Total sleep time, sleep efficiency, and next day subjective sleepiness in a large group of women

The relationship between sleep duration and sleepiness has seen much research, but no data are available on the association between polysomnographically (PSG) determined total sleep time (TST) (or other PSG variables) and subjective sleepiness during the subsequent day in individuals in their habitual life situation. The purpose of the present study was to study the association between TST and sleep efficiency (SE) (and other PSG variables) and next-day sleepiness at 7 times of the day. A large population-based group of women (N = 400) participated. Daytime sleepiness was measured with the Karolinska Sleepiness Scale (KSS). The association was studied through analysis of variance (ANOVA), as well as regression analyses. For SE there was a significant difference in sleepiness across groups with >90%, 80%–89.99%, and <80% SE (F = 7.2, p < .001, eta2 = 0.04), with lowest sleepiness in the first group. In contrast, TST groups of <6 h, 6–6.99 h, and ≥7 h did not differ significantly. In addition, a pronounced U-shape (eta2 > 0.45) was seen for both analyses, with maximum sleepiness at bedtime (≈ 7.5 KSS units). A multiple regression analysis, including all PSG variables (adjusted for age and BMI), showed that SE was a significant predictor (β = 0.16, p < .05) of mean sleepiness, even after depression, anxiety, and subjective sleep duration were entered, but this was eliminated by subjective sleep quality. It was concluded that high SE is modestly associated with lower next-day sleepiness in women in a real-life context, but that TST is not.