PDF
Citation
Print
INTRODUCTION
Individuals who have experienced a stroke often have difficulty in activities of daily living (ADLs), social interactions, and other activities, without physical assistance and supervision from others. These difficulties arise from functional limitations primarily caused by impairments that affect individuals’ health and overall wellbeing [1]. However, life after a stroke can be affected by factors beyond impairments. To address comprehensive stroke rehabilitation, various psychological factors—including self-efficacy, motivation, engagement, and positive enforcement—should be considered.
Among those factors, self-efficacy is a psychological construct comprising belief, confidence, and motivation. It is defined as a belief in one’s ability to execute and change behaviors necessary to achieve valuable outcomes, affecting overall task performance from choosing an activity to completing it [2]. Self-efficacy after stroke is related to functional independence, depression, quality of life, ADL, and physical activities [3,4]. Recognizing and promoting self-efficacy is crucial for promoting and encouraging active engagement in stroke rehabilitation.
In the context of upper extremity (UE) rehabilitation, self-efficacy is associated with an individual’s use of their hands in daily life; those with higher self-efficacy for UE movements use their hands more frequently [5]. Confidence in one’s capacity to use the affected upper limb is highly important because it correlates with motor ability and impacts treatment outcomes [6,7]. Accordingly, an assessment tool that can measure UE self-efficacy in patients who have had a stroke would be meaningful in guiding treatment and the overall rehabilitation process, although many stroke programs objectively assess functional performance without considering the individual’s functional outcomes [3,8].
To date, individuals’ UE self-efficacy after a stroke has been assessed with questionnaires, including the Stroke Self-Efficacy Questionnaire (SSEQ), measure of self-efficacy for reaching, upper limb self-efficacy test (UPSET-stroke), and the Confidence in Hand and Arm Movement (CAHM) scale. The SSEQ comprises 13 items related to general perspectives after stroke, and it measures general self-efficacy related to stroke recovery rather than UE-specific self-efficacy. The self-efficacy for reaching test only assesses confidence in reaching movements, excluding other functional tasks necessary for daily life [6]. The UPSET-stroke assesses UE-specific self-efficacy in various ADL; however, it mainly examines self-efficacy in basic ADL with limited consideration for instrumental ADLs and social contexts, and its questions are not specifically described.
The CAHM scale was developed to assess self-efficacy regarding hand and arm function for people who have had a stroke [9]. It is particularly useful because it assesses an individual’s perceived confidence in relation to ADL and instrumental ADL in social and individual contexts. Unlike previous tools, CAHM includes a broader range of tasks, capturing confidence in both basic and instrumental ADLs, making it a more comprehensive measure of UE-specific self-efficacy. Moreover, as self-efficacy influences motor function, rehabilitation engagement, and long-term recovery, the CAHM scale provides valuable clinical insights into both physical and psychosocial aspects of stroke rehabilitation.
Although the validity and reliability of the CAHM have been reported, the original version has not been published in any peer-reviewed article, which makes it difficult to access and use the assessment tool. Furthermore, cultural norms, values, and daily activities in Korea differ significantly from those in the region where the CAHM was originally developed. These differences can affect how patients perceive and report their abilities and challenges in UE movements. Further, the healthcare system in Korea, including stroke rehabilitation practices, may differ from those where the CAHM scale was initially implemented. These differences can impact patient experiences and expectations. Therefore, a culturally adapted version of the CAHM scale is essential to ensure it is relevant and meaningful for Korean patients. Adapting the scale to fit the Korean healthcare context ensures that it is in accordance with the rehabilitation protocols and patient management strategies. Therefore, this study aimed to translate and modify the CAHM for the Korean cultural context and confirm its reliability and validity.
METHODS
Participants
The study was conducted from May 2022 to June 2023 at a rehabilitation facility with 54 participants who had suffered strokes. The inclusion criteria were diagnosis of ischemic or hemorrhagic stroke, and adequate cognitive and verbal abilities for following instructions (Mini-Mental State Examination score≥24). Those who had severe comorbidity or disability before the stroke or other neurological disorders were excluded. This study was approved by the Institutional Review Board of the National Rehabilitation Center (approval no. NRC-2022-02-013) and registered at Clinical Research Information Service (KCT0007151). The purpose of this study was fully explained to participants, and written informed consent was obtained before enrollment. The questionnaires were self-reported.
Procedures
Translation and cross-cultural adaptation
For this study, permission to translate CAHM into Korean was obtained from the original author. First, the original questionnaire was translated independently by two translators proficient in both English and Korean. Then, a discussion was held to integrate the translations. Next, the amalgamated version was reviewed by a translation committee comprising rehabilitation experts, including physiatrists, psychologists, and occupational therapists. The Korean version of the CAHM—the K-CAHM version 1.0—was created. Subsequently, two additional translators, proficient in both English and Korean and without prior exposure to the questionnaire, reverse-translated K-CAHM version 1.0. Finally, the translation committee reviewed and integrated the original version of the questionnaire, K-CAHM version 1.0, and the reverse-translated version. Through detailed discussions, they created the final version of the K-CAHM (Supplementary Materials S1, S2). During this process, items that were not suitable for Korean culture or had different meanings in Korean were modified (Fig. 1). For readers more familiar with English, we included the English version of K-CAHM in Supplementary Material S3.
Assessment
CAHM: CAHM is a 20-item questionnaire that asks an individual to rate their confidence in performing various functional tasks involving the weaker arm or both arms on a scale of 0 to 100; a rating of 0 indicates “very uncertain/unconfident,” while a rating of 100 indicates “very certain/confident.”
Fugl-Meyer Assessment (FMA): The FMA is the most often-used tool to evaluate motor impairment in patients who have had a stroke [10,11]. In this study, we measured upper limb motor function using FMA.
Jebsen-Taylor Hand Function Test (JTHFT): The JTHFT was designed to assess hand dexterity and coordination. It comprises seven tasks that require individuals to perform various functional hand tasks related to ADL [12]. In this study, we used a standardized scoring system for validity verification [13].
Modified Barthel Index (MBI): The MBI is designed to assess an individual’s ability to independently perform ADL. It includes 10 ADL-related tasks—personal hygiene, bathing, feeding, toileting, stair climbing, dressing, bowel control, bladder control, chair/bed transfers, and ambulation. Each task is rated on a 5-point scale, with the total score ranging from 0 to 100. The assessment focuses on actual patient performance and considers their use of assistive aids during the test. Higher scores indicate greater independence, while lower scores suggest less independence in ADL [14]. We tested the correlation between K-CAHM and each item of MBI; however, only upper limb-relevant items (personal hygiene, feeding, dressing) were reviewed for validation.
Stroke Impact Scale 3.0 (SIS): Health-related quality of life of patients who have had a stroke is assessed by SIS using eight domains: strength, hand function, ADL/instrumental ADL, mobility, communication, emotion, memory and thinking, and social participation [15,16]. Each domain’s score ranges from 0 to 100, with higher scores indicating better quality of life. In addition, stroke recovery was measured by asking “how much have you recovered from the stroke.” We analyzed the association of eight domains and stroke recovery of SIS with K-CAHM and used SIS-hand and stroke recovery as our variables of interest for validation.
Reliability and validity assessment
To determine the reliability of K-CAHM, we assessed internal consistency, test-retest reliability, and agreement. The test-retest reliability was estimated by administering K-CAHM to participants who have had a stroke, twice within 7–10 days. Its validity was scrutinized by correlating it to upper limb relevant items from other measurements: FMA-upper limb, JTHFT, MBI-personal hygiene, MBI-bathing, MBI-feeding, MBI-dressing, SIS-hand, and SIS-stroke recovery. The measurements for validity verification encompass three domains of the International Classification of Functioning, Disability and Health (ICF): body function and structures, activities, and participation [17]. FMA represents body function and structure, JTHFT and MBI represent activity, and SIS indicates participation.
Statistical analysis
The absolute values and percentages were used to show the categorical variables, while the means and standard deviations were employed for the numerical variables. Internal consistency and test-retest reliability were analyzed using a Cronbach’s alpha (α) and an intraclass correlation coefficient (ICC), respectively, with a 95% confidence interval (CI). A Cronbach’s alpha≥0.70 was considered acceptable and ≥0.9 was considered excellent [18]. The ICC was calculated using two-way mixed-effect models. ICC values between 0.75 and 0.9 were interpreted as demonstrating good reliability and ≥0.9 demonstrated excellent reliability [19]. Bland–Altman analysis was used to quantify the agreement between the two measures of K-CAHM by plotting the difference against their mean. An exploratory factor analysis (EFA) was conducted to determine whether the scale measured a single construct or multiple dimensions. To examine the validity, a Pearson’s correlation test was employed with upper-limb related outcomes. All statistical analyses were conducted using R version 4.2, and statistical significance was set at a p<0.05.
RESULTS
Demographic characteristics
Participants comprised 54 patients with a mean age of 62.3±11.1 years and a duration of 101.2±88.1 days after their stroke. Patients’ demographic characteristics are presented in Table 1.
Translation and cross-cultural adaptation
We modified several items for cultural adequacy (Table 2). First, in the instructions, we changed “weaker hand” to “weaker arm” for clarity for patients; “the weaker hand” is repeated in many questions in CAHM. For Question 5, we removed the word “plastic sleeve” as it could be confusing. For Question 11, we modified the unit of measurement from inches to centimeters. For Questions 15 and 16, as the original terms were unfamiliar in Korea, we replaced the word “China platter” with “ceramic platter” and “Styrofoam cup” with “disposable cup.” Finally, for Question 17, we specified wearing “kitchen-mitts” on both hands, as “potholders” are not used when handling a hot pot in Korean culture, and we replaced “casserole dish” with “hot pot.” Moreover, we attached pictures to help patients understand the meaning of some questions (Questions 2, 4, 7, 8, 10, 11, 13, and 19) to avoid any confusion.
Reliability and validity
Reliability of K-CAHM
A Cronbach’s alpha of 0.97 indicated the excellent internal consistency of the K-CAHM. The ICC was 0.895 with 95% CI=0.824–0.938, indicating good to excellent test-retest reliability. Additionally, Bland–Altman plots showed good agreement between the test and retest of K-CAHM, in which 51 out of 54 were within 95% limits of agreement (Fig. 2). An EFA revealed five distinct factors, with factor-specific Cronbach’s alpha values ranging from 0.87 to 0.93, indicating appropriate internal consistency without excessive redundancy (Table 3).
Validity of K-CAHM
Table 4 shows the correlations between K-CAHM and other upper limb-related measures. K-CAHM demonstrated a weak correlation with FMA (r=0.387, p=0.004) and a moderate correlation with JTHFT (r=0.493, p<0.001). Additionally, it showed significant correlations with MBI-personal hygiene (r=0.315, p=0.020) and MBI-bathing (r=0.304, p=0.032), but not with MBI-feeding and MBI-dressing. Furthermore, it had a positive correlation with the SIS-hand (r=0.634, p<0.001) and SIS-stroke recovery (r=0.384, p=0.004) domains.
DISCUSSION
This study developed K-CAHM by modifying the original CAHM based on the cultural environment of Korea. K-CAHM showed good reliability and validity, suggesting that it can be used as an indicator to assess self-efficacy in patients who have had a stroke.
The Cronbach’s alpha of 0.97 and ICC of 0.895 are comparable with those of the original CAHM (0.96 and 0.91, respectively; Lewthwaite, unpublished). Meanwhile, when Cronbach’s alpha exceeds 0.95, it may indicate excessive redundancy rather than a distinct functional assessment [20]. Our Cronbach’s alpha calculation yielded α=0.97, raising concerns about the scale’s redundancy and homogeneity because a highly correlated scale may not effectively capture the complexity of hand function. To explore this, we conducted an EFA to examine the scale’s structure and reassessed reliability for each factor. The EFA identified five distinct factors, suggesting the scale measures multiple dimensions rather than a single, overly redundant construct. Factor-specific reliability values supported internal consistency without excessive overlap, ensuring strong psychometric properties. In addition, the good agreement evidenced by Bland–Altman analysis bolstered the reliability of K-CAHM. These approaches confirmed the reliability of K-CAHM and affirmed its validity in assessing weak hand function across multiple domains. Notably, the K-CAHM showed good concurrent validity, indicated by the correlation with the three domains of the ICF. This is especially meaningful because the K-CAHM considers various aspects of upper limb-related functions. FMA and JTHFT showed significant positive correlations with K-CAHM, suggesting that K-CAHM is indicative of upper limb function, including body structure/function and activities. Consistent with previous findings [5], people with stroke who had better self-efficacy tended to use their upper limbs more in daily living. Specifically, the high correlations between K-CAHM and JTHFT may be because K-CAHM mainly comprises questions that assess upper limb activities using tools. Additionally, K-CAHM significantly correlated with MBI-personal hygiene and MBI-bathing, suggesting its validity regarding upper limb activity. The non-significant findings of MBI-feeding and MBI-dressing might have come from the characteristics of K-CAHM. MBI assesses basic activities, while K-CAHM explores preparatory or more advanced activities, which require higher-level performance. K-CAHM significantly correlated with SIS-hand and SIS-stroke recovery, demonstrating its validity in the area of participation. Interestingly, K-CAHM correlated with SIS-stroke recovery, reflecting that it is indicative of confidence or subjective rating about one’s recovery. Additionally, both K-CAHM and SIS 3.0 are self-report questionnaires. This significant correlation may suggest that K-CAHM could be used as self-report questionnaire. However, correlations between K-CAHM and other variables were not significant, as other items addressed tasks irrelevant to hand function, focusing on lower limb or other functions, such as bladder or bowel control. Overall, K-CAHM proved to be a valid tool to assess functions in diverse domains, including body function/structure, activity, and participation.
Cultural adaptation is crucial to ensure the validity, reliability, and relevance of measurements for the target population. Especially, confidence is inherently subjective, as it reflects an individual’s perception, belief, and self-assessment of their abilities. Thus, unfamiliar elements influencing self-perception should be minimized for more reflective data collection. Notably, the items of K-CAHM are specifically described including concrete objects and actions, thereby facilitating more accurate scoring for individuals. Moreover, individuals who have experienced a stroke and are dealing with disability are likely to lose confidence. Therefore, reducing unfamiliarity allows them to be more cooperative and reflective of their confidence. Therefore, it is imperative to modify the questionnaire to reflect cultural differences, as seen in Table 2. Nevertheless, we conserved the purpose of the original CAHM. For example, in Question 1, we did not change “knife and fork” to “spoon and chopsticks,” although knife and fork are not routinely used in Korean culture. We preserved the term because the intention was to measure confidence in bimanual activity or coordination, and using forks is relatively familiar. Furthermore, it is important to consider whether the task remains relevant today. For example, telephone books (Question 11), although commonplace in the past, are rarely used nowadays because online searches have replaced them. Moreover, we expected that Question 10, which involves lighting matches, could be difficult for respondents to answer since matches are less often used today than they were in the past. Nevertheless, it was hard to find an alternative task that involved bimanual coordination, dexterity, and a safety issue such as a burn. To summarize, our modification ensured that the K-CAHM accurately reflects the experiences and perspectives of individuals in the Korean cultural context, thus contributing to its overall validity and reliability.
K-CAHM reflects patients’ own perspectives of their upper limb function—that is, their self-efficacy or confidence. CAHM comprises questions dealing with diverse aspects. The EFA also demonstrated five distinct factors from K-CAHM; self-regulatory self-efficacy, fine motor skills, gross motor skills, bimanual coordination, risk-sensitive task, indicating that the K-CAHM measures self-efficacy in multiple aspects.
It ascertains confidence during tasks relevant to upper limb functioning, using either the affected or both sides. CAHM includes tasks in diverse environments, not only in a private space where one is alone, but also in public areas such as a restaurant or a restroom, where one’s performance can be observed by others. In addition, it measures self-efficacy when performing the most challenging tasks, especially risk-sensitive ones. Notably, it measures self-efficacy related to self-regulation, in which individuals independently identify their needs, set goals, explore resources, employ appropriate strategies, and evaluate their outcomes [21]. Assessments of individuals’ self-efficacy provide insights into their recovery potential and response to rehabilitation [22]. Self-efficacy can predict whether the affected arm is used following an intervention, even when upper limb function is retained, in people who have had a stroke [23]. Self-efficacy demonstrated a strong correlation with social reintegration and is a crucial factor in assessing performance improvement [24,25].
Moreover, the weak correlations between K-CAHM and specific MBI domains, such as personal hygiene (r=0.315) and bathing (r=0.304), suggest that self-efficacy and actual task performance are distinct constructs. This aligns with Maujean et al. [26], who found similar weak correlations between the self-efficacy scale and Barthel index, supporting the view that self-efficacy and actual function are separate aspects among individuals with stroke. Self-efficacy represents confidence in performing tasks and plays a crucial role in rehabilitation. Along with performance, self-efficacy significantly influences constructive behaviors, attitudes, and beliefs related to recovery after stroke. While actual functioning is influenced by physical ability, social factors, and psychological factors, strong self-efficacy can drive greater engagement in rehabilitation efforts and improve long-term functional outcomes.
Fostering self-efficacy is essential for empowering individuals to persist in challenging tasks and maximize their independence. This underscores the need for a multidimensional rehabilitation approach that prioritizes self-efficacy alongside physical function to enhance overall recovery. Rehabilitation should prioritize self-efficacy as a core component, alongside motor function and psychosocial adaptation, to promote sustained improvements in daily performance.
Therefore, evaluating both performance and self-efficacy provides a comprehensive perspective of an individual’s rehabilitation progress. It is suggested that healthcare professionals in Korea regularly assess patients’ self-confidence using K-CAHM, along with other outcome measures, to optimize rehabilitation outcomes for individuals who have had a stroke.
This study has some limitations. First, the sample size is small; thus, the findings may not be representative of the broader population of stroke patients. Second, intra-rater reliability was not verified. K-CAHM is a self-report questionnaire; therefore, our verification methods showed good reliability. Third, this study was conducted at a single rehabilitation hospital in an urban area; thus, the self-efficacy of participants in acute phase of stroke or in rural areas was not evaluated. Fourth, participants were relatively young, as their mean age was 62.3 years. Self-efficacy may be influenced by numerous personal or environmental factors; thus, future studies should include patients in different age groups; stroke phases; and environmental conditions, including residence (urban or rural), patient status (admitted, outpatient, or living in community), and type of caregiver. Fifth, the K-CAHM in this study retains certain items from the original CAHM; however, some tasks (e.g., using a telephone directory, striking a match) may no longer be relevant in modern contexts. These items were preserved for consistency in data collection, but future revisions should incorporate more applicable alternatives. For example, the telephone book task (Question 11) could be replaced with “finding a 7.5 cm-thick book,” while match-striking (Question 10) could be substituted with “buttoning a shirt” or “removing a price tag from a box.” These modifications require validation to ensure they maintain equivalent cognitive and motor assessment criteria.
XML Download