We recruited 101 people with stroke (58 male, 43 female) from a local self-help group via poster advertisements. People with stroke were included in the study if they: (1) were 50–80 years old; (2) suffered a single stroke that was confirmed by magnetic resonance imaging or computed tomography at least 1 year before the start of the study; (3) scored 7 or higher in the Chinese version of the Abbreviated Mental Test (27 (link)); (4) could speak Chinese (Cantonese); (5) had volitional control of their non-paretic arm; (6) could induce at least minimal anti-gravity movement in the shoulder of their paretic arm; (7) had at least 5° of wrist extension in the anti-gravity position; and (8) could walk independently for at least 10 m with or without an assistive device. People with chronic stroke were excluded if they: (1) had any other unstable medical conditions (e.g., angina pectoris, pain, or arthritis) or other conditions with medications that may intervene the upper limb function (e.g., Parkinson's disease or Multiple Sclerosis); and (2) had any aphasia or hearing impairment that would affect the data collection procedure.
We recruited 50 healthy older adults (14 male, 36 female) aged 50–80 years with stable health as the control group. People with any comorbid neurological, cardiovascular, or musculoskeletal disease that might affect the assessment were excluded.
The 15-item UEFI has demonstrated excellent test–retest reliability (ICC2, 1 = 0.95) in people with upper extremity musculoskeletal disorders (19 (link)). However, its reliability in people with stroke has not been evaluated. Assuming that an ICC value for assessing test–retest reliability of the 15-item UEFI in people with stroke was 0.9, a sample size of ≥46 subjects was required to achieve 80% power to detect an ICC of 0.9 with a null hypothesis ICC of 0.8 and a significance level of 0.05. To evaluate the ability of the C-UEFI to discern differences between different groups, ≥50 people with chronic stroke and ≥50 healthy controls were required with a CA0 value of 0.3 and CA1 value of 0.7 (28 (link)). A sample of >100 people with chronic stroke were regarded as reasonable based on the exploratory factor analysis (EFA) estimation (29 (link)). Thus, 101 people with stroke were recruited for this study.
We recruited 50 healthy older adults (14 male, 36 female) aged 50–80 years with stable health as the control group. People with any comorbid neurological, cardiovascular, or musculoskeletal disease that might affect the assessment were excluded.
The 15-item UEFI has demonstrated excellent test–retest reliability (ICC2, 1 = 0.95) in people with upper extremity musculoskeletal disorders (19 (link)). However, its reliability in people with stroke has not been evaluated. Assuming that an ICC value for assessing test–retest reliability of the 15-item UEFI in people with stroke was 0.9, a sample size of ≥46 subjects was required to achieve 80% power to detect an ICC of 0.9 with a null hypothesis ICC of 0.8 and a significance level of 0.05. To evaluate the ability of the C-UEFI to discern differences between different groups, ≥50 people with chronic stroke and ≥50 healthy controls were required with a CA0 value of 0.3 and CA1 value of 0.7 (28 (link)). A sample of >100 people with chronic stroke were regarded as reasonable based on the exploratory factor analysis (EFA) estimation (29 (link)). Thus, 101 people with stroke were recruited for this study.
Free full text: Click here
