← Research index  ·  Disorder page  ·  Source YAML

Cross-provider research synthesis

Anorexia Nervosa

MONDO:0005351 Curated 2026-07-04T00:00:00Z 9 harmonized findings
falcon · 39 citations openscientist · 71 citations
Concordant asserts the finding Partial supports a weaker/qualified form Contradictory conflicts with it Silent does not address it

Harmonized findings

Anorexia nervosa is defined by persistent restriction of energy intake leading to low body weight, an intense fear of gaining weight, and a disturbance in the perception of one's own body weight or shape.

UNANIMOUS INTEGRATED
phenotypediagnosis
ProviderStanceScoreEvidence
falcon CONCORDANT 95% an eating disorder characterized by the restriction of energy intake, fear of gaining weight despite low weight, and disturbances in the perception of body weight or shape and the influence of these factors on self-worth
Falcon quotes the biomarker meta-analysis definition, which matches the three DSM/ICD-11 core features (restriction, fear of weight gain, body image disturbance).
DOI:10.3390/nu16132095
openscientist CONCORDANT 95% characterized by persistent energy intake restriction, intense fear of gaining weight, and disturbance in self-perceived body weight or shape
OpenScientist states the same triad in its summary, framing AN as a severe metabo-psychiatric disorder.

Anorexia nervosa is a heritable, non-monogenic (polygenic) metabo-psychiatric disorder in which many small-effect loci and substantial genetic overlap with other psychiatric traits confer risk (twin heritability on the order of ~50-74%).

UNANIMOUS INTEGRATED
genetic_factorpathophysiology
ProviderStanceScoreEvidence
falcon CONCORDANT 85% AN is not established as a monogenic disorder in the retrieved evidence
Falcon rejects a monogenic model and elsewhere reports twin-based heritability ~50-60% with shared architecture across psychiatric traits.
DOI:10.1038/s41398-023-02585-1
openscientist CONCORDANT 90% established AN as a disorder with both psychiatric and metabolic genetic origins, with heritability estimated at 48–74% from twin studies and SNP-based heritability of approximately 11–17%
OpenScientist gives the same polygenic, metabo-psychiatric framing with explicit twin (48-74%) and SNP-based (11-17%) heritability estimates.
PMID:31308545, PMID:39988782

Genome-wide association studies have identified genome-wide-significant AN risk loci implicating both psychiatric and metabolic pathways.

MAJORITY LEAD
genetic_factor
ProviderStanceScoreEvidence
openscientist CONCORDANT 85% The 2019 PGC-ED GWAS meta-analysis of approximately 17,000 AN cases and 55,000 controls identified
OpenScientist anchors on the landmark 2019 PGC-ED GWAS, which identified eight genome-wide-significant loci across psychiatric and metabolic pathways.
PMID:31308545
falcon PARTIAL 50% Conditional FDR analysis identified 58 novel AN loci
Falcon instead anchors on a 2023 conditional-FDR cross-trait analysis that leverages overlap with other psychiatric disorders to call 58 loci — a larger, more liberal locus set than the stringent 8 genome-wide-significant loci OpenScientist reports. Compatible but a different primary analysis, so only partial concordance.
DOI:10.1038/s41398-023-02585-1

Anorexia nervosa carries the highest mortality of any psychiatric disorder, with an all-cause standardized mortality ratio around 5.

UNANIMOUS INTEGRATED
prognosis
ProviderStanceScoreEvidence
openscientist CONCORDANT 95% It carries the highest mortality rate of any mental illness, with a standardized mortality ratio (SMR) of 5.52 (95% CI: 4.47–6.82) compared with the general population
OpenScientist reports SMR 5.52 and a suicide-related mortality RR of 9.86, the highest of any mental illness.
PMID:41536100
falcon CONCORDANT 85% AN described as having the highest mortality rate among psychiatric disorders
Falcon agrees AN has the highest psychiatric mortality; it reports a slightly lower SMR of 5.21 (95% CI 4.10-6.62) from a clinical update — a minor quantitative difference, not a conflict.
DOI:10.1007/s00115-025-01820-y, DOI:10.3390/nu16132095

Chronic energy restriction drives a starvation-adaptation endocrine response — suppressed leptin, elevated ghrelin, growth-hormone resistance (high GH, low IGF-1), and hypothalamic-pituitary dysfunction.

UNANIMOUS INTEGRATED
pathophysiology
ProviderStanceScoreEvidence
falcon CONCORDANT 90% A 2024 meta-analysis of peripheral biomarkers reported higher ghrelin forms and lower leptin
Falcon reports the same leptin-down/ghrelin-up pattern and discusses ghrelin resistance and GH resistance (high GH, low IGF-1) as starvation-adaptation.
DOI:10.3390/nu16132095
openscientist CONCORDANT 95% Serum levels of leptin, an anorexigenic adipokine, are suppressed and levels of ghrelin, an orexigenic gut peptide, are elevated in women with anorexia nervosa; however, levels of peptide YY, an anorexigenic gut peptide, are paradoxically elevated
OpenScientist quotes the Misra/Klibanski endocrine review, adding hypogonadotropic hypogonadism, hypercortisolaemia, and the paradoxical peptide-YY elevation.
PMID:27811940

Anorexia nervosa is associated with the largest brain structural (gray-matter and cortical-thickness) deficits of any psychiatric disorder, only partially reversible with weight restoration.

MAJORITY LEAD
pathophysiologyphenotype
ProviderStanceScoreEvidence
openscientist CONCORDANT 90% significant global brain volume reductions in gray matter (GM), white matter (WM), and increases in cerebrospinal fluid (CSF) in acute AN
OpenScientist reports the ENIGMA structural findings (Cohen's d up to 0.95) and a meta-analysis of global GM/WM volume loss that only partially normalizes after 1.5 years of recovery.
PMID:36031441, PMID:41619402
falcon PARTIAL 40% Resting-state fMRI synthesis reports altered large-scale networks in eating disorders; in AN, findings include altered connectivity patterns involving DMN, salience and executive networks and regions tied to social cognition and sensory/aesthetic processing
Falcon documents altered resting-state functional connectivity but does not report the structural gray-matter/cortical-thickness volume deficits or the ENIGMA magnitude — functional-network overlap without the structural claim.
DOI:10.1186/s12880-024-01432-z

Gut microbiome dysbiosis together with immune/cytokine dysregulation contributes to AN pathophysiology via the gut-brain axis.

UNANIMOUS INTEGRATED
pathophysiology
ProviderStanceScoreEvidence
openscientist CONCORDANT 85% AN patients show reduced alpha diversity and lower short-chain fatty acid (SCFA) levels. Dysbiosis affects immune system responses, intestinal permeability, and neurotransmitter production via the gut-brain axis
OpenScientist links dysbiosis (reduced alpha diversity, lower SCFA) to immune and gut-brain-axis effects, and notes decreased zonulin/LBP challenging the leaky-gut model in adolescents.
PMID:33416044, PMID:40789230
falcon CONCORDANT 75% In 63 female adolescents with AN vs 41 controls, IL-1β and IL-6 were lower at admission; IL-1β normalized after weight recovery; IL-15 was elevated at all time points; TNF-α did not differ.
Falcon reports adolescent cytokine dysregulation (differing from adult pro-inflammatory assumptions) plus Mendelian-randomization evidence for causal gut taxa, converging on immune/microbiome involvement.
DOI:10.3390/nu16111596

First-line treatment for adolescent AN is family-based treatment (FBT), with cognitive behavioral therapy for adults, but remission remains incomplete.

UNANIMOUS INTEGRATED
treatment
ProviderStanceScoreEvidence
openscientist CONCORDANT 90% Treatment relies primarily on family-based therapy (FBT) for adolescents, with remission rates of 29–49% at 12-month follow-up, and cognitive behavioral therapy for adults
OpenScientist names FBT as the leading adolescent treatment (remission 29-49% at 12 months) and CBT-E for adults, with no FDA-approved pharmacotherapy and olanzapine as a promising adjunct.
PMID:31466116
falcon CONCORDANT 85% Recovery rates with standard FBT vary by definition: 50–70% when recovery is defined as maintaining >85% body weight, but 28–50% when both weight restoration and reduction in eating-disorder symptoms are required at end of treatment
Falcon likewise centers FBT (recovery 28-70% depending on definition) and cites olanzapine as having only moderate evidence for weight gain, agreeing that remission is incomplete.
DOI:10.3390/psychiatryint5020015

Anorexia nervosa produces multi-system medical complications, prominently including cardiovascular disease (bradycardia/arrhythmias and markedly elevated risk of major adverse cardiovascular events).

MAJORITY INTEGRATED
phenotypecomorbidity
ProviderStanceScoreEvidence
falcon CONCORDANT 90% In 2,081 patients with AN vs 20,810 controls, 4.8% vs 0.8% had major adverse cardiovascular events (MACE) and 6.0% vs 2.3% had any cardiovascular condition
Falcon quantifies cardiovascular risk from a matched cohort (Tseng 2024): elevated MACE and any-cardiovascular-condition incidence with adjusted hazard ratios.
DOI:10.1001/jamanetworkopen.2024.51094
openscientist PARTIAL 55% Medical complications emerge (amenorrhea, bradycardia, electrolyte disturbances)
OpenScientist lists cardiac complications qualitatively (bradycardia, arrhythmias, QTc prolongation) and quantifies renal/liver risk, but does not report the quantitative MACE cohort risk Falcon provides — overlapping but less specific on cardiovascular events.
PMID:41282513

Narrative

Overview

Both providers characterize anorexia nervosa as a severe, heritable metabo-psychiatric eating disorder defined by energy-intake restriction, fear of weight gain, and body-image disturbance, with adolescent onset, the highest mortality of any psychiatric illness, pervasive starvation-driven endocrine adaptation, and family-based treatment as the mainstay. Falcon (Edison Scientific Literature) is a focused 2023-2024 evidence sweep; OpenScientist is a broader synthesis of 89 publications spanning genetics, neurobiology, outcomes, and therapeutics.

Agreement

The reports converge strongly on the core clinical definition, polygenic non-monogenic (metabo-psychiatric) genetic architecture with ~50-74% twin heritability, the highest-of-any-psychiatric-disorder mortality (SMR ~5.2-5.5), the starvation-adaptation endocrine signature (suppressed leptin, elevated ghrelin, GH resistance with low IGF-1, hypothalamic-pituitary dysfunction), gut-microbiome and immune/cytokine dysregulation via the gut-brain axis, cardiovascular and multi-organ complications, and FBT (adolescents) / CBT (adults) as first-line care with incomplete remission and no disorder-specific approved pharmacotherapy.

Divergence

The main divergences are coverage and quantitative anchoring rather than conflict. On GWAS genetics, OpenScientist anchors on the stringent 2019 PGC-ED GWAS (eight genome-wide-significant loci), whereas Falcon anchors on a 2023 conditional-FDR cross-trait analysis calling 58 loci. On the brain, OpenScientist adds the ENIGMA structural finding (largest cortical/gray-matter deficits of any psychiatric disorder, Cohen's d up to 0.95, only partially reversible), while Falcon reports only altered resting-state functional connectivity. OpenScientist provides deeper genetic (candidate genes FOXP1, CADM1, BDNF, HTR2A), neuroimaging, activity-based-anorexia model, and prognostic detail; Falcon provides sharper quantitative cardiovascular-cohort risk (MACE) and adolescent cytokine data. Minor numeric differences (SMR 5.21 vs 5.52; heritability 50-60% vs 48-74%) are compatible, not contradictory.

Integration

Concepts promoted to kb/disorders/Anorexia_Nervosa.yaml align with the existing pathophysiology anchors (polygenic metabo-psychiatric liability, appetite/restrictive-eating circuit dysregulation, starvation-adaptation endocrine response, immune and microbiome-associated dysregulation, cardiovascular complication risk) and phenotype anchors (abnormal eating behavior, low body weight, bradycardia, amenorrhea): the core definition, polygenic/heritable liability, endocrine biomarker signature, microbiome/immune dysregulation, cardiovascular complications, highest-mortality prognosis, and FBT-first-line treatment.

Not integrated (leads)

Retained as research leads rather than promoted: the specific competing GWAS locus counts (8 vs 58) pending reconciliation of the underlying analyses, and the ENIGMA brain structural-deficit magnitude, which is not yet an existing pathophysiology/phenotype node. Candidate-gene detail, activity-based-anorexia model findings, epigenetic (EWAS) leads, and investigational agents (GLP-1 receptor agonists, bone-targeted teriparatide/romosozumab, neuromodulation trials) remain leads for future curation.

Cross-provider synthesis comparing the falcon (focused 2023-2024 sweep, 39 citations) and openscientist (comprehensive, 71 citations) reports. No direct contradictions were found; divergence is coverage, recency, and quantitative anchoring (GWAS locus counts; brain structural vs functional emphasis; SMR and heritability ranges). best_matching_text values are verbatim excerpts from the cited report files; the legacy prose roll-up (Anorexia_Nervosa-research-synthesis.md) informed the narrative only.