An observational study published in _PLOS Global Mental Health_ looks at screentime, sleep and depression in adolescents.
**Prof Ben Carter, Professor of Medical Statistics, Institute of Psychiatry, Psychology & Neuroscience at King’s College London, said:**
“The authors have used a pre-registered and hypothesis testing approach taking data from a previous randomised control trial which included 4810 Swedish adolescents from 55 schools aged 12 to 16. In this well set up secondary analysis where the authors apply causal inference to an observational cohort study, they found that in girls screentime displaced sleep and was associated with increased symptoms of depression over a nine-month period.
“Sleep in the development of adolescents is essential. This study offers biologically plausible data that demonstrates the evidence for widely held views of the impact of displacement of screens and the impact on sleep and symptoms of depression. The authors show that typical screentime in this cohort was in excess of Sweden’s recommended maximum of two to three hours per day. This work supports the introduction of public health leadership and introduction of national recommendations on smartphone use during school evenings for children and adolescents. This would likely lead to collective action to empower parents and help adolescents improve their sleep and prevent daytime sleepiness.
“Weaknesses include that screentime was defined as leisure time on the internet and they were not able to differentiate between the type of screen used and self-reported. The strengths of this work include that the participants were a sample of typical 12- to 16-year-olds from Europe, they applied causal inference with a directed acyclic graph to explain the causal pathways with a pre-registered _a priori_ hypothesis.”
**Prof Chris Ferguson, Professor of Psychology, Stetson University, said:**
“Overall, I don’t find this article to be an impressive addition to our understanding of screen time. The measures of screen time appear to be self-report, a method known for its unreliability. The questions are crude and not validated, and likely to prompt hypothesis guessing among participants that could result in false positive results. There appear to be no proper statistical controls for theoretically relevant variables such as family environment, which might have explained any correlation. Nonetheless, the results from this study are remarkably weak. There is no correlation between screen time and depression which would seem to be important as findings go, given how much angst there is about that topic. Though billed as “longitudinal” the lag between the first and second time frames is probably too short to be meaningful. Associations between screen time and sleep were weak, overall. In conclusion, the data are correlational, not causal, of weak or (in the case of depression) null effects. Unfortunately, the authors also use causal language, which is inappropriate for any correlation design, including longitudinal, particularly given the lack of appropriate statistical controls. Ultimately, this study tells us very little about adolescent screen time, sleep, depression or anything else. With these limitations, I’d worry about making any firm conclusions which could concern parents or carers about screen use in adolescents without the evidence to back it up.”
**‘****Adolescents’ screen time displaces multiple sleep pathways and elevates depressive symptoms over twelve months****’ by** **Sebastian Hökby** **_et al._** **was published in _PLOS Global Mental Health_ at 19:00 UK time April 2nd April.**
**DOI:** **[https://doi.org/10.1371/journal.pgph.0004262](https://doi.org/10.1371/journal.pgph.0004262)**
**Declared interests**
**Prof Ben Carter:** No declarations.
**Prof Chris Ferguson:** No declarations