In the past decade, mental health concerns in adolescents have been on the rise, and so has the use of social media (Orben et al., 2024). This has prompted increasing research to investigate the link between social media use and adolescent mental health. While the evidence suggests small associations between time spent on social media and mental health symptoms (e.g., depressive symptoms, Teague et al., 2026), we do not know whether young people with mental health conditions (e.g., Major Depressive Disorder) have different social media experiences (read Amanda and Louise’s Mental Elf blog to learn more).

Importantly, mental health conditions can be grouped into internalising and externalising based on their underlying features (Achenbach et al., 2016). Internalising conditions, such as anxiety disorders and eating disorders, tend to involve negative self-views, rumination, worries, and social withdrawal, while externalising conditions, such as conduct disorder or attention deficit hyperactivity disorder (ADHD), tend to involve negative emotionality towards others, expressed through impulsivity, risk taking and disinhibition. It is possible that these types of conditions may have very different relationships with social media, but this has not been directly tested in previous studies.

In a recent study, Fassi et al. (2025) examined whether adolescents with mental health conditions use social media differently from those without a condition, and whether this differed by internalising or externalising conditions.

Methods

The researchers analysed data from the 2017 Mental Health of Children and Young People study, a nationally representative survey of children and adolescents in England. To establish whether participants had a mental health condition, adolescents and their parent(s) went through a diagnostic assessment with a clinical rater. Mental health conditions were then grouped into internalising and externalising conditions.

Social media use was examined both quantitatively, in terms of time spent on social media, and qualitatively, which involves seven dimensions of social media engagement. These included:

• Online social comparison
• Perceived lack of control over time spent online
• Monitoring of online feedback
• Perceived impact of online feedback
• Online friendship
• Honest self-disclosure
• Authentic self-presentation.

As part of this paper being a registered report, the authors specified in advance the effect size that they would consider as theoretically meaningful. This means that even if results were positive and statistically significant, the authors would not consider them as meaningful unless the effect was at least g = 0.4, corresponding to a small-to-medium effect size. This threshold was based on previous research on everyday behaviours (e.g., sleep, physical activity) and their link to mental health.

The authors used two complementary statistical approaches:

1. First, they used conventional null hypothesis significance testing to ask: is there evidence of a statistically significant difference between adolescents with and without mental health conditions?
2. Second, they used equivalence testing, which asks: are there meaningful differences between adolescents with and without mental health conditions?

Results

Sample characteristics

The sample included 3,340 adolescents aged 11-19 years (M = 14.77, SD = 2.48) who used social media. The sample had an even split of male and female participants. Among them, 16% (n = 519) had at least one mental health condition, including 8% (n = 282) with an internalising condition and 3% (n = 104) with an externalising condition.

Any mental health condition vs no condition

• Adolescents with any mental health condition reported spending more time on social media than those without a condition, and this difference was meaningful (g = 0.46, 90% CI [0.38 to 0.54]).
• They were also less happy about the number of online friendships (g = -0.37, 90% CI [-0.45 to -0.29]).
• However, there were no meaningful differences in relation to any other dimensions of social media engagement, such as online social comparison and the impact of online feedback on their mood.

Internalising versus no conditions

• Compared to adolescents with no condition, those with internalising conditions:
  • reported spending more time on social media (g = 0.62, 90% CI [0.51 to 0.73]),
  • engaged in more online social comparison (g = 0.54, 90% CI [0.43 to 0.65]),
  • felt more affected by online feedback (g = 0.38, 90% CI [0.27 to 0.49]),
  • were less happy about the number of online friendships (g = -0.45, 90% CI [-0.55 to -0.35]), and
  • were less likely to have honest self-disclosure (g = -0.31, 90% CI [-0.42 to -0.20]).
• They also reported greater lack of control over time spent online (g = 0.43, 90% CI [0.33 to 0.55]); this was interesting because the authors had expected this pattern to be more characteristic of externalising conditions.

Externalising versus no conditions

• By contrast, the only meaningful differences between adolescents with and without an externalising condition was:
  • time spent on social media (g = 0.31, 90% CI [0.13 to 0.48]) and
  • the impact of feedback on mood (g = 0.27, 90% CI [0.10 to 0.45]).

Internalising versus externalising conditions

• When comparing the two clinical groups directly, adolescents with internalising conditions:
  • engaged in more online social comparison (g = 0.64, 90% CI [0.45 to 0.85]),
  • were less happy about their online friendships (g = -0.32, 90% CI [-0.51 to -0.14]), and
  • spent more time on social media (g = 0.27, 90% CI [0.07 to 0.47]), than those with externalising conditions.

Conclusions

Overall, this study suggests that adolescents with mental health conditions not only spend more time on social media compared to their peers, but they also engage with social media differently, especially for those with internalising conditions. The authors concluded that:

This highlights aspects of social media use that might present an increased risk to this already vulnerable group and provides a window for future research to ensure that the digital world is safe for all children regardless of mental health status.

Strengths and limitations

Strengths

• This study used a nationally representative sample, which strengthens confidence that these findings reflect the broader experience of young people in the UK, rather than a selective group. In turn, this makes the findings more generalisable.
• The use of a standardised diagnostic assessment more reliably captures participants with a mental health condition compared to self-reported questionnaire scores. This is important because much of the social media literature relies on symptom scores from community samples, making it difficult to determine whether findings apply to young people with a mental health condition.
• This study was pre-registered, meaning that the methods and analysis plan were peer reviewed before the results were known. The authors pre-defined the smallest effect size of interest, which allowed them to conclude whether findings are theoretically meaningful.
• The inclusion of both quantitative and qualitative measures of social media captures not only how much time adolescents spend on social media, but how they engage with social media. This offers a more nuanced picture of adolescents’ digital lives, which may be particularly relevant for understanding links with mental health.

Limitations

• As acknowledged by the authors, the sample is cross-sectional, meaning that we cannot draw conclusions about causality or directional inference.
• The data used was collected in 2017. Social media platforms, norms, and features have changed substantially in the last decade; therefore, the results need to be interpreted within their context.
• Time spent on social media was measured by self-report. While it is useful, recent studies have shown that young people tend to underestimate their time spent on social media (Lind et al., 2023), meaning it may not be as accurate.
• Some subgroup analyses should be interpreted with caution. The group with externalising conditions was relatively small (n = 104), which may have limited power to detect smaller effects.

Implications for practice

An important finding from this registered report is that mental health conditions should not be treated as a single category when thinking about social media. The results show that young people with internalising and externalising conditions have very different experiences on social media. For example, while adolescents with externalising conditions seem to be spending more time on social media than their peers who do not have a mental health condition, those with internalising conditions also differed meaningfully in how they engage with it. As the authors highlighted, psychoeducation and cognitive-behavioural reappraisal techniques aimed at online social comparison and online feedback could be particularly beneficial for adolescents with internalising conditions (Tibber & Silver, 2022).

A second important takeaway is that difficulties with peer relationships, commonly experienced by young people with mental health conditions offline (Finsaas et al., 2020), appear to extend into their online lives too. This finding is in contrast with the assumption that young people who struggle with offline friendships find connection and community through social media (Bonetti et al., 2010). In fact, the nature of friendship may have fundamental aspects that are similar both in person and online, such as reciprocity, trust, healthy self-disclosure, and conflict resolution. It is possible that social media may exacerbate such vulnerabilities rather than offering an alternative space for connection for some young people with mental health conditions. As adolescents increasingly move fluidly between digital and in-person social worlds, strengthening these foundational skills may help support connection across both contexts. Therefore, interventions that support young people to develop better interpersonal skills that transfer across contexts may be particularly useful.

Future research should move beyond asking whether social media is simply “good” or “bad” for adolescent mental health and instead investigate which young people are most vulnerable, under what circumstances, and through which mechanisms. These questions are particularly relevant to current policy debates, including Australia’s world-first Social Media Minimum Age (SMMA) legislation implemented in December 2025.

Restricting access to social media may be an important step toward protecting youth mental health by reducing exposure to potentially harmful online environments. However, this study suggests that adolescents with internalising conditions differed not only in how much they used social media, but also in how they experienced it; including heightened social comparison and sensitivity to online feedback. Policies aimed solely at limiting access may therefore miss some of the underlying psychological and interpersonal processes linked to risk. Alongside broader regulatory approaches, supporting digital literacy, healthy peer relationships, emotional literacy, and adaptive responses to online social comparison may be equally important targets for future policy and intervention.

Statement of interests

Sylvia Lin and Monika Raniti declare no conflicts of interest. 

Edited by

Dr Nina Higson-Sweeney.

Links

Primary paper

Luisa Fassi, Amanda M. Ferguson, Andrew K. Przybylski, Tamsin J. Ford & Amy Orben (2025). Social media use in adolescents with and without mental health conditions. Nature Human Behaviour, 9(6), 1283–1299. https://doi.org/10.1038/s41562-025-02134-4

Other references

Achenbach, T. M., Ivanova, M. Y., Rescorla, L. A., Turner, L. V., & Althoff, R. R. (2016). Internalizing/Externalizing Problems: Review and Recommendations for Clinical and Research Applications. Journal of the American Academy of Child & Adolescent Psychiatry, 55(8), 647–656. https://doi.org/10.1016/j.jaac.2016.05.012

Bonetti, L., Campbell, M. A., & Gilmore, L. (2010). The Relationship of Loneliness and Social Anxiety with Children’s and Adolescents’ Online Communication. Cyberpsychology, Behavior, and Social Networking, 13(3), 279–285. https://doi.org/10.1089/cyber.2009.0215

Finsaas, M. C., Kessel, E. M., Dougherty, L. R., Bufferd, S. J., Danzig, A. P., Davila, J., Carlson, G. A., & Klein, D. N. (2020). Early Childhood Psychopathology Prospectively Predicts Social Functioning in Early Adolescence. Journal of Clinical Child & Adolescent Psychology, 49(3), 353–364. https://doi.org/10.1080/15374416.2018.1504298

Lind, M. N., Kahn, L. E., Crowley, R., Reed, W., Wicks, G., & Allen, N. B. (2023). Reintroducing the Effortless Assessment Research System (EARS). JMIR Mental Health, 10(1), e38920. https://doi.org/10.2196/38920

Orben, A., Meier, A., Dalgleish, T., & Blakemore, S.-J. (2024). Mechanisms linking social media use to adolescent mental health vulnerability. Nature Reviews Psychology, 3(6), 407–423. https://doi.org/10.1038/s44159-024-00307-y

Sabo, A., & La Sala, L. (2025). Critical lack of evidence about social media use and youth mental health in clinical populations? The Mental Elf.

Teague, S., Somoray, K., Shatte, A., Miller, D., Moss, K., Crawford, A., Wildman, H., Kayal, D., & Hutchinson, D. (2026). Digital Media Use and Child Health and Development: A Systematic Review and Meta-Analysis. JAMA Pediatrics. https://doi.org/10.1001/jamapediatrics.2026.0085

Tibber, M. S., & Silver, E. (2022). A trans-diagnostic cognitive behavioural conceptualisation of the positive and negative roles of social media use in adolescents’ mental health and wellbeing. The Cognitive Behaviour Therapist, 15, e7. https://doi.org/10.1017/S1754470X22000034

Photo credits

• Photo by Getty Images on Unsplash
• Photo by Julian on Unsplash
• Photo by Vitaly Gariev on Unsplash
• Photo by Paul Marlow on Unsplash
• Photo by manas rb on Unsplash