Iranian Rehabilitation Journal (Sep 2024)
Investigating the Validity and Reliability of Bengali Version of Communication Outcome After Stroke Scale
Abstract
Objectives: The speech and language therapy (SLT) profession in Bangladesh has been around for a decade. Yet, therapists are facing challenges in measuring the effectiveness of SLT services after providing this therapeutic intervention to the patient for a certain period. Due to patients’ demands, there is an increasing need to develop and adapt outcome measurement tools for patients with communication difficulties. Accordingly, this study examines the validity and reliability of communication outcomes after the stroke scale developed in the Bengali language for individuals with stroke. Methods: Content validity of the communication outcome after stroke Bengali (COAST-BD) scale was determined by item objective congruence (IOC) as ≥0.67 of three experts. Meanwhile, 110 stroke survivors and their caregivers from the SLT department of CRP hospital were enrolled in this study, considering one-to-two-week intervals for examining the test re-test reliability of the scale. The Cronbach α, correlation matrix, the Kaiser-Meyer–Olkin (KMO) test and the Bartlett test (χ2), in addition to the intraclass correlation coefficient, were applied to compute the psychometric properties of the scale. Exploratory factor analysis was applied to identify the potential item structure in the scale. Results: The Cronbach α coefficient for the internal consistency of the overall scale was 0.97 and 0.89. The interclass correlation coefficient for the test re-test reliability of the COAST-BD scale was 0.96 and the composite reliability score ranged from 0.70 to 0.91. The COAST-BD scale estimated correlated values among its subscales on visit-1 (0.370 to 0.661) and visit-2 (0.288 to 0.520), respectively. The KMO test and χ2 were also calculated correlation values on visit-1 (KMO=0.628; χ2=82.90) and visit-2 (KMO=0.614; χ2=48.10), respectively. Discussion: The COAST-BD scale showed significant internal consistency, test re-test reliability, correlation matrix, and construct validity.
