Translation, validity and reliability of the British Sign Language (BSL) version of the EQ-5D-5L

Abstract

Purpose: To translate the health questionnaire EuroQol EQ-5D-5L into British Sign Language (BSL), to test its reliability with the signing Deaf population of BSL users in the UK and to validate its psychometric properties. Methods: The EQ-5D-5L BSL was developed following the international standard for translation required by EuroQol, with additional agreed features appropriate to a visual language. Data collection used an online platform to view the signed (BSL) version of the tests. The psychometric testing included content validity, assessed by interviewing a small sample of Deaf people. Reliability was tested by internal consistency of the items and test–retest, and convergent validity was assessed by determining how well EQ-5D-5L BSL correlates with CORE-10 BSL and CORE-6D BSL. Results: The psychometric properties of the EQ-5D-5L BSL are good, indicating that it can be used to measure health status in the Deaf signing population in the UK. Convergent validity between EQ-5D-5L BSL and CORE-10 BSL and CORE-6D BSL is consistent, demonstrating that the BSL version of EQ-5D-5L is a good measure of the health status of an individual. The test–retest reliability of EQ-5D-5L BSL, for each dimension of health, was shown to have Cohen’s kappa values of 0.47–0.61; these were in the range of moderate to good and were therefore acceptable. Conclusions: This is the first time EQ-5D-5L has been translated into a signed language for use with Deaf people and is a significant step forward towards conducting studies of health status and cost-effectiveness in this population.