Loneliness and sleep in everyday life: Using ecological momentary assessment to characterize the shape of daily loneliness experience

Loneliness has been linked to an increased risk of sleep problems. Past research has largely relied on trait loneliness or daily recall loneliness when evaluating associations with sleep. The present study extended this work by evaluating the patterns of loneliness throughout the day, including a daily average of all reports, a maximum value, and daily variation. These loneliness patterns predicted daily subjective and objective sleep measures to evaluate whether they provide unique insight to this relationship. Undergraduate students (n = 71; 77% female; age 18-28) completed 2weeks of electronic surveys 4 times a day to assess loneliness. Each morning participants completed a diary of their prior night’s sleep quality, as well as wore actigraphy devices to objectively assess sleep parameters. A total of 778 momentary surveys and 565days of actigraphy-assessed sleep data were collected. Multilevel models tested whether within-person daily aggregates of loneliness were associated with within-person daily sleep outcome variables. Subjective sleep duration, quality, and fatigue were significantly predicted by daily average loneliness. Subjective sleep latency, quality, and fatigue were significantly predicted by daily max loneliness. Only fatigue was significantly predicted by daily loneliness variability. No objective sleep measures were significantly predicted by daily loneliness measures. Patterns of daily loneliness focusing on central tendency (average) or intensity (max) were more consistently associated with subjective (but not objective) assessments of sleep than variability.