Nighttime Social Media Use and Sleep Among Adolescents at High Risk for Suicide: Data from Two Studies Using Mixed Methods and Prospective Designs

• In Event: New Advances in the Study of Adolescents’ Social Media Use: Theoretical and Empirical Perspectives

Abstract

Nighttime social media use is associated with poor sleep among adolescents. Existing research, however, is limited by its reliance on self-reported sleep measures and cross-sectional designs, which makes it unclear whether social media use impacts sleep or whether youth turn to social media because of poor sleep. No study to date has examined this relationship using both prospective designs and objective sleep measures among high-risk adolescents, who may be particularly vulnerable to the effects of social media use and poor sleep. Importantly, shorter sleep duration and later sleep timing is a robust and modifiable risk factor for adolescent suicide. To examine whether nighttime social media use predicts sleep (duration, timing), two separate studies were conducted with adolescents (13-18 years old, mean = 15.17 years) in an intensive outpatient program (IOP) for depression and suicidality for one month. In Study 1, participants (N = 100) completed a baseline measure of social media use before bed (i.e., 0 = less than 30 minutes, 1 = 30-59 minutes, 2 = 1-2 hours, 3 = more than 2 hours) and weekly measures of sleep duration and timing (onset). In Study 2, participants (N =18) completed a baseline measure of nighttime social media use (i.e., minutes of SM use within 2 hours of going to sleep), completed subjective daily measures of sleep onset and offset, and wore an actiwatch for 30 days to objectively assess sleep duration and timing. Multilevel modeling was conducted with social media use predicting the intercept of sleep duration and timing over the next month for each study, covarying for age, gender, depression symptoms, and time. A large proportion of adolescents reported using social media for more than 1 hour before bed (32% in Study 1 and 44% in Study 2). Per actigraphic and self-reported sleep measures across both studies, adolescents reported falling asleep around 11:30AM and getting around 7 hours of sleep per night. Multilevel models indicated that more use of social media before bed significantly predicted later sleep timing in both studies using different methods, including self-reported in Study 1 (B = .51; SE = .15; p = .001) and Study 2 (B = .02; SE = .01; p = .006) and actigraphic sleep timing (Study 2: B = .02; SE = .01; p = .008). Nighttime social media use also predicted shorter sleep duration, but only when it was self-reported in Study 1 (B = -.43; SE = .19; p = .02) and Study 2 (B = -.02; SE = .01; p = .01), but not for actigraphy-assessed sleep (Study 2: B = -.01; SE = .01; p = .43). Overall, findings suggest that social media use before bed predicts later sleep timing among adolescents with depression and suicidality, and that nighttime social media use may predict adolescents’ perceptions of getting insufficient sleep. However, future studies are needed to disentangle the daily effects of social media use on sleep, and examine whether reducing nighttime social media use improves sleep health and subsequent risk for depression and suicide.

Authors

• Jessica Leigh Hamilton, Rutgers University - New Brunswick
  Presenting Author

• Candice Biernesser, University of Pittsburgh
  Non-Presenting Author

• Craig Sewall, University of Pittsburgh
  Non-Presenting Author

• Jamie Zelazny, University of Pittsburgh
  Non-Presenting Author

• Noelle Rode, University of Pittsburgh
  Non-Presenting Author

• Giovanna Porta, University of Pittsburgh
  Non-Presenting Author

• Edward Hamilton, University of Pittsburgh
  Non-Presenting Author

• Kimberly D. Poling, University of Pittsburgh
  Non-Presenting Author

• David Brent, University of Pittsburgh
  Non-Presenting Author

• Peter Franzen, University of Pittsburgh School of Medicine
  Non-Presenting Author

• Tina Goldstein, University of Pittsburgh
  Non-Presenting Author