Think about the next situation. You’re mendacity in mattress, telephone in hand, scrolling via TikTok. A video pops up in your For You feed: “5 indicators you could have ADHD.” One other video claims trauma rewires your mind in methods therapists received’t inform you about. It feels relatable, maybe even reassuring – however is it correct, and may it’s trusted?
Social media platforms corresponding to Fb, Instagram, TikTok and YouTube have turn into a serious supply of psychological well being data, notably for younger individuals. They provide fast solutions, shared experiences, and validation (Loades et al., 2025). However as earlier Psychological Elf blogs have highlighted, the identical platforms may amplify misery and unfold false or deceptive data – in any other case referred to as misinformation (learn blogs by Margherita and Sarah).
Well being-related misinformation is more and more frequent on social media, with earlier research advised 80% of health-related content material is misinformation, and that it’s extra frequent than correct well being data (Suarez-Lledo & Alvarez-Galvez 2021; Wang et al., 2019). However what about content material that’s particular to psychological well being and neurodivergence?
To handle this hole within the literature, Carter and colleagues (2026) carried out a scientific evaluate to grasp how frequent psychological well being and neurodivergence-related misinformation is on social media, in addition to assess the accuracy, high quality and reliability of the data discovered.
Psychological well being and neurodivergence-related content material on social media will be skilled as reassuring and validating, however is it all the time correct, and may it’s trusted?
Strategies
The authors searched 4 databases for articles written in English that evaluated the standard and/or accuracy of psychological well being and neurodivergence-related data on social media. Research have been screened by one creator on the title and summary stage and the full-text stage, with 25% double-screened by one other creator; this course of was repeated for information extraction. Google Scholar and the reference lists of included articles have been additionally searched to establish any lacking papers.
Research high quality was assessed utilizing a instrument developed in a earlier evaluate on well being misinformation on social media (Suarez-Lledo & Alvartez-Galvez, 2021). Once more, one creator critically appraised all research, with 25% appraised by a second creator. The imply high quality ranking for included research was 65%, indicating good high quality; nonetheless, research ranged from 41% (poor high quality) to 80% (prime quality).
Outcomes
Research traits
Twenty-seven research have been included on this systematic evaluate, with the bulk evaluating YouTube (n = 18) and TikTok (n = 5). Nearly a 3rd of research targeted on neurodivergence, particularly autism (n = 4) and attention-deficit hyperactivity dysfunction (ADHD; n = 4), with the remainder exploring varied psychological well being diagnoses, together with anorexia (n = 3), bipolar (n = 2), and obsessive compulsive dysfunction (OCD; n = 2). A complete of 5,057 social media posts have been analysed throughout research.
What’s misinformation?
13 research supplied clear definitions of misinformation, most of which outlined it as, “content material which contained factually inaccurate and/or scientifically unsubstantiated claims”.
How a lot data is on social media?
Prevalence charges for misinformation have been reported in 17 research. Misinformation was highest on TikTok (35%), whereas misinformation on YouTube was usually decrease (22%), though this did differ by subject (e.g., 6.7% for dissociative id dysfunction, 57% for MRI claustrophobia). YouTube Youngsters had the bottom price of misinformation, with 0% for nervousness and melancholy and 9% for ADHD. The imply prevalence of misinformation on Fb was 15% (n = 2), and the reported prevalence of misinformation on X/Twitter was 19% (n = 1). Typically, misinformation was extra frequent for neurodivergence than psychological well being situations.
What’s the reliability and high quality of data on social media?
YouTube content material was usually extra dependable and of upper high quality than different social media platforms. Nevertheless, this was not constant and doesn’t essentially imply that the content material was good high quality or dependable.
Content material created by professionals was often extra dependable and better high quality than content material by non-professionals; nonetheless, some research did recommend that skilled and patient-created content material was equally dependable.
Psychological well being and neurodivergence-related misinformation seem like highest on TikTok and customarily decrease on YouTube, with only a few research contemplating misinformation on Fb, X/Twitter, or Instagram.
Conclusions
Carter et al. (2026) conclude that the reliability and high quality of psychological well being and neurodivergence-related data on social media is very variable, each between and inside platforms, and that this could possibly be as a consequence of a wide range of causes.
Curiously, the authors spotlight that, “this variability means that platform-specific elements, corresponding to algorithmic programs and content material moderation, might affect the unfold of misinformation”, with the search-based designs of YouTube and Fb maybe being much less problematic than algorithm-driven TikTok, however rather more analysis is required to correctly perceive what drives misinformation on social media, and what we will do to mitigate it.
The standard of psychological well being and neurodivergence-related data varies extensively throughout platforms, and algorithm-driven feeds might play a key function in shaping what individuals see.
Strengths and limitations
It is a well-conducted systematic evaluate that addresses a sequence of necessary and well timed questions on psychological well being and neurodivergent-related misinformation on social media, offering a beneficial contribution to the literature. The evaluate was pre-registered on the Open Science Framework (however not PROSPERO, which is extra generally used) and adhered to acceptable tips, guaranteeing clear and correct reporting, which will increase its reliability.
Nevertheless, I can’t assist however query why solely 25% of articles have been double screened at every stage, and inter-rater reliability statistics weren’t calculated or reported. We’re more and more seeing this in systematic opinions as a consequence of pragmatics – 100% double screening is a process that takes a substantial period of time. However in a technique that’s recognized for its rigour exactly due to this thoroughness, I can’t assist however marvel if doubtlessly eligible and informative research have been missed, and if this evaluate is as complete because it might have been. Then once more, the search technique for the evaluate was fairly broad, with no restrictions on research sort, inhabitants, or publication date – however this does make it much more crucial that 100% of articles have been double screened, because the inclusion standards was comparatively open to interpretation.
The included research have been extremely heterogeneous, making direct comparability tough, and there was a transparent platform imbalance, with 18 research assessing YouTube and solely two assessing Fb, for instance. This necessitates warning when deciphering the findings from this evaluate; whereas we will be extra assured of the findings in relation to YouTube, findings from different platforms appear tentative at greatest.
Lastly, as is commonly the case with systematic opinions, there are methodological weaknesses related to the included research that have an effect on the evaluate itself. Whereas the imply ranking of research high quality was 65%, indicating good high quality, it did dip as little as 41%. The authors present a useful desk summarising the research high quality rankings, and it looks as if research high quality was usually lowered by the search technique, corresponding to mentioning search instruments, utilizing multiple search engine, and reporting preliminary hits, which limits replicability and general transparency. Extra high-quality analysis is required on this space.
Whereas this can be a well-conducted and well timed systematic evaluate, the shortage of inter-rater reliability statistics and restricted proof throughout platforms means findings ought to be interpreted with warning.
Implications for follow
Whereas the findings from this systematic evaluate must be handled with warning as a result of general lack of knowledge that may be drawn on, there are fascinating implications for anybody concerned in public psychological well being. The necessary factor to recollect is that persons are already utilizing these platforms for data on psychological well being and neurodivergence, no matter reliability or high quality. Now that we’ve got some synthesised details about which platforms are usually roughly dependable, we will take into consideration how we’d enhance the reliability of this data, or how we will steer individuals in direction of extra correct, useful content material.
For clinicians, this research is a crucial reminder concerning the potential affect of social media on service customers, and the sorts of conversations that will come up in follow. As just a few of us at The Psychological Elf wrote in a current debate article, it is very important have open and sincere conversations with people who’re sharing or imagine misinformation, guaranteeing that it’s approached with out judgement of dismissal (Higson-Sweeney et al., 2026). There are quite a few methods for clinicians to assist service customers to suppose critically concerning the data they could see on social media, and to seek out content material that it relatable and validating, in addition to correct and dependable.
For researchers, this reinforces the truth that sharing evidence-based findings with most of the people is a crucial a part of the analysis course of; if you happen to’re not sharing your findings, then maybe another person will, and it is probably not as correct. There may be additionally a necessity for additional analysis on this space. Personally, I might additionally like to see larger-scale analysis that focuses on particular psychological well being and neurodivergence-related misinformation on social media platforms. Analysis ought to straight examine misinformation throughout situations, platforms and forms of content material (e.g., written posts, short-form movies, long-form movies), in addition to contemplating how individuals interpret and use this data – simply because it’s obtainable doesn’t imply persons are taking it at face worth (Loades et al., 2025).
For policymakers, there’s a have to develop clearer requirements for social media platforms concerning the moderation of health-related data, together with content material on psychological well being and neurodivergence. Complete definitions relating to what counts as misinformation and the way it ought to be addressed (together with discussions of algorithms) may assist to scale back the quantity of misinformation that’s circulated or at the least immediate these participating with this content material to be vital and to not essentially imagine every part they’re studying.
Lastly, I believe it is very important caveat that misinformation shouldn’t be conflated with sharing lived expertise. I wrote a weblog final yr about portrayals of CAMHS on TikTok, and whereas the movies analysed might not have introduced a complete image of the service and accounted for various views, it nonetheless mirrored how these younger individuals felt and what they’d skilled. Once more, I believe there’s a have to make clear what misinformation is and isn’t, to be sure that correct data is shared with out dismissal or disempowerment.
Persons are already utilizing social media to hunt out psychological well being and neurodivergence-related data, so what can we do to mitigate the impression of misinformation on these platforms?
Assertion of pursuits
Nina Higson-Sweeney incessantly collaborates with one of many authors of the present research, however had no data or involvement on this research. Past this, she has no different conflicts of curiosity to declare.
Hyperlinks
Main paper
Alice Carter, Fergus Gracey, Joanna Moody, Amber Ovens, & Eleanor Chatburn. (2026). High quality, reliability and misinformation in psychological well being and neurodivergence content material on social media: a scientific evaluate. Journal of Social Media Analysis, 3(1), 30-47. https://doi.org/10.29329/jsomer.84
Different references
Hetrick, S. (2018). Social media: good and unhealthy experiences and the impression on melancholy. The Psychological Elf.
Higson-Sweeney, N. (2025). “I don’t want a cup of tea, I would like some @#$%&! assist”: #camhs via the lens of TikTok. The Psychological Elf.
Higson‐Sweeney, N., Badenoch, D., & Tomlin, A. (2026). Debate: Standing up for science–methods to fight misinformation in youngster psychological well being? 5 suggestions for disentangling truth from fiction. Little one and Adolescent Psychological Well being, 31(1), 74-76. https://doi.org/10.1111/camh.70055
Loades, M. E., Higson‐Sweeney, N., Teague, B., Leas, J., Payne‐Prepare dinner, C., Slastikova, A. V., … & Biddle, L. (2025). What do they search for and what do they discover? A coproduced qualitative research on younger individuals’s experiences of looking for psychological well being data on-line. Psychology and Psychotherapy: Principle, Analysis and Observe, 98(2), 373-395. https://doi.org/10.1111/papt.12550
Suarez-Lledo, V., & Alvarez-Galvez, J. (2021). Prevalence of well being misinformation on social media: systematic evaluate. Journal of Medical Web Analysis, 23(1), e17187. https://doi.org/10.2196/17187
Wang, Y., McKee, M., Torbica, A., & Stuckler, D. (2019). Systematic literature evaluate on the unfold of health-related misinformation on social media. Social Science & Drugs, 240, 112552. https://doi.org/10.1016/j.socscimed.2019.112552
Zenoni, M. (2021). Social media peer assist teams for OCD and associated problems: useful or dangerous? The Psychological Elf.
