Validation of the Malay version of Epworth sleepiness scale for children and adolescents (MESS-CHAD)

Abstract

Abstract Background Epworth Sleepiness Scale for Children and Adolescents (ESS-CHAD) is a valid and reliable eight-item self-administered questionnaire for the assessment of excessive daytime sleepiness and is commonly used to screen sleep-disordered breathing for children and adolescents. The cross-sectional study aimed to translate and cross-culturally adapt ESS-CHAD into a Malay version of Epworth Sleepiness Scale for Children and Adolescents (MESS-CHAD) for the Malaysian population, and to assess the validity and reliability of MESS-CHAD. Methods Forward-backward translation method was used to translate and cross-culturally adapt ESS-CHAD. Three linguistic experts and two paediatricians content validated the translated version. Face validity was conducted through audio-recorded semi-structured in-depth interviews with 14 native Malay-speaking children and adolescents followed by thematic analysis. The revised questionnaire was then proofread by a linguistic expert. A total of 40 subjects answered the MESS-CHAD twice, 2 weeks apart, for test-retest reliability and internal consistency. For criterion validity, 148 eligible subjects and their parents answered MESS-CHAD and the Malay version of Sleep-Related Breathing Disorder scale extracted from the Paediatric Sleep Questionnaire (M-PSQ:SRBD) concurrently. Variance Inflation Factor (VIF) and P values of the model’s outer weight and outer loading were analysed using SmartPLS software to assess the indicator’s multicollinearity and significance for formative construct validity. Results Intraclass Correlation Coefficient (ICC) ranging from 0.798 to 0.932 and Cronbach’s alpha ranging from 0.813 to 0.932 confirmed good to excellent test-retest reliability and internal consistency, respectively. Spearman Correlation Coefficient value of 0.789 suggested a very strong positive correlation between MESS-CHAD and M-PSQ:SRBD. VIF ranging from 1.109 to 1.455 indicated no collinearity problem. All questionnaire items in MESS-CHAD were retained as the P value of either outer model weight or outer model loading was significant (P < 0.05). Conclusion ESS-CHAD has been translated and cross-culturally adapted into Malay version for the Malaysian population, and found to be valid and reliable.