555 - Systematic review of studies measuring social media use and mental health in adolescents, 2018-2020

Poster Number: 555
Publication Number: 555.245

Quinlan D. Alfredson, University of Wisconsin School of Medicine and Public Health, Muskego, WI, United States; Megan A. Moreno, University of Wisconsin, Madison, WI, United States

Background: Previous systematic reviews have found associations between increased social media usage and worsened adolescent mental health. These studies have generally focused on mental health symptom burden and “screen time” on social media. However, the current literature has yet to examine study characteristics that may influence these associations.

Objective: The purpose of this study was to examine social media measures and demographic stratification in analyses in recent research on the association between social media usage on depression, anxiety, and psychological distress in adolescents.

Design/Methods: This systematic review study involved a search of PubMed, Embase, PsychINFO, Cumulative Index to Nursing and Allied Health, and Social Sciences Citation Index for dates 2018-2020. Eligible studies were empirical and observational with both measures of social media usage and validated measures of mental health concerns (depression, anxiety, or psychological distress). Eligible studies were evaluated for three focus areas: 1) valence (positive or negative associations found for social media usage and mental health measures), 2) measurement of social media usage, 3) stratification by demographics (gender, race, and socioeconomic status).

Results: Of the 3,131 articles identified, 19 studies met eligibility criteria. The majority (n=12, 63%) reported a positive valence for depression, anxiety, and psychological distress. Studies reporting a positive valence found that social media usage was positively associated with symptoms of depression, anxiety, or psychological distress. Social media measurements and their validation varied between studies, with time spent on social media (n=7, 37%), Bergen Social Media Addiction Scale (n=4, 21%), and Facebook Intrusion Questionnaire (n=2, 11%) being the most common. While a majority of studies reported results stratified by gender (n=11, 58%), few did so by race (n=0, 0%) or socioeconomic status (n=2, 11%).Conclusion(s): Our systematic review supports that many social media and mental health studies are framed around a risk-centered model, assessing measures of problematic social media use. Future work should investigate how demographic characteristics may mediate the relationship between social media use and adolescent mental health. Future reviews and media reports should consider whether social media measures focus on problematic or other features of use.