The Barthel Index (BI) is a widely recognized and reliable tool for measuring an individual's ability to perform ADLs, its use is limited in Nepal due to the unavailability of a Nepali version and the presence of culturally inappropriate items. Thus, the objective of this study was to undertake the translation, cross-cultural adaptation, and validation of the BI into Nepali, aiming to facilitate its meaningful application in the rehabilitation of stroke survivors in Nepal.
The first step was to translate and cross-culturally adapt BI into Nepali using the Linguistic Validation Guidance of a Clinical Outcome Assessment (COA). The measurement properties of BI-NP were calculated according to the COSMIN criteria. For testing validity, two physiotherapist interns administered the BI-NP, the Nepali versions of WHO Disability Assessment Schedule 2.0 (WHODAS 2.0 -NP), Patient Specific Functional Scale (PSFS-NP), and Time Up and Go Test (TUG-NP) among 76 stroke survivors recruited by convenience sampling. For testing intra- and inter-rater reliability and responsiveness, Global Rating of Change (GROC-NP) was added as an external anchor. Intra rater reliability was tested across two consecutive days. Responsiveness was tested at baseline and follow up at two weeks. Receiver operating characteristic (ROC) curves were plotted to determine the responsiveness of the BI-NP using the area under the curve (AUC), and minimum important changes (MIC).
The BI was successfully translated and cross culturally adapted into Nepali. Cronbach's alpha was 0.899 indicating good internal consistency. The intra-rater and inter-rater reliability were 0.999 (95% CI: 0.999-0.999) and 0.997 (95% CI: 0.996-0.998) respectively using the intraclass correlation coefficient (ICC) of stable group, indicating excellent reliability. The MDC was 7.24. The exploratory factor analysis of BI-NP revealed two factors: somatic function and physiological self-care, accounting for 70.1% total variance. Convergent validity was supported by strong to moderate negative correlations between BI-NP and WHODAS 2.0-NP items, and strong positive correlations with PSFS-NP and TUG-NP. Responsiveness analysis demonstrated an area under curve of 0.798, indicating acceptable discriminative ability to differentiate between improved and non-improved groups. The identified optimal threshold, with a maximum Youden Index of 7.50 on the BI-NP scale, balances sensitivity (0.844) and specificity (0.818) effectively.
The Nepali translated and cross-culturally validated Barthel index shows good internal consistency and excellent intra and inter-rater reliability. BI-NP also presents as a valid and responsive tool for assessing functional independence among Nepalese stroke survivors.
The BI-NP is expected to improve functional outcome assessment and intervention planning among Nepali stroke survivors aiding clinical decision-making. The BI-NP is expected to instigate further research scope in stroke rehabilitation.
Measurement properties
Nepali
