Translation, cross-cultural adaptation and psychometric testing of Yoruba Version of the EQ-5D Questionnaire in patients with musculoskeletal disorders

Fatoye, Francis ORCID: https://orcid.org/0000-0002-3502-3953, Akinfala, Abiodum Emmanuel, Gebrye, Tadesse ORCID: https://orcid.org/0000-0001-7976-2013, Fatoye, Clara, Ojelade, Titilope Faith, Oyewole, Oyeleye Olufemi and Chidozie, Emmanuel Mbada (2022) Translation, cross-cultural adaptation and psychometric testing of Yoruba Version of the EQ-5D Questionnaire in patients with musculoskeletal disorders. Frontiers in Public Health, 10. p. 902680. ISSN 2296-2565

Preview

Published Version Available under License Creative Commons Attribution. Download (531kB) | Preview

Abstract

Background: The EuroQol- 5 Dimension (EQ-5D) is a generic self-administered questionnaire used for the measurement and economic valuation of a wide range of health conditions, which necessitates its existence and adaptation in different languages. Currently, the tool does not exist in any Nigerian language. This study was aimed to translate, cross-culturally adapt, and determine the reliability and validity of the Yoruba version of the EQ-5D-5L questionnaire. Methods: The International Quality of Life Assessment (IQOLA) project guidelines, involving forward translation, reconciliation and harmonisation, backward translation, and reconciliation of problematic items was used in the Yoruba translated version of the EQ-5D-5L (EQ-5D-Yor). A total of 113 and 109 persons with musculoskeletal disorders participated in the validity, and seven-day test-retest reliability testing of the EQ-5D-Yor respectively. Convergent and discriminant validity of the EQ-5D-Yor were determined using the Yoruba version of the SF-12 (SF-12-Y) and Visual Analogue Scale (VAS). Data was analysed using descriptive and inferential statistics of Spearman correlation, Intra-Class Correlation, Cronbach alpha and multi trait scaling analysis. Alpha level was set as p < 0.05. Results: The construct validity of the EQ-5D-Yor yielded Spearman rho ranging from 0.438 to 1.000, with the EQ-VAS having the highest co-efficient (r = 1.000; p = 0.001). The convergent validity of the EQ-5D-Yor index with scales and domains of the SF-12-Y yielded no significant correlations (p < 0.05), except the physical functioning scale (r = -0.709, p = 0.001). On the other hand, the divergent validity of the EQ-5D-Yor index with VAS yielded a moderate negative correlation, (r = -0.482; p=0.001). The Intra-class Correlation Coefficient and Cronbach alpha for the test-retest reliability of the EQ-5D-Yor were 1.000 and 0.968. The confirmatory factor analysis showed the factor loadings were poor when including VAS in the model. Conclusion: The EQ-5D-Yor has acceptable validity and reliability and can be used as a valid tool among Yoruba speaking population with musculoskeletal disorder.