167 - Sleep health on social media: Prevalence and content of adolescents' social media sleep posts

Publication Number: 1157.167

Kyle Palmberg, University of Wisconsin School of Medicine and Public Health, Madison, WI, United States; Abigail Rose. Hommer, Marquette University, Milwaukee, WI, United States; Babayosimi O. Fadiran, University of Wisconsin School of Medicine and Public Health, Waunakee, WI, United States; Nina E. Anglim, University of Wisconsin School of Medicine and Public Health, Madison, WI, United States; Qianqian Zhao, University of Wisconsin at Madison, ANN ARBOR, MI, United States; Ariel A.. Williamson, University of Oregon, Portland, OR, United States; Christopher N. Cascio, University of Wisconsin School of Medicine and Public Health, Madison, WI, United States; Ellen Selkie, University of Wisconsin School of Medicine and Public Health, Madison, WI, United States; Megan A. Moreno, University of Wisconsin School of Medicine and Public Health, Madison, WI, United States

Background: Getting sufficient, high-quality sleep is especially important for adolescents yet is often difficult to achieve. Social media is an ever-present component of adolescent social lives and commonly associated with their sleep health. Previous literature has not established how adolescents choose to share their sleep behaviors and experiences on social media.

Objective: The purpose of this study was to investigate adolescents' social media posts that displayed sleep.

Design/Methods: This 6-month longitudinal content analysis evaluated social media profiles of 13-15-year-old adolescents living in Wisconsin. Facebook, Instagram, X, and TikTok accounts were evaluated monthly. Demographic information was collected at enrollment. Trained coders using a theory-driven codebook recorded posts with sleep-related content in the post's images, videos, text and audio. The codebook included sleep actions, sleep disorders, and the six dimensions of the Pediatric B-SATED sleep health model (pediatric sleep behaviors, satisfaction, alertness, timing, efficiency, and duration) to categorize posts. Definitions of these categories and example posts are in Table 1. Coders tested interrater agreement with a subset of 20% of posts, which was 94.1%. A zero-inflated negative binomial model tested associations with age, gender and/or race (including American Indian or Alaskan Native, Asian, Black, and Native Hawaiian or Pacific Islander participants grouped together for analysis) for presence and frequency of sleep posts.

Results: Among the 315 participants, 53.0% identified as female and 79.7% as white. A total of 117 posts contained sleep-related content originating from 55 participants (17.5%). Most participants who displayed sleep did so through image/video content (N = 38, 12.1%) (Table 2), and on Instagram only (N = 24, 7.6%) (Figure 1). Sleep actions were the most frequently recorded category (N = 42, 13.3%) (Table 2). White participants were less likely to present zero sleep posts compared to non-white participants (OR=0.29, p<.05). Additionally, the number of posts was 3.35 times higher in female participants (p <.05) and 4.44 times higher in other gender participants (p <.05) when compared to males.

Conclusion(s): Although displayed sleep content in social media posts was not widely witnessed, specific subgroups showed a higher likelihood of displaying sleep compared to others. Future research should build on these findings by examining whether online sleep posts are associated with offline sleep and health behaviors.

Theory Driven Codebook Definitions and Post Examples
This table shows the different codebook variables used to evaluate social media posts for sleep-related content. Variables have a definition, and dimensions of the B-SATED Pediatric Sleep Health model have example posts generated based off of similar posts seen in the study.

Coding Variables and Content Location Within Sleep Posts
This table displays the number of participants who displayed the different dimensions of the codebook and where in these posts sleep-related content was seen to be displayed. This table also displays the median and interquartile range of the number of posts per participants who displayed these different categories.

Platforms with Participant Sleep Posts
This figure displays the social media platforms that sleep-related content was found on based on the number of participants who displayed on one or more of these platforms.