Published by Unseen Progress, an independent publisher of caregiver research. Last reviewed 2026-05-10. Part of the adolescent eating disorder research overview.
Short answer. ARFID (Avoidant/Restrictive Food Intake Disorder) and anorexia nervosa (AN) can look superficially similar — both involve restricted intake, often weight loss, and significant family disruption — but they are driven by different mechanisms and respond to different treatments. The DSM-5 and DSM-5-TR distinguish them on the basis of body-image disturbance: ARFID does not include fear of weight gain or body-image preoccupation, while AN does (APA, 2022). FBT for AN has the strongest evidence base; FBT-ARFID is a more recent adaptation with growing but smaller evidence (Thomas & Eddy, 2019; Lock et al., 2019). Diagnosis matters, and a child with ARFID treated as anorexia will not respond to the same protocol.
What the research says about the diagnostic difference
DSM-5 introduced ARFID in 2013 to capture restrictive eating in children and adolescents that was clinically significant but did not involve the body-image disturbance characteristic of anorexia nervosa. DSM-5-TR (APA, 2022) maintains the distinction. The core diagnostic criteria for ARFID are:
- Avoidance or restriction of food intake based on sensory characteristics, fear of aversive consequences (choking, vomiting), or apparent lack of interest in eating.
- Significant weight loss, nutritional deficiency, dependence on supplements, or psychosocial impairment as a result.
- No body-image disturbance — no fear of weight gain, no over-evaluation of weight or shape.
- Not better explained by another eating disorder (anorexia nervosa, bulimia nervosa).
Thomas and Eddy's clinical synthesis identifies three commonly recognised ARFID presentations: the sensory-based presentation (food avoided because of texture, smell, or appearance), the fear-based presentation (food avoided because of choking, vomiting, or other aversive event), and the low-interest presentation (food avoided because of low appetite or interoception; Thomas & Eddy, 2019).
Why the distinction matters clinically
The two conditions overlap in their consequences — weight loss, growth disruption, social impairment — but diverge sharply in their drivers. This has direct treatment implications.
- In anorexia, the restriction is in the service of the illness's preoccupation with weight and shape. The treatment target is weight restoration plus addressing the body-image disturbance over time.
- In ARFID, the restriction is in the service of sensory aversion, fear of aversive consequences, or low interest. The treatment target is the underlying mechanism — sensory desensitisation, fear reduction, or appetite/interest scaffolding — alongside restoration of intake.
A child with sensory-based ARFID treated as anorexia will be asked to eat foods that are physiologically aversive, with no addressing of the underlying sensory mechanism. A child with anorexia treated as ARFID will have the body-image driver left unaddressed, and intake increases will not generalise. The research base treats accurate diagnosis as a precondition for the right protocol (Thomas & Eddy, 2019; Lock et al., 2019).
Where the two conditions can overlap
The literature also recognises that the boundary is not always clean. Several patterns appear in the research and clinical observations:
- A child can have ARFID and develop AN later, particularly during early adolescence when body-image concerns become more salient.
- A child can have AN that initially looks like ARFID if the body-image disturbance is hidden or denied. Diagnostic workup needs to probe this directly rather than relying on the child's self-report.
- Sensory aversion can co-occur with AN, particularly in autistic adolescents, where the sensory driver may be present alongside genuine body-image concerns.
- Comorbid anxiety, OCD, and autism are more common in ARFID than in AN, and the comorbidity profile often shapes the treatment approach (problem 9 below for anxiety/OCD; Thomas & Eddy, 2019).
The clinical implication is that the differential diagnosis is the treatment team's work, not the parent's, and that a re-evaluation across treatment is sometimes warranted if the response does not match the diagnosis.
What the research says about treatment for ARFID
ARFID's evidence base is younger than AN's, but several approaches have growing support:
- Family-based treatment for ARFID (FBT-ARFID), adapted from the FBT-AN protocol, has shown promise in case series and small trials, particularly for adolescents with low-weight ARFID where weight restoration is a clinical priority (Lock et al., 2019). The structural elements — parents leading meals, externalisation of the problem, treatment team coaching — are similar to FBT-AN, but the in-meal work is calibrated to the underlying mechanism (sensory exposure, fear reduction).
- Cognitive behaviour therapy for ARFID (CBT-AR), developed by Thomas and Eddy, is a manualised individual therapy targeting the specific ARFID presentation (sensory, fear-based, or low-interest). It has the most-developed evidence base for older adolescents and adults with ARFID (Thomas & Eddy, 2019).
- Multidisciplinary outpatient management — dietitian, paediatrician, psychologist, sometimes occupational therapist for sensory work — is the typical structure across both modalities.
- Inpatient stabilisation is sometimes required for severe weight loss or nutritional deficiency, particularly with the low-interest presentation in younger children.
What the research suggests not doing
- Do not assume the diagnosis based on the visible behaviour. A child who refuses food can be doing so for very different reasons; the workup is the treatment team's specialised task.
- Do not apply unmodified FBT-AN to a clear ARFID presentation. The body-image-focused parts of FBT-AN do not address the ARFID driver, and the unmodified plate-and-portion approach can intensify sensory or fear-based aversion.
- Do not delay assessment because "it's just a phase." Both ARFID and AN can produce serious medical and developmental consequences in adolescents, and both have better outcomes with earlier treatment (AED, 2021).
- Do not assume sensory issues alone rule out AN. Sensory and body-image drivers can coexist, particularly in autistic adolescents. The diagnostic question is which drivers are present and active, not which one is "the real one."
What the research suggests doing if the diagnosis is unclear
- Get a specialised assessment with a clinician experienced in both ARFID and AN. The referral pathway typically goes through a paediatrician to a specialist eating-disorder service.
- Bring concrete examples to the assessment — specific foods refused and the reason given, history of aversive events, family history of eating disorders or autism, observed body-image-related behaviours (clothes, mirrors, weighing).
- Be open to a revised diagnosis across treatment. A child initially diagnosed as ARFID who develops body-image-driven behaviours during early adolescence may need re-evaluation; a child initially diagnosed as AN who shows sustained sensory aversion may benefit from sensory-specific work alongside the AN treatment.
- Treat the diagnosis as a working hypothesis informing the protocol, not a label fixed for life. The clinical literature supports flexibility in revisiting the diagnosis when the treatment response does not match expectations.
How the protective frames apply across both
Several research-backed frames apply across both ARFID and AN, even where the protocols differ:
- Externalisation of the food-related problem (the sensory aversion, the fear, the illness) is a useful frame in both, though the language is calibrated to the diagnosis.
- Caregiver-led structure at meals — predictable timing, clear plate construction, end conditions — supports both protocols, though the in-meal work differs.
- Co-parent alignment matters in both. Inconsistency between caregivers undermines treatment regardless of diagnosis.
- Sibling protection matters in both. The household reorganisation around restrictive eating affects siblings whether the driver is sensory aversion or body-image disturbance.
Related questions
References
- American Psychiatric Association (2022). Diagnostic and Statistical Manual of Mental Disorders (5th ed., text rev.). American Psychiatric Publishing.
- Thomas, J. J., & Eddy, K. T. (2019). Cognitive-Behavioral Therapy for Avoidant/Restrictive Food Intake Disorder: Children, Adolescents, and Adults. Cambridge University Press.
- Lock, J., Sadeh-Sharvit, S., & L'Insalata, A. (2019). Feasibility of conducting a randomized clinical trial using family-based treatment for avoidant/restrictive food intake disorder. International Journal of Eating Disorders, 52(6), 746–751.
- Lock, J., & Le Grange, D. (2013). Treatment Manual for Anorexia Nervosa: A Family-Based Approach (2nd ed.). Guilford Press.
- Academy for Eating Disorders (2021). Medical Care Standards Guide (4th ed.).
---
Unseen Progress publishes long-form caregiver research and builds research-backed daily trackers for the families covered. See the full adolescent eating disorder research overview for the complete framework.