Can we get more from the Epworth Sleepiness Scale (ESS) than just a single score?: a confirmatory factor analysis of the ESS

Purpose: The Epworth Sleepiness Scale (ESS) is a widely used tool for measuring sleepiness. In addition to providing a single measure of sleepiness (a one-factor structure), the ESS also has the capacity to provide additional information about specific factors that facilitate sleep onset, including a person's posture, activity and environment. These features of sleepiness are referred to as somnificity. This study evaluates and compares the fit of a one-factor structure (sleepiness) and three-factor structure (reflecting low, medium and high levels of somnificity) for the ESS. Methods: All participants (a community sample N = 356 and a clinical sample N = 679) were administered the ESS. Confirmatory factor analysis was used to evaluate and compare the fit of one- and three-factor models of the ESS. Results: In both samples, a three-factor structure (community sample adjusted X 2 = 2.95, root mean square error of approximation (RMSEA) = 0.07, Comparative Fit Index (CFI) = 0.95; clinical sample adjusted X 2 = 3.98, RMSEA = 0.07, CFI = 0.98) provided a level of model fit that was at least as good as the one-factor structure (community sample adjusted X 2 = 5.01, RMSEA = 0.11, CFI = 0.87; clinical sample adjusted X 2 = 8.87, RMSEA = 0.11, CFI = 0.92). Conclusions: In addition to a single measure of sleepiness, the ESS can provide subscale scores which relate to three underlying levels of somnificity. These findings suggest that the ESS can be used to measure an individual's overall sleep propensity as well as more specific measures of sleep propensity in low, moderate and high levels of situational somnificity. © 2012 Springer-Verlag.