Emergence of different dimensions of sleepiness in a general population sample: An EPISONO study

Abstract

IntroductionRecent evidence supports the idea that sleepiness has several dimensions, comprising psychophysiological phenomena, such as sleep propensity and sleepiness perception. The Epworth Sleepiness Scale (ESS) is among one of the most used sleepiness assessment tools, but recent data suggest that it might not detect problematic impairments in sleep health. More research is warranted investigating sleepiness dimensions with large sample sizes, using both objective and subjective methods of analyzing sleep. MethodsWe used data from the Epidemiological Sleep Study (EPISONO), a major study carried out in Brazil using a representative sample of 1042 participants of the general population of Sao Paulo city, who completed questionnaires and underwent type I polysomnography. Sleepiness was measured by the ESS (sleep propensity) and the UNIFESP Sleep Questionnaire, which asked about the frequency of feeling sleepy during the day (sleepiness perception). The participants were distributed into 4 groups in respect of excessive daytime sleepiness (EDS) according to the following criteria: 1) having an ESS score >10 (ESS group). 2) feeling sleepy ≥3 times per week (Frequent EDS group). 3) the combined criteria (Combined EDS group). 4) having no evidence of EDS (no EDS group). The UNIFESP Sleep Questionnaire was used to obtain self-reported sleep duration, sleep debt, and sleep duration variability. Participants also completed the Pittsburgh Sleep Quality Index, the Insomnia Severity Index, and a pre-polysomnography sleep questionnaire. ResultsAfter removing participants with missing data, there were 620 participants with no EDS, 255 individuals in the ESS group, 68 in the Frequent EDS group, and 73 in the Combined EDS group. Compared to the no EDS group, the ESS and Combined EDS groups had significantly increased sleep efficiency, diminished sleep latency, wake after sleep onset, and self-reported weekly sleep duration. The Frequent and Combined EDS groups had significantly higher scores in the Insomnia Severity Index and Pittsburgh Sleep Quality Index. All EDS groups had an increased likelihood of nonrestorative sleep. EDS frequency and the ESS scores were significantly correlated. ConclusionsParticipants with a high sleep propensity had a profile suggesting prior sleep deprivation, while high sleepiness perception was associated with impaired sleep quality. These findings indicated that, in the general population, these instruments evaluated separate sleepiness dimensions with different associations, highlighting that screening using a combination of instruments might be more effective in detecting impairments in sleep health.