mo, but not between 3 & 6 mo. For patients with minor strokes SD scale scores did not $\mathrm{\uparrow}$ SGNF over any time period.
Edwards & O’Connell (2003) [25]	Australians with chronic stroke ( $n=$ 74)	Version 2.0 & 16	Scores: SD scale scores ranged from 0 to 100 (mean 52.9).
Validity: SD scale scores correlated SGNF with SIS domain scores of Emotion ( $r=$ 0.27), ADL/IADL ( $r=$ 0.60), Mobility ( $r=$ 0.60), Hand Function ( $r=$ 0.70), & Participation ( $r=$ 0.42) & SIS-16 score ( $r=$ 0.65) but not SIS domain scores of Memory ( $r=$ 0.16) or Communication ( $r=$ 0.14). SD scale scores correlated SGNF with Physical ( $r=$ 0.40) & Environment ( $r=$ 0.34) domains of World Health Organization Quality of Life Scale but not the Psychological ( $r=$ 0.20) or Social Relationships ( $r=$ 0.13) domains of the Scale or Zung Depression Scale ( $r=-$ 0.14).
Reliability: SD item scores internally consistent ( $\alpha=$ 0.90).
Essers et al. (2018) [26]	Belgians with chronic stroke ( $n=$ 32)	Version not stated	Scores: SD scale scores ranged from a median 36.8 to 78.9 depending on category of perceived function.
Validity: SD scale scores differed SGNF between patients with different observed & perceived function.
Fulk et al. (2020) [27]	Americans with chronic stroke included in study ( $n=$ 408 @ 2 mo, $n=$ 380 @ 6 mo, $n=$ 360 @ 12 mo)	Version not stated	Scores: SD scale scores ranged from a median 37.5 to 62.5 depending on time and sleep problems group.
Validity: SD scale scores SGNF less in patients with sleep problems.
Guidetti et al. (2014) [28]	Swedes with chronic stroke enrolled/included in study ( $n=$ 204, $n=$ 201 @ 3 mo, $n=$ 203 @ 12 mo)	Version 3.0	Scores: SD scale scores were a mean 69 or 71 depending on time.
Responsiveness: SD scores $\mathrm{\uparrow}$ SGNF between 3 & 12 mo, but $\mathrm{\uparrow}$ largely in patients with moderate/severe stroke.
Hsieh et al. (2017) [29]	Taiwanese with subacute stroke assigned to bilateral priming $+$ task-oriented ( $n=$ 16) or task-oriented ( $n=$ 15) intervention	Version 3.0	Scores: SD scale scores were a mean 27.5 to 45.3 depending on time & group.
Responsiveness: SD scores $\mathrm{\uparrow}$ but $\mathrm{\uparrow}$ in group receiving priming SGNF ${>}$ other group.

Table 1, continued
Study	Participants	Stroke impact scale	Findings
Huang et al. (2010) [30]	Taiwanese with chronic stroke ( $n=$ 58)	Version 3.0	Scores: SD scale score was a mean 40.7 @ baseline.
Responsiveness: SD scale scores $\mathrm{\uparrow}$ in 60.3% of patients after constraint -induced intervention.
Hung et al. (2016) [31]	Taiwanese with chronic stroke assigned to robot $+$ task-specific ( $n=$ 11) or robot $+$ impairment-oriented ( $n=$ 10) intervention	Version 3.0	Scores: SD scale scores ranged from a mean 35.6 to 53.4 depending on time & intervention group.
Responsiveness: SD scores $\mathrm{\uparrow}$ after intervention but not SGNF.
Hung et al. (2019) [32]	Taiwanese with chronic stroke assigned to/completed follow-up for unilateral hybrid therapy ( $n=$ 14/14), bilateral hybrid therapy ( $n=$ 16/15) or robot assisted therapy ( $n=$ 15/13) intervention	Version 3.0	Scores: SD scale scores ranged from 32.5 to 52.1 depending on time & intervention.
Responsiveness: SD scores $\mathrm{\uparrow}$ SGNF after intervention albeit not differently for interventions.
Kamel et al. (2010) [33] ${}^{3}$	Egyptians ( $n=$ 50) with subacute stroke	Version 2.0	Scores: SD scale scores were a mean 54.0 or 57.1 depending on time.
Reliability: SD item scores internally consistent ( $\alpha=$ 0.71).
Responsiveness: SD scores $\mathrm{\uparrow}$ SGNF between 1 & 3 mo.
Kamwesiga et al. (2018) [34]	Ugandans with acute, subacute & chronic stroke ( $n=$ 120)	Version 3.0	Scores: Among 62 patients in rehabilitation unit SD scale scores ranged from 0 to 100 (mean $=$ 44.7).
Khanittanuphong & Tipchatyotin (2017) [35]	Thais with chronic stroke ( $n=$ 92)	Version 3.0	Scores: SD scale scores ranged from a mean 29.4 to 68.9 depending on gait speed.
Validity: SD scores SGNF different between gait speed categories. SD scores correlated SGNF with gait speed ( $r=$ 0.64).
Kim & Park (2019) [36]	Koreans with chronic stroke assigned to/completed bilateral occupational-based ( $n=$ 10/10) or task-oriented ( $n=$ 10/10) interventions	Version not stated	Scores: SD scores (?) ranged from a mean 9.6 to 14.1 depending on time & group.
Responsiveness: SD scores $\mathrm{\uparrow}$ SGNF for both groups between pre-test and post-test.
Kirac-Unal et al. (2018) [37]	Turks with acute stroke assigned to/completed conventional ( $n=$ 13/11) or EMG-triggered electrical stimulation ( $n=$ 14/12) intervention	Turkish version 3.0	Scores: SD scale scores ranged from a mean 37.0 to 62.0 depending on group & time.
Responsiveness: SD scores in both groups $\mathrm{\uparrow}$ over time but $\mathrm{\uparrow}$ not SGNF different.
Lai et al. (2002) [38]	Americans with mild chronic stroke ( $n=$ 81) or without stroke living in community ( $n=$ 66 & 180)	Version 16 with modification for individuals without stroke	Scores: SD scale scores a mean 73 for stroke sample, 70 & 74 for nonstroke samples & 72 for diabetic & nondiabetic samples.
Validity: SD scores not SGNF different between strokes & nonstrokes or between diabetics & nondiabetics.
Lee et al. (2017) [39]	Americans with chronic stroke assigned to control ( $n=$ 8) or day rehabilitation ( $n=$ 9) intervention	Version not stated	Scores: SD scores (?) ranged from a median 10.5 to 16.0 depending on group & time.
Responsiveness: SD scores $\mathrm{\uparrow}$ for rehabilitation group but not the control group. Effect size for difference in pre- & post- SD scores was 0.04 for control & 0.62 for rehabilitation groups.
Lewthwaite et al. (2018) [40]	Americans with chronic stroke assigned to/completed usual therapy ( $n=$ 122/96), usual occupational therapy ( $n=$ 120/100), or skill acquisition ( $n=$ 119/104) interventions	Version 3.0	Scores: SD scale scores ranged from a mean 54.3 to 69.5 depending on group and time.
Responsiveness: SD scores $\mathrm{\uparrow}$ SGNF during the treatment phase. MCID of SD score reported as 9.2.
Lieshout et al. (2020) [41]	Dutch with chronic stroke enrolled/completed ( $n=$ 250/242) 1 of 2 interventions	Version 3.0	Scores: SD scale scores were a mean 51.8 @ baseline & 59.6 @ follow-up.
Responsiveness: SD scores $\mathrm{\uparrow}$ SGNF between baseline & end of 12 wk intervention
Lin et al. (2009) [42]	Taiwanese with chronic stroke assigned to/completed conventional ( $n=$ 20/20), bilateral upper limb ( $n=$ 20/20), or constraint induced ( $n=$ 20/20) interventions	Version 3.0	Scores: SD scale scores ranged from a mean 39.7 to 57.2 depending on group & time.
Responsiveness: SD scores $\mathrm{\uparrow}$ between pretreatment & posttreatment.
Lin et al. (2010) [43]	Taiwanese with chronic stroke ( $n=$ 74) assigned to/completed conventional, bilateral upper limb, or constraint induced interventions	Version 3.0	Scores: SD scale score standard deviation was 19.3.
Reliability: Test-retest ICC $=$ 0.80.
Responsiveness: SD scores $\mathrm{\uparrow}$ between pretreatment & posttreatment. SEM $=$ 8.7, MDC ${}_{\text{95\%}}$ $=$ 24.0.

Table 1, continued
Study	Participants	Stroke impact scale	Findings
Lou et al. (2020) [44]	Chinese with chronic stroke & diabetes assigned to/completed control ( $n=$ 37/30) or diabetes management ( $n=$ 38/33) nterventions	Chinese version	Scores: SD scale scores ranged from a mean 48.3 to 70.8 depending on group & time.
Responsiveness: SD scores $\mathrm{\uparrow}$ SGNF more in diabetes management group.
Mühr et al. (2017) [45]	Swedes with chronic stroke assigned to/retained in reperfusion ( $n=$ 31/31) & reference ( $n=$ 31/26) groups based on history	Version not stated	Scores: SD scale scores were a median 81.3 & 56.3 depending on group.
Validity: SD scores were SGNF higher in reperfusion group.
Muren et al. (2008) [46]	Norwegians with mild chronic stroke ( $n=$ 30)	Norwegian version 3.0	Scores: SD scale scores ranged from 18.8 to 100 (mean $=$ 65.0).
Validity: SD scores correlated SGNF with 6 min walk distance ( $r=$ 0.52).
Ochi et al. (2017) [47]	Japanese with chronic stroke ( $n=$ 32)	Japanese version 3.0	Scores: SD scale score was a mean 32.3 on 2 test days.
Reliability: SD item scores internally consistent ( $\alpha=$ 0.82). Test-retest ICC for SD $=$ 0.87. Test-retest coefficient ( $K_{W}$ ) for individual SD items $=$ 0.62 to 0.70.
Palmcrantz et al. (2012) [48]	Swedes with chronic stroke ( $n=$ 192)	Version 3.0	Scores: SD scale scores were a median 84 for age $\mathrm{<}$ 65 y & 75 for age $\mathrm{>}$ 65 y.
Validity: SD scores for younger group SGNF $\mathrm{>}$ for older group.
Park et al. (2019) [49]	Koreans with chronic stroke assigned to/completed conventional ( $n=$ 13/13) or virtual reality ( $n=$ 13/12) intervention	Korean version 3.0	Scores: SD scale scores ranged from mean 19.7 to 29.4 depending on group & time.
Responsiveness: SD scores $\mathrm{\uparrow}$ SGNF in both groups after intervention.
Pramanick et al. (2020) [50]	Indians with chronic stroke assigned to/completed PT $+$ Curare ( $n=$ 25/22) or PT $+$ homeopathy ( $n=$ 25/23) intervention	Bengali version 3.0	Scores: SD sum scores (?) ranged from mean 8.9 to 14.1 depending on group & time.
Responsiveness: SD scores $\mathrm{\uparrow}$ SGNF in both groups at 3 & 6 mo.
Rosenfeldt et al. (2019) [51]	Americans with chronic stroke assigned to/completed forced exercise $+$ repetitive task practice ( $n=$ 17/16), voluntary exercise $+$ repetitive task practice ( $n=$ 20/16), or education $+$ repetitive task practice ( $n=$ 11/8) intervention	Version 3.0	Scores: SD scale scores ranged from 51.0 to 69.0 depending on group & time.
Responsiveness: SD scores $\mathrm{\uparrow}$ in all 3 groups with $\mathrm{\uparrow}$ SGNF in forced exercise & voluntary exercise groups.
Rossit et al. (2019) [52]	Scots with chronic stroke & neglect assigned to/completed visual feedback training ( $n=$ 10/9) or control ( $n=$ 10/10) intervention	United Kingdom version 3.0	Scores: SD scale scores ranged from a mean 29.8 to 53.4 depending on group & time.
Responsiveness: SD scores $\mathrm{\uparrow}$ in control group but not visual feedback group.
Sims et al. (2009) [53]	Australians with chronic stroke & depression assigned to/completed progressive resistance exercise ( $n=$ 23/21) or control ( $n=$ 22/22) intervention	Version 3.0	Scores: SD scale scores were a mean 65.2 or 73.0 depending on group.
Skoglund et al. (2019) [54]	Swedes with chronic stroke ( $N=$ 45)	Version not stated	Scores: SD scale scores ranged from 0 to 100. The mean score was 68.2 or 58.7 depending on time.
Validity: SD scores correlated SGNF with MRS scores ( $r_{s}=-$ 0.38)
Responsiveness: SD scores $\mathrm{\downarrow}$ SGNF between 1 & 5 years post stroke.
Stuart et al. (2009) [55]	Italians with mild to moderate chronic stroke assigned to/completed adaptive physical activity ( $n=$ 49/40) or control (44/38) intervention	Version not stated	Scores: SD scale scores were a mean 35.8 & 38.0 in the 2 groups.
Studenski et al. (2005) [56]	Americans with chronic stroke assigned to/completed home exercise program ( $n=$ 50/44) or usual care ( $n=$ 50/49) for intervention	Version not stated	Scores: SD scale scores ranged from a mean 53.2 to 66.1 depending on group & time.
Responsiveness: SD scores $\mathrm{\uparrow}$ over time in both groups but $\mathrm{\uparrow}$ not SIGNF different.
Tistad et al. (2012) [57]	Swedes with chronic stroke enrolled in/completed ( $n=$ 175 @ baseline, $n=$ 170 m @ 3 mo, 175 @ 6 m) study	Version 3.0	Scores: SD Scale score was a median 75.0 @ 3 & 12 mo after stroke.
Validity: Dissatisfaction with care @ 12 mo related SGNF to SD @ 3 & 12 mo.

Table 1, continued
Study	Participants	Stroke impact scale	Findings
Van de Port et al. (2008) [58]	Dutch with chronic stroke ( $n=$ 26)	Version 2.0	Scores: SD scale score was a median 52.1.
Validity: SD score correlated SGNF with Motricity Index ( $r_{s}=$ 0.64).
Reliability: Test-retest ICC for SD $=$ 0.69. Inter-tester ICC for SD $=$ 0.61
Wall et al. (2019) [59]	Swedes with chronic stroke assigned to/completed hybrid assistive ( $n=$ 17/15) or conventional ( $n=$ 17/13) intervention	Version 3.0	Scores: SD scale score was a mean 31.0 or 25.0 depending on group.
Validity: SD scores not correlated SGNF with SIS recovery score.
Wang et al. (2015) [60]	Taiwanese with chronic stroke assigned to/completed home-based ( $n=$ 25/25 or control ( $n=$ 26/26) intervention	Version 3.0	Scores: SD scale score ranged from a mean 36.8 to 52.3 depending on group and time.
Responsiveness: SD scores $\mathrm{\uparrow}$ . In home-based intervention $\mathrm{\uparrow}$ was SGNF.
Westerlind et al. (2020) [61]	Swedes with chronic stroke enrolled in completed ( $n=$ 266/234) study	Version not stated	Scores: SD scale scores ranged from 0 to 100 (median $=$ 69 or 100 depending on group).
Validity: SD scores of functionally independent SGNF $\mathrm{>}$ SD scores of functionally dependent.
Wolf et al. (2008) [62]	Americans with chronic stroke assigned to/completed constraint-induced movement (CIM) ( $n=$ 98/70) intervention	Version not stated	Scores SD scale scores with least square means 54.0 to 62.3 depending on time.
Responsiveness: SD scores $\mathrm{\uparrow}$ SGNF after intervention.
Wolf & Koster (2013) [63]	Americans with chronic stroke continuing participation in high demand leisure activities: no ( $n=$ 53), yes ( $n=$ 55)	Version not stated	Scores: SD scale scores were a mean 71.2 or 85.5 depending on activity group.
Validity: SD scores in high demand activity group SGNF $\mathrm{>}$ in lower demand activity group.
Wu et al. (2012) [64]	Taiwanese with chronic stroke assigned to/completed CIM ( $n=$ 19), CIM with trunk stabilization ( $n=$ 20), or control ( $n=$ 18) intervention	Version not stated	Scores: SD sum scores (?) were a mean 2.2 to 2.8 depending on group & time.
Responsiveness: SD scores $\mathrm{\uparrow}$ in CIM groups.
Wu et al. (2007) [65]	Taiwanese with subacute or chronic stroke assigned to/completed CIM ( $n=$ 13/13) or traditional ( $n=$ 13/13) intervention	Version 2.0	Scores: SD scale score ranged from a mean 35.6 to 51.9 depending on group & time.
Responsiveness: SD scores of CIM group $\mathrm{\uparrow}$ between pre- and post-treatment.
Wu et al. (2012) [66]	Taiwanese with chronic stroke assigned to/completed therapist-based bilateral ( $n=$ 14/14), robot-assisted bilateral ( $n=$ 14/14), or control ( $n=$ 14/14) intervention	Version 3.0	Scores: SD scale scores ranged from a mean 36.2 to 51.3 depending on group & time.
Responsiveness: SD scores $\mathrm{\uparrow}$ in both bilateral training groups but not control group.
Ytterberg et al. (2017) [67]	Swedes with chronic stroke ( $n=$ 100)	Version 3.0	Scores: SD scale scores ranged from 0 to 100 & were a mean 72 or 77 depending on time of measurement.
Validity: SD scores for mild stroke classified by BI were SGNF $\mathrm{<}$ for moderate/severe strokes. SD scores for patients greater, but not SGNF, for patients $\mathrm{<}$ 65 y compared with patients $\mathrm{\geqslant}$ 65 y.
Responsivness: SD scores $\mathrm{\downarrow}$ SGNF between 1 & 6 years post-stroke, but only 27% of patients $\mathrm{\downarrow}$ .

${}^{*}$ ADL $=$ activities of daily living, BI $=$ Barthel Index, CIM $=$ constraint induced movement, FIM $=$ Functional Independence Measure, HADS $=$ Hospital Anxiety & Depression scale, IADL $=$ instrumental activities of daily living, MCID $=$ minimal clinically important difference, MMSE $=$ Mini-mental State Examination, MRS $=$ Modified Rankin Scale, NIH $=$ National Institutes of Health, NIHSS $=$ National Institutes of Health Stroke Scale, SGNF $=$ significant/significantly.

Anxiety and Depression Scale scores ( $r_{s}=-$ 0.20 and $-$ 0.24) [16].

Three primary aspects of relative reliability have been addressed in the literature: internal consistency, test-retest reliability, and inter-tester reliability. Internal consistency of SD item scores has been documented in 7 studies ( $\alpha=$ 0.71 to 0.94) with all but one reporting Cronbach’s alpha to exceed 0.80 [11, 13, 15, 16, 25, 33, 47]. Seven studies used intraclass correlation coefficients (ICCs) to describe test-retest reliability of the SD of the SIS (ICC $=$ 0.57 to 0.87) [11, 13, 16, 20, 43, 47, 58] whereas 2 studies used weighted kappa ( $K_{w}$ ) to describe the test-retest reliability of individual SD items ( $K_{w}=$ 0.56 to 0.75) [16, 47]. Four studies used ICCs to describe the inter-tester reliability of the SD of the SIS (ICC $=$ 0.61 to 0.92) [11, 15, 24, 58] while 1 study used $K_{w}$ to describe the inter-tester reliability of individual SD items ( $K_{w}=$ 0.39 to 0.60) [15].

Responsiveness, or the sensitivity of a measurement to changes in strength over time, found considerable support in studies showing SD scores to increase, often significantly, over time or concomitant with interventions [3, 4, 5, 6, 7, 8, 13, 14, 15, 22, 23, 28, 29, 30, 31, 32, 33, 36, 37, 38, 39, 40, 41, 42, 43, 44, 49, 50, 51, 52, 56, 60, 62, 64, 65, 66]. These increases could vary depending on the intervention applied, the severity of stroke, or the chronicity of stroke [3, 28, 29, 39, 51, 52, 56, 60, 64, 65, 66]. In regard to chronicity, the SD of the SIS actually decreased significantly between 1 and 5 years in one study [54] and between 1 and 6 years in another study [67]. Formal statistical summaries of responsiveness over time in the review were scarce. With 1 exception, they involved distribution-based statistics including effect size [39], standardized response mean [13], standard error of measurement [15, 43], and smallest real difference/minimal detectable change [15, 43]. The exception was the purported minimal clinically important difference of 9.2 points referred to by Lewthwaite et al. [40].

4. Discussion

This review focused on literature addressing use of the SD of the SIS as a self- or proxy-report measure of limb muscle strength. The large number of individuals in different settings whose diverse limb strength has been measured with different versions of the SD of the SIS studies provides support for its use. Nevertheless, use of the SD of the SIS almost exclusively with patients with stroke limits its generalizability. While known groups and convergent validity of the SD of the SIS is well established in this review, its relationship with other self-report measures of limb muscle strength remains unexplored [2]. Discriminant validity is mostly supported, but whether strength should be or not be related to specific nonphysical variables is debatable. Also unaddressed is self-report of trunk muscle strength.

The reliability of SD measures of the SIS, whether test-retest or inter-tester, is generally acceptable in this study. However, ICCs less than 0.80 are not uncommon. Moreover, the SD of the SIS is a essentially ordinal measure not best addressed by the ICC. So the reliability of the SD of the SIS would appear to warrant further scrutiny. The same applies to information on responsiveness. This review has several limitations beyond those already noted. First, a single examiner was engaged in the selection and review of articles. Second, by intent, no quality rating was conducted.

5. Conclusions

Use of the SD of the SIS is supported by considerable evidence. However, further research Is required, particularly in its use with nonstroke populations.

Ethical considerations

This study, as a literature review, was exempt of Institutional Review Board approval.

Funding

The author reports no funding.

Footnotes

Acknowledgments

The author has no acknowledgments.

Conflict of interest

The author claims no conflict of interest.

References

International Classification of Functioning, Disability and Health. Geneva, World Health Organization, 2001, 96–97.

Bohannon

. Patient-report as an option for describing muscle weakness: an integrative review. Isokinet Exerc Sci. 2019; 27(3): 79–82.

Duncan

Wallace

Lai

Johnson

Emberetson

Laster

. The Stroke Impact Scale Version 2.0. Evaluation, of reliability, validity, and sensitivity to change. Stroke. 1999; 30(10): 2131–2140.

Ahmad

Singh

DKA

Nordin

NAM

Nee

Ibrahim

. Virtual reality games as an adjunct in improving upper limb function and general health among stroke survivors. Int J Environ Res Pub Health. 2019; 16(24): 5144.

Allen

Richardson

McIntyre

Janzen

Meyer

Ure

, et al. Community stroke rehabilitation teams: providing home-based stroke rehabilitation in Ontario, Canada. Can J Neurol Sci. 2014; 41(6): 697–703.

Badke

Sherman

Boyne

Page

Dunning

. Tongue-based biofeedback for balance in stroke: results of n 8 week pilot study. Arch Phys Med Rehabil. 2011; 92(9): 1364–1370.

Barclay-Goddard

Lix

Tate

Weiberg

Mayo

. Health-related quality of life after stroke: does response shift occur in self perceived physical function? Arch Phys Med Rehabil. 2011; 92(11): 1762–1769.

Barry

Ross

Worhrle

. Therapy incorporating a dynamic wrist-hand orthosis versus manual assistance in chronic stroke: a pilot study. J Neurol Phys Ther. 2012; 36(1): 17–24.

Bohannon

. Adequacy of simple measures for characterizing impairment in upper limb strength following stroke. Percept Mot Skills. 2004; 99(3): 813–817.

10.

Branco

Oliveira

Pinheiro

Ferreira

. Assessing upper limb function: transcultural adaptation and validation of the portuguese version of the stroke upper limb capacity scale. BMC Sports Sci Med Rehabil. 2017; 9: 15.

11.

Brandão

Teixeira

Brandão

Vidotto

Jardim

Gazzoyyi

. Translation and cultural adaptation of the stroke impact scale 2.0 (SLS): a quality-of-life scale for stroke. Sao Paulo Med J. 2018; 136(2): 144–149.

12.

Byun

Kohen

Becker

Kirkness

Khot

Mitchell

. Stroke impact syndromes are associated with sleep-related impairment. Heart Lung. 2020; 49(2): 117–122.

13.

Caël

Decavel

Binquet

Nenaim

Puyaveau

Chotard

, et al. Stroke Impact Scale Version 2: validation of the French version. Phys Ther. 2015; 95(5): 778–790.

14.

Caleffi Segura

Fontes

Fukujima

de Andrade Matas

. The impact evaluation of physical therapy on the quality of life of cerebrovascular stroke patients. Int J Rehabil Res. 2006; 29(3): 243–246.

15.

Carod-Artal

Coral

Trizotto

Moreira

. Self and proxy-report agreement on the Stroke Impact Scale. Stroke. 2009; 40(10): 3308–3314.

16.

Carod-Artal

Coral

Trizotto

Moreira

. The Stroke Impact Scale 3.0. Evaluation of acceptability, reliability, and validity of the Brazilian version. Stroke. 2008; 39(9): 2477–2484.

17.

Carod-Artal

Trizotto

Coral

Moreira

. Determinants of quality of life in Brazilian stroke survivors. J Neurol Sci. 2009; 284(1–2): 63–68.

18.

Chen

C-M

Tsai

C-C

Chung

C-Y

Chen

C-L

KPH

. Potential predictors for health-related quality of life in stroke patients undergoing inpatient rehabilitation. BMC Health Qual Life Outcomes. 2015; 13: 118.

19.

Cheng

Nelson

Brooks

Salbach

. Validation of stroke-specific protocols for the 10-meter walk test and 6-minute walk test conducted using 15-meter and 30-meter walkways. Top Stroke Rehabil. 2020; 27(4): 251–261.

20.

Choi

Lee

Shin

Koo

Park

, et al. Stroke Impact Scale 3.0: reliability and validity evaluation of the Korean version. Ann Rehabil Med. 2017; 41(3): 387–393.

21.

Cook

Sunnerhagen

Persson

. Level of physical activity is positively correlated with perceived impact on life 12 months after stroke: a cross-sectional study. J Rehabil Med. 2020; 52(5)(11): jrm00056.

22.

Dettmers

Teske

Hamzei

Uswatte

Taub

Weiller

. Distributed form of constraint-induced movement therapy improves functional outcome and quality of life after stroke. Arch Phys Med Rehabil. 2005; 86(2): 204–209.

23.

Dobkin

BHK

Nadeau

Behrman

Rose

Bowden

, et al. Prediction of responders for outcome measures of locomotor experience applied post stroke trial. J Rehabil Res Dev. 2014; 51(1): 39–50.

24.

Duncan

Lai

Tyler

Perera

Reker

Studenski

. Evaluation of proxy responses to Stroke Impact Scale. Stroke. 2002; 33(11): 2593–2599.

25.

Edwards

O’Connell

. Internal consistency and validity of the Stroke Impact Scale 2.0 (SIS 2.0) and SIS-16 in an Australian sample. Qual Life Res. 2003; 12(8): 1127–1135.

26.

Essers

Meyer

DeBruyn

Van Gils

Boccuni

Triccas

, et al. Mismatch between observed and perceived upper limb function: an eye-catching phenomenon after stroke. Disabil Rehabil. 2019; 41(13): 1545–1551.

27.

Fulk

Duncan

Klingman

. Sleep problems worsen health-related quality of life and participation during the first 12 moths of stroke rehabilitation. Clin Rehabil. 2020; 34(11): 1400–1408.

28.

Guidetti

Ytterberg

Ekstam

Johansson

Eriksson

. Changes in the impact of stroke between 3 and 12 moths post-stroke, assessed with the Stroke Impact Scale. J Rehabil Med. 2014; 46(1): 963–968.

29.

Hsieh

Y-W

C-Y

Wang

W-E

Lin

K-C

Chang

K-C

Chen

C-C

, et al. Bilateral robotic priming before task orientation approach in subacute stroke rehabilitation: a pilot randomized controlled trial. Clin Rehabil. 2017; 31(2): 225–233.

30.

Huang

Y-H

C-Y

Hseieh

Y-W

Lin

K-C

. Predictors of change in quality of life after distributed constraint-induced therapy in patients with chronic stroke. Neurorehabil Neural Repair. 2010; 24(6): 559–566.

31.

Hung

C-S

Hsieh

Y-W

C-Y

Lin

Y-T

Lin

K-C

Chen

C-L

. The effects of combination of robot-assisted therapy with task-specific or impairment oriented training on motor function and quality of life in chronic stroke. PM R. 2016; 8(8): 721–729.

32.

Hung

C-S

Lin

K-C

Chang

W-Y

Huang

W-C

Chang

Y-J

Chen

C-L

, et al. Unilateral vs bilateral hybrid approaches for upper limb rehabilitation in chronic stroke: a randomized controlled trial. Arch Phys Med Rehabil. 2019; 100(12): 2225–2232.

33.

Kamel

Ghani

Zaiton

El-Motayam

El-Fattah

. Health related quality of life in stroke survivors measured by the Stroke Impact Scale. Egypt J Neurol Psychiat Neurosurg. 2010; 47(2): 267–274.

34.

Kamwesiga

, von Kock Eriksson

Guidetti

SGE

. The impact of stroke on people living in central Uganda: a descriptive study. African J Disabil. 2018; 7(0): a438.

35.

Khanittanuphong

Tipchatyotin

. Correlation of the gait speed with quality of life and the quality of life classified according to speed-based community ambulation in Thai stroke survivors. Neurorehabilitation. 2017; 41(1): 135–141.

36.

Kim

Park

. The effect of occupation-based bilateral upper extremity training in a medical setting for stroke patients: a single-blinded, pilot randomized controlled trial. J Stroke Cerebrovasc Dis. 2019; 28(12): 104335.

37.

Kirac-Unal

Gencay-Can

Karaca-Umay

Cakci

. The effect of task-oriented electromyography-triggered electrical stimulation of the paretic wrist extensors on upper limb motor function early after stroke: a pilot randomized controlled trial. Int J Rehabil Res. 2019; 42(1): 74–81.

38.

Lai

S-M

Studenski

Duncan

Perera

. Persisting consequences of stroke measured by the Stroke Impact Scale. Stroke. 2002; 33(7): 1840–1884.

39.

Lee

Fischer

Zera

Robertson

Hammel

. Examining a participation-focused stroke self-management intervention in a day rehabilitation setting: a quasi-experimental pilot study. Top Stroke Rehabil. 2017; 24(8): 601–607.

40.

Lewthwaite

Winstein

Lane

Blanton

Wagenheim

Nelson

, et al. Accelerating stroke recovery: body structures and functions, activities. participation, and quality of life outcomes from a large rehabilitation trial. Neurorehabil Neural Repair. 2018; 32(2): 150–165.

41.

Lieshout

ECCV

van de Port

Dijkhuizen

Visser-Meily

JMA

. Does upper limb strength play a prominent role in health-elated quality of life in stroke patients discharged from inpatient rehabilitation? Top Stroke Rehabil. 2020; 27(7): 525–533.

42.

Lin

K-C

Chang

Y-F

C-Y

Chen

Y-A

. Effects of constraint-induced therapy versus bilateral arm training on motor performance, daily functions, and quality of life in stroke survivors. Neurorehabil Neural Repair. 2009; 23(5): 441–448.

43.

Lin

K-C

C-Y

Wang

Y-H

Liu

J-S

Hsieh

C-J

, et al. Minimal detectable change and clinically important difference of the Stroke Impact Scale in stroke patients. Neurorehabil Neural Repair. 2010; 24(5): 486–492.

44.

Lou

Yuan

Hao

Miller

Yan

Zuo

, et al. Effects of glucose fluctuation targeted intervention on the prognosis of patients with type 2 diabetes following the first episode of cerebral infarction. J Diabtes Res. 2020; Article ID 2532171.

45.

Mühr

Persson

Sunnerhagen

. Long-term outcome after reperfusion-treated stroke. J Rehabil Med. 2017; 49(4): 316–321.

46.

Muren

Hütler

Hooper

. Functional capacity and health related quality of life in individuals post stroke. Top Stroke Rehabil. 2008; 15(1): 51–58.

47.

Ochi

Ohashi

Hachisuka

Saeki

. The reliability and validity of the Japanese version of the Stroke Impact Scale version 3.0. J UOEH. 2017; 39(3): 215–221.

48.

Palmcrantz

Widén

Sommerfeld

Tisdad

Ytterberg

Von Koch

. Differences between younger and older individuals in their use of care and rehabilitation but not in self-perceived global recovery 1 year after stroke. J Neurol Sci. 2012; 321(1–2): 29–34.

49.

Park

M-H

S-W

Lee

J-Y

Ham

, et al. Effects of virtual reality-based planar motion exrcises on upper extremity function, range of motion, and health-related quality of life: a multicenter, single-blinded randomized, controlled pilot study. J Neuroengineer Rehabil. 2019; 16: 122.

50.

Pramanick

Uchat

Chattopadhyay

Mir

Koley

Saha

. An open-label randomized pragmatic non-inferiority pilot trial comparing the effectiveness of Curare 30CH against individualized homeopathic medicines in post-stroke hemiparesis. Advan Integr Med. 2020; 7(2): 79–88.

51.

Rosenfeldt

Linder

Davidson

Clark

Zimmerman

Lee

, et al. Combined aerobic exercise and task practice improve health-related quality of life poststroke: a preliminary analysis. Arch Phys Med Rehabil. 2019; 100(5): 923–930.

52.

Rossit

Benwell

CSY

Szmanek

Learmonth

McKernan-Ward

Corrigan

, et al. Efficacy of home-based visuomotor feedback training in stroke patients with hemispatial neglect. Neuropsychol Rehabil. 2019; 29(2): 251–272.

53.

Sims

Galea

Taylor

Dodd

Jespersen

Joubert

, et al. Regenerate: assessing the feasibility of a strength-training program to enhance the physical and mental health of chronic post stroke patients with depression. Int J Geriatr Psychiat. 2009; 24(1): 76–83.

54.

Skoglund

Westerlind

Persson

Sunnerhagen

. Self-perceived impact of stroke: a longitudinal comparison between one and five years post-stroke. J Rehabil Med. 2019; 51(9): 660–664.

55.

Stuart

Benvenuti

Macko

Taviani

Segenni

Mayer

, et al. Community-based adaptive physical activity program for chronic stroke: feasibility, safety, and efficacy of the Empoli model. Neurorehabil Neural Repair. 2009; 23(7): 726–734.

56.

Studenski

Duncan

Perera

Reker

Lai

Richards

. Daily functioning and quality of life in a randomized controlled trial of therapeutic exercise for subacute stroke survivors. Stroke. 2005; 36(8): 1764–1770.

57.

Tistad

Tham

Von Koch

Ytterberg

. Unfulfilled rehabilitation needs and dissatisfaction with care 12 months after a stroke: an explorative observational study. BMC Neurology. 2012; 12: 40.

58.

Van de Port

IGL

Leenes

, Sellmeijer Zuidgeest

Kwakkel

. Reliability and criterion validity of the Dutch Stroke Impact Scale 2.0 in patients with stroke. Ned Tijdschr Fysiother. 2008; 118(1): 12–18.

59.

Wall

Borg

Palmcrantz

. Self-perceived functioning and disability after randomized conventional and electromechanically-assisted gait training stroke AND gait in subacute stroke: a 6 months follow-up. Neurorehabil. 2019; 45(4): 501–511.

60.

Wang

T-C

Tsai

Wang

J-Y

Lin

Y-T

Lin

K-L

Chen

, et al. Caregiver-mediated intervention can improve physical functional recovery of patients with chronic stroke: a randomized controlled trial. Neurorehabil Neural Repair. 2015; 29(1): 3–12.

61.

Westerlind

Hörsell

Persson

. Different predictors after stroke depending on functional dependency at discharge: a 5-year follow up study. BMC Neurol. 2020; 20: 263.

62.

Wolf

Winstein

Miller

Thompson

Taub

Uswatte

, et al. Retention of upper limb function in stroke survivors who have received constraint-induces movement therapy: the EXCITE randomized trial. Lancet Neurol. 2008; 7(1): 33–40.

63.

Wolf

Koster

. Perceived recovery as a predictor of physical activity participation after mild [stroke]. Disabil Rehabil. 2013; 35(14): 1143–1148.

64.

C-Y

Chen

Y-A

Lin

K-C

Chao

C-P

Chen

Y-T

. Constraint-induced therapy with trunk restraint for improving functional outcomes and trunk-arm control after stroke: a randomized controlled trial. Phys Ther. 2012; 92(4): 483–492.

65.

C-Y

Chen

C-L

Tsai

W-C

Lin

K-C

Chou

S-H

. A randomized controlled trial of modified constraint-induced movement therapy for elderly stroke survivors: changes in motor impairment, daily functioning, and quality of life. Arch Phys Med Rehabil. 2007; 88(3): 273–278.

66.

C-Y

Yang

C-L

Chuang

L-L

Lin

K-C

Chen

H-C

Chen

M-D

, et al. Effect of therapist-based versus robot-assisted bilateral arm training on motor control, functional performance, quality of life after chronic stroke: a clinical trial. Phys Ther. 2012; 92(8): 1006–1016.

67.

Ytterberg

Dybäck

Bergström

Guidetti

Eriksson

. Perceived impact of stroke six years after onset, and changes in impact between one and six years. J Rehabil Med. 2017; 49(8): 637–643.

68.

Feinstein

. Clinimetrics New Haven, Yale University Press, 1987, 204–205.