Day-to-day associations between objectively and subjectively measured sleep parameters and daytime emotions in older adults

Abstract We investigated the relationship between subjective and objective sleep, as well as (subjective and objective) sleep and emotions in older adults in a multi-day study with a particular focus on night-to-day associations. 72 participants aged 50–79 were enrolled in a seven-day home-based study, where they rated their sleep and emotions in the mornings (and afternoons). Objective sleep data were collected with an electroencephalographic headband each night. Positive and negative morning and afternoon emotion ratings were aggregated separately, while objective sleep variables were reduced with principal component analysis. Temporal associations were analyzed with multilevel models. At the intraindividual level, poorer subjective sleep quality was associated with less positive and more negative emotions in the morning and afternoon, but not vice versa. Furthermore, objectively assessed sleep parameters were linked to subjective sleep quality, and the composite measure of objective hyperarousal was also associated with morning emotions. Beyond the day-to-day effects, lower subjective sleep quality was associated with less positive and more negative emotions at the between-person level. In conclusion, we found unidirectional temporal relationships between subjective sleep quality, objective sleep parameters and daytime emotions. It was corroborated that self-assessed sleep is associated with emotional states the upcoming day, but contrary to previous findings, objective sleep quality was also associated with morning emotions. Moreover, in naturalistic settings with repeated assessments, objective sleep seems more closely related to subjective sleep quality at the within-person level than previously observed in the sleep laboratory cross-sectionally.