l

The content of this website is intended for healthcare professionals only

High social media use linked to poor sleep in teens

Optimal outcomes for sleep associated with minimal use – so we should support teens in balancing online time and sleep

Louise Prime

Wednesday, 23 October 2019

Teenagers who spend at least three hours a day using social media are more likely than others to have poor sleep patterns including getting to sleep late on school nights and awakening during the night, researchers said today in their study* published in BMJ Open. They found that optimal outcomes for sleep were associated with minimal – not moderate – social media use, and so recommended supporting young people in finding a balance between their online interactions and getting sufficient sleep on school nights.

The authors, from the University of Glasgow, pointed out that although there is significant current interest in the possible impact of screen time and social media on adolescents’ health, previous research in this area has tended to consider associations between poor wellbeing and ‘screen time’ as a whole rather than social media use specifically. So they used cross-sectional data from the Millennium Cohort Study (a national birth cohort study) covering 11,872 adolescents aged 13-15 years, to investigate associations between social media use and six self-reported sleep parameters – sleep onset and wake times on school days and on free days, sleep onset latency, and trouble falling back asleep after night-time awakening – controlling for a wide range of covariates.

The median time spent using social media on a typical day was 1 –<3 hours (32% of participants); however, 14% were ‘high’ users of social media (3 <5 hours daily), and a further 21% were ‘very high’ users (at least five hours); 34% were low users, i.e. less than an hour daily. Girls tended to use social media much more than boys.

The researchers reported that very high social media use was associated with higher odds of all six sleep outcomes, and high social media use with higher odds of all outcomes except for late rise times on free days; low social media use was associated with lower odds of late sleep onset on both school days and free days, and late wake times on free days.

When they took account of a comprehensive list of covariates (exact age, sex, ethnic minority status, family income, number of siblings in household, the presence of both parents in household, parent’s age, Strengths and Difficulties score, Mood and Feelings score, self-esteem, general health, social support and physical activity), high social media use was no longer significantly associated with long sleep onset latency or frequent night-time awakenings, and very high social media use was no longer significantly associated with long sleep onset latency, although its association with frequent night-time awakenings remained significant but smaller. However, patterns of significant associations for late sleep onset and wake times remained unchanged, although effect sizes were reduced, particularly for very high social media use.

The study authors commented: “In terms of sleep timing, social media use remained significantly associated with late sleep onset and wake times after controlling for covariates, with the strongest effect for sleep onset. Very high social media users were roughly 70% more likely than comparable average users to fall asleep later than average, that is, after 23:00 on school days and after midnight on free days. Low social media users were least likely to fall asleep late, indicating that unlike mental well-being, optimal outcomes for sleep are associated with minimal – not moderate – use.”

They said that their results provide robust evidence on associations between social media use and sleep outcomes that can be used to support evidence-based decision-making policy and practice, “rather than relying on assumptions around prevalence of high social media use”.

They acknowledged that restricting adolescent social media use is likely to be especially challenging as peer interactions, belonging and acceptance are highly valued during this developmental period of increasing autonomy.

They concluded: “Interventions should focus on addressing delayed sleep onset, by supporting young people to balance online social interactions with an appropriate and consistent sleep schedule that allows sufficient sleep on school nights, with benefits for health and educational outcomes.”


*Scott H, Biello SM, Woods HC. Social media use and adolescent sleep patterns: cross-sectional findings from the UK millennium cohort study. BMJ Open 2019; 9: e031161. doi:10.1136/bmjopen-2019-031161

Registered in England and Wales. Reg No. 2530185. c/o Wilmington plc, 5th Floor, 10 Whitechapel High Street, London E1 8QS. Reg No. 30158470