Dimensionality and reliability of the Epworth Sleepiness Scale in dental patients referred for oral appliance therapy.

BACKGROUND: There is a limited amount of published research on the dimensionality and reliability of the Epworth Sleepiness Scale (ESS) questionnaire for adult patients referred for oral appliance therapy. This information is crucial for dentists, who often lack objective measures of sleep-disordered breathing (SDB) during titration process of an oral appliance. This study investigated the dimensionality and reliability of ESS scores in adult dental patients with SDB undergoing oral appliance treatment. METHODS: In 103 dental patients with SDB referred by a physician (mean age: 52.3 ± 13.0 years; 35% female), the dimensionality of the ESS was investigated using exploratory (EFA) and confirmatory factor analyses (CFA) to determine how many scores are needed to characterize the construct. ESS questionnaires were administered twice before treatment. Internal consistency and test-retest reliability were assessed. RESULTS: Horn's parallel analysis suggested a one-factor model. Extracting one factor and standardizing loadings led to strong loadings for all items, ranging from 0.53 to 0.82. The fit indices indicated a good model fit (Comparative Fit Index = 0.999, Root Mean Square Error of Approximation = 0.020, and Standardized Root Mean Square Residual = 0.064). Cronbach's alpha with 95% confidence interval (CI) was 0.85 (0.82-0.88), indicating strong internal consistency. The intraclass correlation coefficient type 2,1 (95% CI) was 0.86 (0.79-0.90), and the weighted kappa ranged from 0.50 to 0.81. CONCLUSIONS: In this patient population, the ESS reliably characterizes excessive daytime sleepiness with a single score and appears suitable for individual assessment in dental patients undergoing oral appliance treatment for SDB.