Skip to content

Anorexia nervosa — overview

TL;DR — Anorexia nervosa (AN) is a restrictive eating disorder defined by inadequate energy intake and low weight, fear of weight gain or persistent weight-preventing behaviour, and disturbance in weight/shape experience. It is neither simply a choice nor only a weight state: genetic, metabolic, developmental, cognitive and social processes contribute, while starvation itself amplifies many abnormalities. Mortality is the organizing clinical fact: a 36-study meta-analysis estimated 5.1 deaths per 1,000 person-years and an SMR of 5.86, with one in five deaths by suicide (Arcelus 2011, PMID 21727255); two 2026 syntheses reproduce that magnitude on far larger samples — a pooled all-cause mortality risk ratio of 5.52 (95% CI 4.47–6.82) and a suicide-specific RR of 9.86 (95% CI 5.63–17.27) for AN (Semchishen 2026, PMID 41536100), and a pooled SMR of 5.06 (95% CI 3.47–7.38) across 30 studies and 33,176 patients (Lai 2026, PMID 41277145). Adolescent family-based approaches have the clearest psychotherapy evidence, whereas adult trials show improvement across therapies without a consistently superior modality (Lock 2010, PMID 20921118; Solmi 2021, PMID 33600749). Weight restoration is necessary but is not a complete definition of recovery.

What the diagnosis contains

Domain Operational meaning Important qualification
Energy restriction Intake insufficient for physiological needs Restriction may be overt or concealed; appetite loss is not required
Weight criterion Significantly low weight in AN Atypical AN can carry comparable psychopathology and medical instability without low BMI (Walsh 2023, PMID 36508318)
Fear/behaviour Fear of gain or persistent interference with gain Behaviour can demonstrate the criterion when fear is not verbalized
Self-evaluation Disturbed experience of weight/shape, undue influence, or failure to recognize seriousness Ego-syntonicity complicates engagement and capacity assessment
Subtype Restricting; binge-eating/purging Subtype is descriptive and can change over time

Scale and burden

Epidemiological estimates vary materially with age, sampling frame and diagnostic system. A worldwide review of DSM-5 eating disorders in young people reported lifetime AN in 0.8–6.3% of women and 0.1–0.3% of men, within a total DSM-5 eating-disorder burden of 5.5–17.9% of young women and 0.6–2.4% of young men by early adulthood (Silén 2022, PMID 36125216). The US nationally representative NESARC-III survey of 36,306 adults estimated lifetime DSM-5 AN at 0.80% (SE 0.07%) and 12-month AN at 0.05% (SE 0.02%), showing that community cases extend well beyond specialist services (Udo 2018, PMID 29859631). Global burden estimates should be read as modelled estimates constrained by sparse primary data in many regions; GBD 2019 could compute years of life lost for only two mental disorders — anorexia nervosa and bulimia nervosa — and its authors state plainly that the resulting eating-disorder figure of 17,361.5 YLLs is "extremely low" and does not reflect actual premature mortality (GBD 2019 Mental Disorders Collaborators 2022, PMID 35026139).

The long course is heterogeneous. Across 415 eating-disorder studies (88,372 participants), pooled recovery increased with follow-up, from 42% before two years to 67% at ten years or longer; these are transdiagnostic estimates and not a prognosis for an individual with AN (Solmi 2024, PMID 38214616). Health-related quality of life is substantially impaired across eating disorders: a seven-study SF-36 meta-analysis found significantly lower scores than population norms in every diagnostic group, and — importantly — could not establish differences between AN, BN, EDNOS and BED (Winkler 2014, PMID 24857566). The AN evidence within it rests on five studies and 227 patients, so the impairment finding is robust while the between-diagnosis comparison is simply underpowered.

A metabo-psychiatric disorder

The largest landmark GWAS in the seed evidence studied 16,992 cases and 55,525 controls, identified eight loci, and found genetic correlations with psychiatric traits, physical activity, glycaemic, lipid and anthropometric traits that were not explained only by common BMI-associated variants (Watson 2019, PMID 31308545). Twin-based heritability estimates summarized in that report were 50–60%. “Metabo-psychiatric” is therefore a research model, not evidence that a metabolic test diagnoses AN or that a metabolic drug has been shown to treat it.

Neurobiological studies implicate reward, threat, interoception and cognitive-control systems, but state effects are pervasive: starvation, dehydration, endocrine adaptation and treatment exposure can each change brain and cognitive measures (Bulik 2022, PMID 35524137; Frank 2019, PMID 31443880). Cross-sectional differences cannot by themselves distinguish cause, scar and adaptive response.

Treatment map

Clinical task Best-supported conclusion Evidence boundary
Medical stabilization Correct acute physiological compromise and begin monitored nutrition Setting and rate depend on instability, not BMI alone (Society for Adolescent Health and Medicine 2022, PMID 36058805)
Nutritional rehabilitation Higher-calorie inpatient refeeding can restore stability sooner in monitored adolescents/young adults Does not license unsupervised rapid refeeding or extrapolation to extreme malnutrition (Garber 2021, PMID 33074282)
Adolescents Eating-disorder-focused family therapy/FBT has repeated randomized support Access, family context and illness duration affect feasibility (Fisher 2019, PMID 31041816)
Adults CBT-E, MANTRA, SSCM and focal psychodynamic therapy are reasonable specialist options Network evidence does not identify a robust winner (Solmi 2021, PMID 33600749)
Medication No medication is established as sole treatment Olanzapine gave a modest weight signal (BMI gain 0.259 vs 0.095 kg/m² per month) with no obsessionality benefit in a 152-patient RCT (Attia 2019, PMID 30654643)
Neuromodulation/psychedelics Experimental Largest sham-controlled rTMS trial is a 34-person feasibility study; psilocybin evidence is a 10-person open-label phase 1 (McClelland 2018, PMID 30012789; Peck 2023, PMID 37488291; Gallop 2022, PMID 35179712)

Economic burden

Burden is usually stated in mortality and prevalence; the money is less often quoted and is more concentrated outside the health system. A systematic review of 26 studies (11 formal cost-of-illness studies) published between August 2013 and June 2024, with all figures converted to 2024 USD purchasing-power parity, estimated the nationwide annual financial cost of eating disorders at PPP-USD 70.5 billion, with indirect costs contributing 70–93% of that total, and estimated intangible costs (burden of disease) at PPP-USD 355.6 billion. About half the cost-of-illness studies met 60% or more of a standard quality checklist, and the number of such studies has more than doubled in a decade (Ahmed 2025, PMID 39542867; PROSPERO CRD42022358136). The dominance of indirect cost is the substantive point: most of the economic burden is lost productivity and unpaid care, so health-service spending is a small fraction of the total and is not where cost reduction would mainly be found.

At the individual level, German statutory-insurance claims for 1,242 women and 71 men with incident AN and 1,104 women and 64 men with incident bulimia nervosa, followed over five years from two years before the index diagnosis, showed median total costs of roughly EUR 14,000–20,000 across groups, with about two thirds of AN costs attributable to mental illness (versus about half in bulimia nervosa). AN was distinguished by a steeper cost increase shortly after onset and higher inpatient treatment costs; men incurred lower outpatient costs than women (Bothe 2022, PMID 34599621). Intensive treatment costs have been quantified directly in a US hospital-based inpatient/partial-hospitalization programme over 314 consecutive adult first admissions: average charge of $2,295 per inpatient day and $1,567 per partial-hospitalization day, yielding $4,089 and $7,050 per pound of weight gained respectively, with 70% of patients reaching a discharge BMI ≥19; the authors' comparison against national insurance-claims data for residential treatment found residential care associated with higher cost per pound and both higher cost and lower likelihood of weight restoration for most patients (Guarda 2017, PMID 28130794).

The condition in one contemporary review

A 2025 JAMA review of eating disorders provides an external summary against which this knowledge base can be checked, and its numbers align with the pages here. It puts lifetime prevalence of eating disorders at 2–5% worldwide; lifetime depression at 49.5% in AN (against 76.3% in bulimia nervosa and 65.5% in binge-eating disorder); AN mortality at 5.1 deaths per 1,000 person-years (95% CI 4.0–6.1), nearly six times that of same-age individuals, with 25% of deaths from suicide; and family-based treatment remission at 6–12 months of 48.6% versus 34.3% for individual treatment (OR 2.08, 95% CI 1.07–4.03, p = .03). It states plainly that there are currently no effective medications for anorexia nervosa, while fluoxetine and lisdexamfetamine have measurable effects in bulimia nervosa and binge-eating disorder respectively (Attia 2025, PMID 40048192).

Two of those figures deserve to be read against the detail elsewhere in this wiki. The 25% suicide-death proportion sits at the top of the range reported by the AN-specific meta-analyses (20–21%), and the depression figure — roughly half of people with AN meeting lifetime criteria — is the reason pharmacotherapy records a depression-moderated fluoxetine subgroup finding as the one live antidepressant question in this condition.

Why weight is necessary but insufficient

Weight restoration reverses many starvation-mediated abnormalities and is a core objective of acute treatment. Yet the same BMI can conceal different trajectories, body compositions, rates of loss, premorbid percentiles and physiological risks. Atypical AN demonstrates the failure of a low-weight gate: its eating-disorder psychopathology is as high as or higher than AN in comparative studies, and many physiological complications occur, although some are less frequent (Walsh 2023, PMID 36508318). An invited 2026 update of that review, now spanning 64 publications, reaches the same conclusion and sharpens it — atypical AN shows greater eating-disorder psychopathology, comparable non-eating-disorder psychopathology, and the same complications (notably menstrual disturbance and reduced bone mineral density) at lower frequency — while still finding the longitudinal course evidence missing (Lee 2026, PMID 42557659).

Research also uses inconsistent recovery definitions. Some require weight only; others combine weight, menstruation, eating-disorder psychopathology and psychosocial function. Consequently, apparent differences between studies can be definitional rather than therapeutic (Solmi 2024, PMID 38214616).

Safety frame

Immediate risk can arise from bradycardia, hypotension, hypothermia, electrolyte disturbance, dehydration, hypoglycaemia, organ dysfunction, rapid weight loss, purging, suicide risk and refeeding shifts. Clinical severity must not be inferred from appearance or BMI alone. Medical complications span cardiovascular, renal/electrolyte, endocrine, skeletal and gastrointestinal systems (Westmoreland 2016, PMID 26169883; Puckett 2023, PMID 36803805).

This knowledge base is research synthesis, not individualized medical advice. Thresholds are summarized in red flags and safety concerns, with jurisdiction-specific guidance in guidelines.

Open questions

  • Can genetically implicated metabolic biology yield a causal, targetable mechanism rather than a correlational label (Watson 2019, PMID 31308545)?
  • Which outcome set best captures durable recovery beyond weight, and can it be harmonized across trials (Solmi 2024, PMID 38214616)?
  • Which components of adolescent family treatment are transportable to adults, and for whom (Lock 2010, PMID 20921118; Solmi 2021, PMID 33600749)?
  • How should treatment systems incorporate atypical AN without replacing one rigid weight rule with another (Walsh 2023, PMID 36508318)?
  • Where should resources go if 70–93% of the economic burden of eating disorders is indirect rather than health-service cost (Ahmed 2025, PMID 39542867)?
  • Does the cost-per-pound advantage of hospital-based over residential care survive comparison on relapse and function rather than discharge weight (Guarda 2017, PMID 28130794)?

References

  1. Arcelus J, et al. Mortality rates in patients with anorexia nervosa and other eating disorders: a meta-analysis of 36 studies. Arch Gen Psychiatry. 2011;68:724-731. PMID 21727255.
  2. Solmi M, et al. Outcomes in people with eating disorders: a systematic review and meta-analysis. World Psychiatry. 2024. PMID 38214616.
  3. Watson HJ, et al. Genome-wide association study identifies eight risk loci and implicates metabo-psychiatric origins for anorexia nervosa. Nat Genet. 2019. PMID 31308545.
  4. Lock J, et al. Randomized clinical trial comparing family-based treatment with adolescent-focused individual therapy. Arch Gen Psychiatry. 2010. PMID 20921118.
  5. Solmi M, et al. Comparative efficacy and acceptability of psychological interventions for adult outpatients with anorexia nervosa. Lancet Psychiatry. 2021. PMID 33600749.
  6. Walsh BT, et al. A systematic review comparing atypical anorexia nervosa and anorexia nervosa. Int J Eat Disord. 2023. PMID 36508318.
  7. Bulik CM, et al. Genetics and neurobiology of eating disorders. Nat Neurosci. 2022. PMID 35524137.
  8. Frank GKW. The neurobiology of eating disorders. Child Adolesc Psychiatr Clin N Am. 2019. PMID 31443880.
  9. Garber AK, et al. Short-term outcomes of the Study of Refeeding to Optimize Inpatient Gains. JAMA Pediatr. 2021. PMID 33074282.
  10. Attia E, et al. Olanzapine versus placebo in adult outpatients with anorexia nervosa. Am J Psychiatry. 2019. PMID 30654643.
  11. Society for Adolescent Health and Medicine. Medical management of restrictive eating disorders in adolescents and young adults. J Adolesc Health. 2022. PMID 36058805.
  12. Westmoreland P, et al. Medical complications of anorexia nervosa and bulimia. Am J Med. 2016. PMID 26169883.
  13. Puckett L, et al. Renal and electrolyte complications in eating disorders. J Eat Disord. 2023. PMID 36803805.
  14. Winkler LA, et al. Quality of life in eating disorders: a meta-analysis. Psychiatry Res. 2014. PMID 24857566.
  15. Semchishen SN, et al. All-cause and cause-specific mortality risk in individuals with eating disorders: systematic review and meta-analysis. World Psychiatry. 2026;25:125-140. PMID 41536100.
  16. Lai ET, et al. Systematic review and meta-analysis of mortality in patients with anorexia nervosa. Int J Eat Disord. 2026;59:424-449. PMID 41277145.
  17. Lee V, Hagan KE. An invited updated systematic review and meta-analysis comparing atypical anorexia nervosa and anorexia nervosa. Int J Eat Disord. 2026. PMID 42557659.
  18. McClelland J, et al. Randomised controlled feasibility trial of real versus sham rTMS in adults with severe and enduring anorexia nervosa: the TIARA study. BMJ Open. 2018. PMID 30012789.
  19. Peck SK, et al. Psilocybin therapy for females with anorexia nervosa: a phase 1, open-label feasibility study. Nat Med. 2023;29:1947-1953. PMID 37488291.
  20. Fisher CA, et al. Family therapy approaches for anorexia nervosa. Cochrane Database Syst Rev. 2019. PMID 31041816.
  21. Silén Y, Keski-Rahkonen A. Worldwide prevalence of DSM-5 eating disorders among young people. Curr Opin Psychiatry. 2022. PMID 36125216.
  22. Udo T, Grilo CM. Prevalence and correlates of DSM-5-defined eating disorders in US adults. Biol Psychiatry. 2018. PMID 29859631.
  23. GBD 2019 Mental Disorders Collaborators. Global burden of 12 mental disorders, 1990–2019. Lancet Psychiatry. 2022. PMID 35026139.
  24. Gallop L, et al. Neuromodulation and eating disorders. Curr Psychiatry Rep. 2022. PMID 35179712.
  25. Ahmed M, et al. Global and regional economic burden of eating disorders: a systematic review and critique of methods. Int J Eat Disord. 2025;58:91-116. PMID 39542867.
  26. Bothe T, et al. Gender-related differences in health-care and economic costs for eating disorders. Int J Eat Disord. 2022;55:61-75. PMID 34599621.
  27. Guarda AS, et al. Intensive treatment for adults with anorexia nervosa: the cost of weight restoration. Int J Eat Disord. 2017;50:302-306. PMID 28130794.
  28. Attia E, Walsh BT. Eating disorders: a review. JAMA. 2025;333:1242-1252. PMID 40048192.