social media syndrome

SMS

Not a diagnosis. Thirteen things keep turning up in the research. Nobody has checked whether they turn up in the same person.

There is already a scale

Social Media Disorder counts the using — nine criteria, and five of them means you qualify.11

That scale is not a formal diagnosis anywhere. And it measures only how much, which says nothing about what the using leaves behind. This page is the other half.

Every study here measures one sign, alone. Nobody has tested whether the signs happen to the same person.

If any of this frightens you, talk to a doctor. Under each sign, three marks — the more of them filled, the harder the evidence.

the mind

Four at once

Depression, anxiety, self-harm and substance use all rose in the same Italian hospital records — the one place where the signs are seen travelling together. One study, though, not four.1

An American study of one social network’s college rollout — 430,000 responses — found the first two: severe depression up seven percent, anxiety up twenty.2

— hospital records, exam results, minutes counted

the body

Sleep

Spain's fibre rollout cut adolescents' sleep. Girls carried almost all of it.3

— hospital records, exam results, minutes counted

the body

Your face stops being enough

144 girls, randomised. Filtered photographs lowered body image on the spot.6

— randomised, small sample

the body

And then you ask a surgeon

Fifty-five percent of facial plastic surgeons report patients asking to look better in their selfies. Two years earlier it was forty-two.8

— observed, not demonstrated

the day

Time you did not choose

Given a cap people could set themselves, they used much less — and the authors estimate self-control problems drive 31 percent of all social media use.5

— hospital records, exam results, minutes counted

the day

Falling behind

Schools that banned phones saw scores rise. The bottom quarter gained twice the average; the top gained nothing.4

— hospital records, exam results, minutes counted

the day

The phone only has to be there

Face down and silent, your own phone still costs you a little memory. Memory only — attention did not hold.12, 13

— randomised, small

the day

Pulling away

Less time with family. Less with friends.3

— hospital records, exam results, minutes counted

the edges

Money, through the same door

The more someone spends on loot boxes — paid random rewards inside games — the more problem gambling they report. Fifteen studies, a small-to-moderate link.7, 14

— real, direction unknown

the edges

Identity that will not settle

Thirty-two studies, nearly twenty thousand adolescents. What you do there matters more than how long you stay — and comparing yourself raises self-exploration and distress at once.9

— direction unclear

every generation panics about a machine

Zero

Broadband reached Italian towns at different times. Those who got it before they were twenty turned up in hospital more often. For those born ten years earlier, the effect was zero. Same country, same hospitals, same years.1

against this page

Potatoes

Six hundred million defensible ways to analyse three national datasets. Screen time explained at most 0.4 percent of a teenager’s wellbeing — between asthma and eating potatoes. Bullying was four times worse.15

That is self-reported hours. The hardest rows here measure a hospital admission, an exam result, a counted minute. The weakest are exactly what this dissolves.

not one of the thirteen

The binge that is not scrolling

In adults, binge-watching and loneliness go together.10 The same hollow turns up in people who never opened the apps. The substance is interchangeable. The shape is not.

One sign at a time

Nobody has asked whether the one who lost the sleep is the one who edited the selfie, who bought the loot box, who stopped seeing friends.

Every study above measured one sign, on its own. None measured two. Until one does, this is a suspicion, not a finding.

Sources

  1. Donati, D., Durante, R., Sobbrio, F. & Zejcirovic, D. (2025). Lost in the net? Broadband internet and youth mental health. Journal of Health Economics, 103, 103017.
  2. Braghieri, L., Levy, R. & Makarin, A. (2022). Social Media and Mental Health. American Economic Review, 112(11), 3660–3693.
  3. Arenas-Arroyo, E., Fernández-Kranz, D. & Nollenberger, N. (2025). High speed internet and the widening gender gap in adolescent mental health: Evidence from Spanish hospital records. Journal of Health Economics, 102, 103014.
  4. Beland, L.-P. & Murphy, R. (2016). Ill Communication: Technology, distraction & student performance. Labour Economics, 41, 61–76.
  5. Allcott, H., Gentzkow, M. & Song, L. (2022). Digital Addiction. American Economic Review, 112(7), 2424–2463.
  6. Kleemans, M., Daalmans, S., Carbaat, I. & Anschütz, D. (2018). Picture Perfect: The Direct Effect of Manipulated Instagram Photos on Body Image in Adolescent Girls. Media Psychology, 21(1), 93–110.
  7. Garea, S. S., Drummond, A., Sauer, J. D., Hall, L. C. & Williams, M. N. (2021). Meta-analysis of the relationship between problem gambling, excessive gaming and loot box spending. International Gambling Studies, 21(3), 460–479. Fifteen studies of problem gambling; r = 0.26, rising to 0.37 under a trim-and-fill correction, which the authors call small-to-moderate. Every included study is cross-sectional, and they say plainly that they cannot determine the causal direction — loot boxes may be producing the harm, or already-vulnerable people may be the ones buying. That sentence is why this row is graded weakest.
  8. Rajanala, S., Maymone, M. B. C. & Vashi, N. A. (2018). Selfies—Living in the Era of Filtered Photographs. JAMA Facial Plastic Surgery, 20(6), 443–444. This one is a viewpoint, not a study, and the surgeon percentages come from trade-association surveys of their own members — which is why that row is graded weakest.
  9. Avci, H., Baams, L. & Kretschmer, T. (2025). A Systematic Review of Social Media Use and Adolescent Identity Development. Adolescent Research Review, 10(2), 219–236. The panel study is Valkenburg, P. M., Koutamanis, M. & Vossen, H. G. M. (2017). The concurrent and longitudinal relationships between adolescents' use of social network sites and their social self-esteem. Computers in Human Behavior, 76, 35–41: three waves, 852 adolescents aged 10–15, in which self-esteem longitudinally predicted more use while use only marginally predicted improvements in self-esteem — the opposite direction from the one a page like this would want.
  10. Sun, J. J. & Chang, Y. J. (2021). Associations of Problematic Binge-Watching with Depression, Social Interaction Anxiety, and Loneliness. International Journal of Environmental Research and Public Health, 18(3), 1168.
  11. van den Eijnden, R. J. J. M., Lemmens, J. S. & Valkenburg, P. M. (2016). The Social Media Disorder Scale. Computers in Human Behavior, 61, 478–487. Nine criteria adapted from the DSM-5's Internet Gaming Disorder, validated on 2,198 Dutch adolescents aged 10–17. Neither social media disorder nor this scale is a formal diagnosis: Internet Gaming Disorder sits in the DSM-5's Section III, conditions for further study, and only Gaming Disorder reached ICD-11. Some researchers argue the addiction framework over-pathologises ordinary behaviour — that argument is live, and this page does not settle it.
  12. Ward, A. F., Duke, K., Gneezy, A. & Bos, M. W. (2017). Brain Drain: The Mere Presence of One's Own Smartphone Reduces Available Cognitive Capacity. Journal of the Association for Consumer Research, 2(2), 140–154.
  13. Böttger, T., Poschik, M. & Zierer, K. (2023). Does the Brain Drain Effect Really Exist? A Meta-Analysis. Behavioral Sciences, 13(9), 751. Twenty-two studies, 43 effects. Pooled g = −0.14 (95% CI −0.24 to −0.03), with no sign of publication bias. Only memory held up (g = −0.23, p < 0.001); attention (p = 0.29) and general cognitive performance (p = 0.76) did not. Split by region, only the Asian studies reached significance (g = −0.39); Europe (p = 0.12) and North America (p = 0.60) did not — so the pooled figure is carried by one part of the world, and the honest reading of this row is that something small is there and we do not yet know its shape.
  14. Zendle, D. (2019). Problem gamblers spend less money when loot boxes are removed from a game: a before and after study of Heroes of the Storm. PeerJ, 7, e7700. The one study here that speaks to direction: a game withdrew loot boxes, and problem gamblers then spent significantly less in it than other players did. It is a before-and-after of 112 players, and the author calls the pattern complex and says only that loot boxes may be the cause — so the row keeps its weakest grade. One small study pointing at a direction is not the same as knowing it.
  15. Odgers, C. L. & Jensen, M. R. (2020). Annual Research Review: Adolescent mental health in the digital age — facts, fears, and future directions. Journal of Child Psychology and Psychiatry, 61(3), 336–348. The strongest case against a page like this one, and it belongs here rather than in a footnote: their synthesis is that most of the research is correlational, mixed in direction, and that the most rigorous large-scale preregistered studies find associations too small to be of clinical or practical significance. The 0.4 per cent is Orben and Przybylski (2019), read direct.16 The daily figure comes from roughly 400 young people, 13,017 observations over 5,270 days: no evidence that mental health was worse on heavier days, and where associations did appear they ran the other way. One more thing worth knowing about that whole literature — self-reported screen time matches measured screen time at about r = .20, and of the 29 studies they tabulate, two used an objective measure at all.
  16. Orben, A. & Przybylski, A. K. (2019). The association between adolescent well-being and digital technology use. Nature Human Behaviour, 3(2), 173–182. Specification-curve analysis over three national datasets: 372 defensible ways to analyse one, 40,966 another and 603,979,752 the third. Their own summary is that the association is negative but small, explaining at most 0.4% of the variation in wellbeing. The benchmarks are the useful part. Against a median β of −0.049 for technology use, being bullied ran −0.212, binge-drinking −0.144, asthma −0.066, eating potatoes −0.042 — and on the other side, eating breakfast +0.116 and getting enough sleep +0.150. Sleep is associated with three times more wellbeing than screens are associated with less.

Unsigned on purpose. Every number has a source you can go and check, and the sign closest to the writer's own trade is graded the weakest here.

← mutinoise