Repeating administration of Epworth Sleepiness Scale is clinically useful

Abstract

Purpose: We aimed to verify whether it is clinically useful to repeat the Epworth Sleepiness Scale (ESS) in individuals with suspected sleep-disordered breathing (SDB).

Methods: In this cross-sectional, prospective study, results of the repeated administration of the ESS were analyzed. In 929 consecutive patients, ESS was obtained as usual in the laboratory routine, immediately before the sleep study (ESS1) and was repeated in the morning, after the polysomnography (ESS2). ROC curve, classical psychometry, and item response theory (IRT) Rasch analysis were used to assess measurement properties of ESS.

Results: The ESS1 score was (mean ± SD), 11 ± 5.1, and the ESS2, 13 ± 4.7 (p < 0.001). Change in ESS score is explained in multivariate analysis by age, but not by gender, educational level, change in Stanford Sleepiness Scale, symptoms of sleep disorder, or polysomnography results. Accuracy of an ESS score >10 to predict apnea-hypopnea index ≥5 increased from 56% (ESS1) to 72% (ESS2). IRT psychometric properties (unidimensionality, invariance, local independence) were maintained in ESS2.

Conclusions: Repeating the administration of the Epworth Sleepiness Scale in a clinical setting increases its score and diagnostic accuracy and correlation with SDB variables, without changing the psychometric properties of the scale. This experiment indicates the clinical usefulness of repeating the ESS. The scale can be repeated at a negligible cost, before dismissing individual patients on the basis of a low ESS score, discontinuing a potentially lifesaving diagnostic and therapeutic process.