Validation of the Arabic Version of the Epworth Sleepiness Scale among the Yemeni Medical Students

Abstract

The study was conducted with the aim to assess the psychometric measures of an adapted Arabic version of the Epworth Sleepiness Scale (ESS) among medical students at Sana’a University, Yemen. The cross-sectional study targeted 360 students (males: 176; females: 184) from the preclinical 3rd year (N: 197) and the final clinical year (N: 163). Participants self-filled an Arabic and slightly modified version of the 8-item Epworth Sleepiness Scale. Exploratory Principal Component Analysis (PCA) and Confirmatory Factor Analyses (CFA) were conducted on two equal subsets of the sample (N: 180 each). The PCA yielded a two-dimension model subsequently confirmed by factor analysis. The first dimension was grouped on three items while the second dimension had five items reflecting the respondents’ propensity to sleep during “interactive situations” and “sitting and lying,” respectively. The model had an acceptable goodness of fit measures for the overall ESS (CMINDF=2.362, CFI=0.91, IFI=0.92) and acceptable reliability indicators (factor 1 α=0.65, factor 2 α=0.62). However, due to weak variance explanation (0.07) of item 6 (sitting and talking) in factor 1, analysis was repeated excluding this item. The 7-item model was also two-dimensional, valid, and reliable. The reliability indicators were acceptable with α=0.65 for factor 1 (4 items of interactive situations) and 0.62 for factor 2 (3 items of sitting) and overall α=0.68. Overall, the ESS is a useful tool. Factor analysis produced a two-factor model of 7 items with good validity and reasonable reliability that can be used in diagnosing daytime sleepiness among young Yemeni adults.