Social Media and Eating Disorder Risk: What Research Shows

Author: Angela Bogdanova · Published: October 10, 2026 · Editorial Policy

Research cluster: Eating Disorders & Disordered Eating · Written by Artificial Sapiens

Social media is associated with eating-disorder symptoms, particularly when users repeatedly encounter appearance ideals, pressure to change their bodies, and content that normalizes risky eating or exercise practices. The research is more precise than the familiar headline that social media causes eating disorders. A 2026 systematic review and meta-analysis of 45 studies including 33,086 adolescents found a small overall association between social media use and body-image or eating-related outcomes; a separate meta-analysis of social comparison found a moderate association with eating-disorder symptoms. Neither result means an individual user's exposure is a clinical diagnosis or that all platform use produces harm. Fortunato et al., 2026; Bonfanti et al., 2025.

The evidence points toward differences in what people see, how they engage with it, their developmental context, and preexisting vulnerabilities. Exposure to idealized fitness or appearance imagery can affect body-related feelings in short-term experiments, while the path from an online experience to a diagnosed eating disorder remains considerably more difficult to demonstrate. The 2026 review focused on young adults included only seven eligible studies that directly addressed eating disorders as diagnostic entities. This distinction between measured symptoms and established clinical diagnoses is fundamental to interpreting the field.

What Does Research Actually Show About Social Media and Eating Disorders?

Current studies support a link between some forms of social media engagement and disordered eating, while the strength and interpretation of that link depend on the exposure and outcome being measured. In research, 'social media use' might mean self-reported minutes per day, frequency of opening a platform, appearance-related comparison, viewing particular videos, posting photos, engagement with online communities, or problematic use. These behaviors should be analyzed separately because they involve different psychological experiences.

Likewise, 'eating-disorder risk' is used in several ways. Some researchers measure attitudes such as concern about body shape. Others measure behavioral symptoms, such as restrictive rules or episodes of loss-of-control eating. A smaller body of research examines clinically assessed disorders. These outcomes have different meanings, and correlations involving body dissatisfaction cannot be interpreted as the number of people who developed anorexia nervosa, bulimia nervosa, or binge eating disorder.

The 2026 adolescent meta-analysis

In Fortunato and colleagues' 2026 meta-analysis, 45 studies involving 33,086 adolescents aged 13–18 contributed data on social media and body-image or eating outcomes. Across outcomes, the overall pooled association was approximately r = .19. For eating-disorder symptoms specifically, the pooled association across 11 studies was approximately r = .15. These are small average correlations, meaning the exposure accounted for a limited share of individual variation. The studies included cross-sectional and longitudinal designs and differed substantially in how they measured social media and clinical outcomes.

The review found stronger associations in research focused on Instagram than in research combining platforms, and some associations varied with the percentage of girls in a study. Those are study-level moderator findings. They do not establish that Instagram is intrinsically more dangerous than every alternative platform, and they cannot predict the effect of any one user's feed. The reviewed material was collected through April 2024, so the publication year should not be confused with a 2026 sample of platform behavior.

Social comparison has a closer association with symptoms than simple exposure totals

A different question is whether people who compare their appearances with others online report more eating-disorder symptoms. Bonfanti and colleagues' meta-analysis synthesized 83 studies and 55,440 participants, finding a pooled correlation of r = .36 between online social comparison and eating-disorder symptoms. Its pooled correlation with body-image concerns was r = .454. These values describe associations between specific constructs in a large literature, not a direct estimate of future disease caused by social media.

The two meta-analyses therefore answer different questions. The smaller average correlation for broad use and the larger correlation for appearance comparison can coexist. Specific cognitive engagement may correlate more closely with problematic eating than the amount of time a person reports spending online. The evidence does not justify treating all comparisons as causal or assuming every person who compares their body will experience a disorder.

The clinical-disorder evidence is narrower

A 2026 systematic review by Fomínová and colleagues specifically selected research on social media and eating disorders in young adults aged 18–30, rather than treating generic body-image concern as a diagnosis. Only seven studies met its eligibility criteria. Six reported a positive association between social media and eating-disorder symptoms or diagnoses, while one did not. Most designs were cross-sectional, and one was experimental. The review suggests clinically relevant concern but also reveals how limited the direct diagnostic evidence remains.

Can Social Media Cause an Eating Disorder?

Social media can contribute to experiences and behaviors implicated in eating-disorder risk, and short-term causal effects have been demonstrated for some narrower outcomes, especially body dissatisfaction following exposure to appearance-ideal images. Research has not established a simple universal pathway in which social media exposure alone produces a diagnosable eating disorder. Eating disorders arise through interacting biological, psychological, social, developmental, and environmental influences, described in the National Institute of Mental Health's clinical overview and our separate article on what causes eating disorders.

Three kinds of evidence must be distinguished. Cross-sectional studies observe exposure and outcomes at the same time, so they cannot determine which came first. Longitudinal studies test whether earlier use predicts later symptoms, often adjusting for earlier symptom levels, but still face confounding and changes in exposure. Experiments randomly assign short-term content exposure and can test immediate responses, but a brief change in mood, comparison, or body satisfaction is not equivalent to a new DSM-5-TR eating-disorder diagnosis.

In De Valle and colleagues' meta-analyses of experimental and longitudinal studies, exposure to social media appearance-ideal images had a moderate short-term negative impact on body image in experimental comparisons, whereas the longitudinal association between general social media use and subsequent body image was very small. Those patterns support a plausible contribution from content, especially appearance ideals, and caution against overstating an effect for general use or a clinical diagnosis. Other causes, protective factors, and selection into particular feeds remain relevant.

Why Some Feeds May Increase Eating-Disorder Risk

Appearance comparison and internalization of ideals

People can encounter idealized bodies repeatedly through photos, videos, advertising, influencer content, and peer posts. Comparison can become personally consequential when an image is taken as a standard against which someone evaluates their own appearance or worth. Internalization means adopting such a standard as a personally important goal. These processes may contribute to body dissatisfaction and eating concerns for some users, especially when social feedback reinforces attention to appearance. Bonfanti et al., 2025; Fortunato et al., 2026.

This mechanism overlaps with the field of body-image psychology, which has its own canonical treatment in Social Media and Body Image. The eating-disorder-specific concern starts when body-focused pressure connects to rigid dietary rules, distress about eating, preoccupation with food or weight, compensatory behavior, or clinically important impairment. Appearance dissatisfaction alone does not establish an eating disorder. For the broader comparison process, see Social Comparison on Social Media.

Fitness, wellness, diet, and 'healthy living' content

Health and fitness language can convey useful information, but appearance-centered fitness content sometimes promotes rigid or misleading ideals. In a 2024 content analysis of 200 TikTok fitspiration videos, researchers identified idealized physiques, objectifying themes, and incorrect or harmful diet and exercise information in many sampled videos. Because the sample consisted of selected popular fitness hashtags, these proportions should not be generalized to all fitness creators or to the platform as a whole.

A newer 2026 meta-analysis of 26 experimental studies involving 6,111 participants reported that fitspiration exposure among young adults increased comparison and unhealthy dieting or exercise motivations while worsening several body-image-related outcomes. These are causal experimental findings for short-term responses and motivations. They do not show that every fitness video is harmful or that the experiments documented the incidence of clinical eating disorders.

Normalization of harmful behavior and repeated exposure

Some posts frame eating-disordered behaviors as achievements, identity markers, discipline, or social belonging. Other posts may use apparently ordinary health language while encouraging persistent fear, guilt, or rigid rules around food. Repetition and social reinforcement can make messages seem ordinary or normative. The American Psychological Association's adolescent social media advisory recommends minimizing minors' exposure to content that encourages restrictive eating, purging, or excessive exercise and emphasizes media literacy and age-sensitive safeguards.

Recommendation feeds may amplify material that a user has viewed or interacted with, and eating-disorder-related searches may affect later exposures. However, proprietary recommendation systems are difficult for independent researchers to observe consistently, and a particular person's feed cannot be reconstructed from general claims about an algorithm. A defensible explanation examines both the types of content and the user's interactions without attributing every symptom to a single technical mechanism.

Social approval, self-monitoring, and persistent evaluation

Likes, comments, repeated photo checking, appearance-editing, and audience feedback can reinforce self-evaluation through appearance. For some people, the need to monitor how their body looks may spill into eating rules or exercise behavior. For others, social media primarily supports friendship, creativity, and community, with little appearance investment. The APA advisory identifies appearance-oriented comparison and photo-related feedback as relevant risk pathways while emphasizing that social media experiences vary with developmental and individual context.

Is It Social Media Time or the Content That Matters?

Screen time is an imprecise exposure measure. Two people can spend the same amount of time online while encountering entirely different content and emotional experiences. One may talk with supportive peers and watch unrelated entertainment; another may repeatedly compare bodies or encounter messages that intensify food-related distress. The total time figure cannot distinguish these exposures and is a poor standalone clinical screening tool.

The 2026 adolescent synthesis reported associations for both general time-based measures and appearance-related use, with heterogeneity between studies. Some specialized exposures show stronger links than broad usage averages, but research also faces self-report errors, correlated mental-health vulnerabilities, differences in age and gender composition, and changing platforms. It is therefore useful to ask what appears on a feed, how it is interpreted, and whether there are changes in eating, physical health, or functioning, rather than applying a universal daily-minute threshold.

Which Eating-Disorder Symptoms Are Relevant?

Social media research commonly assesses disordered-eating symptoms rather than confirmed diagnoses. Concerning changes can include increasing rigidity around food, intense distress after eating, secretive or socially isolating eating practices, persistent preoccupation with weight or shape, loss-of-control eating, or compensatory behaviors. These signs require context and clinical assessment. A person may show medically consequential restriction at many body sizes, and visible weight alone cannot establish the presence or severity of an illness. See our early warning signs guide and disordered eating overview.

Anorexia nervosa, atypical anorexia nervosa, bulimia nervosa, binge eating disorder, ARFID, OSFED, UFED, pica, and rumination disorder are clinically distinct diagnostic presentations. Some involve intense body-shape or weight concerns, while ARFID can involve sensory sensitivity, low interest in eating, or fear of aversive consequences without the same weight- or shape-based motivation. An online pattern of food avoidance, a screening score, or a diagnosis named in a social media bio cannot determine the clinical category. Formal evaluation requires a thorough history and, when indicated, medical examination. NIMH; NICE NG69.

Do Social Media Effects Differ for Adolescents and Adults?

Children and teenagers

Adolescence combines rapid biological development, peer evaluation, identity formation, and shifts in body composition. During this period, online comparison and reinforcement can become personally salient, but teenagers vary greatly in maturity, vulnerability, supportive relationships, and media skills. The meta-analysis focused on 13–18-year-olds found small pooled eating-related associations, not a deterministic adolescent response. The American Academy of Pediatrics clinical report emphasizes careful recognition and medical assessment when eating-disorder symptoms are suspected in children or adolescents.

For younger teens, the APA advisory favors developmentally appropriate discussion and monitoring, gradual autonomy, reduced exposure to harmful material, and training in social media literacy. Caregivers can pay attention to changes in eating, mood, social withdrawal, and functioning without turning ordinary body concerns or internet use into automatic accusations. A collaborative conversation about what appears in a feed is generally more informative than inferring disease from hours online.

Young adults and older adults

Young adulthood can involve academic transitions, occupational pressure, dating, athletic goals, and changing access to support. The 2026 young-adult systematic review indicates that appearance-focused content and problematic social media engagement are associated with more severe eating-disorder outcomes in several studies, but diagnostic research remains limited. Evidence derived from college or young-adult samples should not be applied uncritically to middle-aged adults, older adults, or people with different clinical and cultural contexts.

Boys, men, and diverse populations

Muscularity, leanness, performance, and fitness-focused ideals may be especially relevant to some boys and men. Gendered marketing and social expectations also affect people of different gender identities. The selected TikTok fitspiration study showed different content themes for videos featuring women and men, but platform content analysis is not a clinical prevalence estimate. Pryde and colleagues, 2024. More research is needed among transgender and nonbinary users, racial and ethnic minorities, people with disabilities, and populations outside the heavily studied adolescent and young-adult samples.

The APA advisory explicitly recognizes limited representation of marginalized populations in the underlying evidence. Lack of adequate subgroup research is a reason for tailored assessment, not evidence that a group is unaffected. When evaluating symptoms, clinicians should consider discrimination, exposure to appearance standards, access to culturally responsive care, and differences in how distress is expressed.

What About TikTok, Instagram, and Other Platforms?

Platform names are useful descriptions of study exposure, but platform-level comparisons are easily overstated. Features differ: image presentation, short-video feeds, public metrics, private messaging, and search-based recommendations are not interchangeable. The 2026 adolescent meta-analysis found stronger relationships for some Instagram-focused studies, yet the included studies varied in outcome measures, sampling, time period, and risk of bias. These differences cannot isolate an intrinsic causal effect of a brand. Fortunato et al., 2026.

TikTok studies provide important examples of rapidly circulated fitness and recovery material, but examining selected hashtags cannot establish what the average user sees. Platforms and recommendation systems change faster than research programs can often evaluate them. General claims about any single algorithm should be treated as hypotheses unless supported by an appropriate audit or experimental design.

Can Diet and Fitness Tracking Apps Have Similar Associations?

Digital food and exercise monitoring overlaps with the wider social-media environment but is a different exposure category. A 2025 systematic review of 38 studies found that diet and fitness app users, particularly frequent users, commonly reported more disordered-eating symptoms than nonusers. Most studies were cross-sectional; they cannot establish whether app use increased symptoms, whether people with concerns were more likely to begin tracking, or whether another factor influenced both.

This literature underscores the importance of individual meaning and context. A digital tool may be used within supervised clinical care for a defined purpose, while repeated self-monitoring may be distressing or compulsive for someone else. Decisions about keeping, changing, or discontinuing a tool during eating-disorder treatment should be informed by the person's clinical situation and treatment team, especially when nutrition or medical safety is at stake.

Can Positive or Recovery-Focused Content Help?

Online communities may offer connection, language for recognizing problems, hope about treatment, or access to lived experiences that feel otherwise isolated. Social support is a plausible benefit, and the APA advisory recognizes meaningful opportunities for connection. However, positive intentions and recovery labels do not guarantee that a post is clinically appropriate, accurate, or helpful for every viewer.

A 2025/2026 qualitative analysis of 312 prominent TikTok recovery videos identified themes involving recovery milestones, social relationships, bodies presented as evidence of recovery, and the perceived power of the disorder. These observations illuminate the culture of recovery content but do not establish treatment effectiveness. A 2026 mixed-methods study found that recovery-labeled and pro-eating-disorder elements could overlap in sampled TikTok material and feeds; its quantitative sample was small and did not provide strong evidence of a protective relationship between recovery-content exposure and symptom severity.

A person in recovery may therefore benefit from choosing online communities carefully, identifying content that supports treatment goals, and discussing distressing exposures with clinicians. Some people find online stories supportive; others find comparison and numerical or visual cues triggering. Online support can complement care, but it cannot replace individualized medical assessment, nutritional rehabilitation when needed, or evidence-based psychotherapy. See Eating Disorder Recovery and Eating Disorder Treatment.

Does Social Media Literacy or Prevention Work?

Media literacy aims to help people recognize persuasive content, constructed appearance ideals, commercial incentives, manipulated images, and unhelpful comparisons. It may also encourage critical reflection on recommendation feeds and healthier choices about engagement. The APA advisory recommends age-appropriate social media literacy training, but the evidence for any single lesson, school workshop, or platform feature preventing clinically diagnosed eating disorders remains incomplete.

Digital prevention is a separate clinical-research category from ordinary social media content. A 2025 randomized controlled trial of the Media Smart Targeted program enrolled 555 participants aged 13–25 and examined an internet-delivered indicated prevention or early-intervention program for disordered eating. It tested a structured intervention with defined outcome measures, not the effect of simply following body-positive accounts or reducing scrolling. Outcomes from this trial should not be interpreted as proof that any generic online wellness campaign prevents an eating disorder.

Prevention efforts are most credible when they specify the population, intervention, comparison condition, measured outcomes, follow-up period, participation, and adverse effects. Prevention of symptom worsening, improvements in body image, greater help-seeking, and prevention of new clinical cases are all meaningful but different outcomes.

Practical Ways to Reduce Harmful Exposure

If particular types of posts consistently heighten food-related distress or body comparison, reducing their visibility can be a reasonable environmental adjustment. Options include unfollowing accounts that repeatedly intensify distress, using reporting or 'not interested' functions, muting triggering topics, seeking supportive and evidence-informed communities, and protecting time for sleep and offline relationships. These strategies are practical adaptations of exposure reduction and media-literacy principles; the research does not establish them as stand-alone treatments for eating disorders.

Notice patterns in the consequences of use rather than treating all use as a problem. Questions include whether a session leads to rigid changes in eating, rising shame, compulsive checking, interference with daily activities, or avoidance of meals and social situations. An individual pattern can guide a conversation with a qualified professional. There is no validated self-monitoring threshold that by itself determines an eating-disorder diagnosis or who must stop using a particular app.

Parents and educators can ask adolescents what kinds of content they encounter, why it feels compelling, and how they evaluate claims. Conversations work better when they make room for genuine enjoyment, humor, friendships, and supportive communities. For children with substantial behavioral or physical changes, digital boundary-setting belongs alongside prompt professional assessment, rather than becoming the only response. AAP clinical report; APA health advisory.

When Social Media Concerns Call for Clinical Assessment

Professional help is appropriate when eating behavior, thoughts about food or body shape, or exercise patterns become rigid, distressing, persistent, or interfere with health, relationships, school, work, or normal daily functioning. Physical warning signs such as marked weakness, fainting, rapid or substantial changes in intake or weight, dehydration, or symptoms associated with purging require medical attention. A person can have serious complications at any body size. The NIMH and AAP emphasize comprehensive assessment rather than deciding based on appearance.

Urgent medical evaluation is warranted for collapse, chest pain, severe dehydration, confusion, fainting, dangerous electrolyte concerns, or other signs of acute instability. Medical risk can be present even when the person does not look underweight. Clinicians evaluate eating and weight history, vital signs, physical symptoms, mental-health risk, and relevant laboratory findings. Society for Adolescent Health and Medicine guidance and NICE NG69 describe why medical monitoring and appropriate levels of care matter.

An online checklist or questionnaire can suggest that further assessment is appropriate, but it does not establish anorexia nervosa, bulimia nervosa, binge eating disorder, ARFID, OSFED, or another diagnosis. The diagnostic process considers duration, specific behaviors, underlying motivations, distress, functional impairment, developmental stage, differential medical explanations, and possible co-occurring conditions. For a clinical pathway, consult How Are Eating Disorders Diagnosed?.

How Clinicians Can Use Social Media Information

Asking about social media can enrich an eating-disorder assessment when done sensitively and without assuming the platform is the cause. Useful clinical questions include which content types are followed, whether the person compares their body or eating to online examples, how digital exposure affects behaviors, and whether supportive online connections exist. The answers contribute to a broader formulation that includes eating patterns, personal history, family and social context, medical status, and comorbid symptoms.

Clinical treatment targets the diagnosed disorder and the individual's health needs. Social media management may become one component of a treatment plan, alongside therapies appropriate to the diagnosis and age group, nutritional care, family involvement where indicated, and medical monitoring. Recommendations supported for adolescent anorexia nervosa cannot simply be carried over to ARFID, adult binge eating disorder, or another presentation. The NICE eating-disorders guideline provides diagnosis-sensitive treatment recommendations.

Research Limitations and What Scientists Still Need to Learn

First, a large share of available research uses cross-sectional surveys and self-reported social media use. Selection effects and reciprocal pathways are plausible: people experiencing eating-related distress may seek out more appearance or food content, and such content may then reinforce the distress. Even longitudinal adjustment cannot eliminate every potential confounder. Randomized experiments improve causal inference for immediate effects but rarely follow participants long enough to test incidence of clinical disorders.

Second, samples and constructs vary. Studies frequently recruit adolescents, college students, or girls and young women, and report body dissatisfaction or eating-attitude scores as outcomes. Much less evidence addresses long-term course of formally diagnosed disorders, relapse, medically complex patients, men and gender-diverse participants, and low-resource settings. The narrowly focused Fomínová et al. review illustrates how few studies remain when the question specifically concerns diagnostic eating disorders.

Third, different instruments measure different exposure and outcome constructs. An overall correlation may combine time spent online, platform-specific behaviors, and symptom questionnaires with different validity properties. Reviews therefore provide useful population-level patterns, while they cannot yield a personally calibrated risk probability from someone's usage time or feed composition. Heterogeneity, incomplete access to platform data, rapid product changes, and publication bias complicate comparison across studies.

Fourth, the scientific question is broader than whether an app is 'good' or 'bad.' Researchers need controlled studies of exposure types, longitudinal clinical outcomes, harmful and protective processes, developmental differences, recommender systems, social support, and interventions that produce meaningful behavioral and medical benefits. These studies should make adverse effects and subgroup differences visible. These priorities follow the gaps identified in recent evidence syntheses. Fortunato et al., 2026; Fomínová et al., 2026.

Frequently Asked Questions

Does social media cause anorexia nervosa or bulimia nervosa?

Social media can expose users to pressures and behavior models that may contribute to vulnerability or symptom maintenance, but research does not establish social media alone as a sufficient cause of anorexia nervosa or bulimia nervosa. These are distinct multifactorial clinical conditions requiring assessment. The strongest immediate causal evidence is for narrower outcomes such as changes in body image following specific imagery exposure. De Valle et al.; NIMH.

Is TikTok worse than Instagram for eating-disorder risk?

There is no robust clinical head-to-head evidence that establishes a universal ranking. Studies differ in their participants, definitions, historical periods, and outcomes. Platform-specific features and a person's particular feed may matter more than a brand label. Fortunato et al., 2026.

Is seeing fitness content the same as having an eating disorder?

No. Fitness content is an exposure category, and an eating disorder is a clinical condition assessed through behavior, cognition, health, and functioning. Some fitness-oriented material contains harmful misinformation or appearance pressure, while other material is benign or useful. The viewer's experience and any clinically meaningful symptoms require separate evaluation. Pryde et al., 2024; NIMH.

Can social media affect binge eating?

Research links some forms of social media engagement with disordered-eating symptoms, but studies often combine restrictive and binge-related behaviors in composite measures. The evidence does not establish that a particular social-media pattern causes binge eating disorder, which has its own diagnostic criteria and treatment literature. Fomínová et al., 2026; NICE.

Should someone recovering from an eating disorder stop using social media?

The most appropriate decision depends on the person's triggers, supportive relationships, treatment goals, and medical and psychological needs. Reducing harmful exposures can be helpful as one environmental measure; total abstinence is not a universally validated treatment. Recovery-focused content may also contain distressing comparison cues. Recovery-content study, 2026; NICE.

Are body-positive and recovery accounts protective?

They may offer community and reassurance, but following an account does not amount to a tested eating-disorder treatment. Recovery-labeled posts vary in quality and can sometimes mix supportive messages with problematic imagery or comparisons. Evidence for protection against clinical disorders is preliminary and context-dependent. Recovery-content analyses; 2026 TikTok recovery study.

Can an eating disorder occur at a normal or higher body weight?

Yes. Serious disordered-eating behaviors and medical complications can occur across body sizes. Weight and BMI cannot be used alone to exclude an eating disorder or decide its medical severity. Clinical assessment considers eating and weight history, vital signs, symptoms, and other findings. NIMH; SAHM.

Does a high eating-disorder screening score confirm a diagnosis?

No. Screening instruments identify possible concern and can support a referral for assessment. Diagnosis requires professional evaluation of the clinical criteria, associated impairment, medical risk, and differential explanations. AAP; NICE.

What is the most useful first step if online content is changing my eating?

Take the changes seriously, reduce exposure to material that repeatedly worsens them, and speak with a qualified clinician if food-related behavior, distress, or physical health has changed. Prompt clinical evaluation is especially important when eating is substantially restricted, compensatory behaviors occur, or medical symptoms develop. Digital changes support safety but do not replace appropriate care. NIMH.

Related Articles

What Causes Eating Disorders? Genetics, Psychology, and Risk Factors — the parent article on multifactorial causes and vulnerabilities.

Eating Disorders: Types, Symptoms, Causes, and Treatment — a clinical overview of the diagnostic spectrum.

Disordered Eating: Signs, Examples, and When to Seek Help — the distinction between patterns of concern and formal diagnoses.

Early Signs of an Eating Disorder: Behavioral and Physical Changes — early recognition across body sizes.

How Are Eating Disorders Diagnosed? Assessment and Criteria — professional evaluation and screening limitations.

Eating Disorder Treatment: Evidence-Based Therapies and Care — diagnosis- and age-sensitive clinical treatments.

Social Media and Body Image: Appearance Comparison, Ideals, and Mental Health — the distinct body-image search intent.

Social Comparison on Social Media: Mental Health, Body Image, and Self-Evaluation — the psychological comparison pathway.

Social Media and Mental Health: Risks, Benefits, and What Research Shows — broader platform effects beyond eating disorders.

References

American Academy of Pediatrics (Hornberger, L. L., Lane, M. A., & Committee on Adolescence). (2021). Identification and management of eating disorders in children and adolescents. Pediatrics, 147(1), e2020040279. doi:10.1542/peds.2020-040279.

American Psychological Association. (2023). Health advisory on social media use in adolescence. Official health advisory.

Anderberg, I., Kemps, E., & Prichard, I. (2025). The link between the use of diet and fitness monitoring apps, body image and disordered eating symptomology: A systematic review. Body Image, 52, 101836. doi:10.1016/j.bodyim.2024.101836.

Bloxham, I., Taumoepeau, M., & Treharne, G. J. (2025). 'Actually, let's not give sick people any more ideas': Eating disorder recovery content on TikTok. Eating Disorders. doi:10.1080/10640266.2025.2589521.

Bonfanti, R. C., et al. (2025). The association between social comparison in social media, body image concerns and eating disorder symptoms: A systematic review and meta-analysis. Body Image, 52, 101841. doi:10.1016/j.bodyim.2024.101841.

De Valle, M. K., Gallego-García, M., Williamson, P., & Wade, T. D. (2021). Social media, body image, and the question of causation: Meta-analyses of experimental and longitudinal evidence. PubMed PMID 34695681.

Fomínová, V., Sebalo Vňuková, M., Anders, M., Kitzlerová, E., & Trávníčková, E. (2026). The association between social media use and eating disorders in young adults: A systematic review. Journal of Eating Disorders, 14, 170. doi:10.1186/s40337-026-01653-3.

Fortunato, L., Gullo, S., Muscolino, A., La Tona, A., Rodgers, R. F., & Lo Coco, G. (2026). Relation between social media use and body image concerns and disordered eating in adolescence: A systematic review and meta-analysis. Adolescent Research Review, 11, 433–453. doi:10.1007/s40894-026-00285-7.

Griffiths, S., Martyn, I., Kannan, H., Stone, B., Austen, E., Tescari, E., Nattala, U., Grey, W., Brittain, N., Mitchison, D., Linardon, J., Fardouly, J., & Harris, E. (2026). Blurring the boundaries: An investigation of eating disorder recovery content on TikTok. International Journal of Eating Disorders. doi:10.1002/eat.70153.

Gruest, V., & Walter, N. (2026). Lifting the screen on fitspiration: A meta-analysis. Health Communication. doi:10.1080/10410236.2026.2653191.

National Institute for Health and Care Excellence. (2017; updated). Eating disorders: Recognition and treatment (NG69). Official guideline.

National Institute of Mental Health. (n.d.). Eating disorders: What you need to know. Official clinical information.

Pryde, S., Kemps, E., & Prichard, I. (2024). A content analysis of fitspiration videos on TikTok. Body Image, 51, 101769. doi:10.1016/j.bodyim.2024.101769.

Society for Adolescent Health and Medicine. (2022). Medical management of restrictive eating disorders in adolescents and young adults. Journal of Adolescent Health. doi:10.1016/j.jadohealth.2022.08.006.

Wilksch, S. M., O'Shea, A., Crosby, R., Lawson, R., & Wade, T. D. (2026). Pragmatic indicated prevention and early intervention for disordered eating: A randomized controlled trial of Media Smart Targeted internet program in youth. Journal of the American Academy of Child & Adolescent Psychiatry, 65(3), 392–407. doi:10.1016/j.jaac.2025.02.017.