Measurement properties of the Epworth sleepiness scale.

Obstructive sleep apnea (OSA) has been estimated to affect 2–4% of middle aged adults [1]. Excessive daytime sleepiness is an important criterion both for establishing the severity of OSA and for determining the response to specific treatment such as nasal continuous positive airway pressure (CPAP) [2]. Objective assessment of daytime sleepiness with tests such as the multiple sleep latency test (MSLT) is both time-consuming and costly and is not offered by many sleep disorders clinics [3]. Consequently, a questionnaire that reliably quantifies the severity of daytime sleepiness and that is responsive to changes in daytime sleepiness over time would greatly assist clinicians and researchers in the management and investigation of OSA. Since its publication in 1991, the Epworth sleepiness scale (ESS) has been used by several groups of investigators to measure daytime sleepiness in patients with known or suspected OSA [4]. Furthermore, the ESS has also been used to track changes in daytime sleepiness during treatment of OSA [5,6]. This usage of the ESS as an evaluative instrument that measures change over time may not be appropriate given the original goals of its development and the consequent design of the questionnaire. Furthermore, the use of the ESS as a diagnostic tool may be premature. The objectives of this review are to determine what the ESS actually measures and whether that measurement truly reflects objective sleepiness, and to determine if the ESS can be used to diagnose pathological sleepiness and follow its response to treatment. 2. What does the ESS measure?