Autism-affirming
eating disorder
care runs into
a longstanding
fear of
accommodating
the disorder

A rule that works — until it meets autism

Much of traditional eating disorder care rests on a principle that is easy to state and hard to hold: do not accommodate the disorder. Accommodation, in the literature the study below draws on, means the behaviours carers or clinicians may engage in that unintentionally maintain an eating disorder. The belief in its risks runs deep, especially for anorexia nervosa.

Now put an autistic patient in front of that principle. One clinician in the study below described patients who find it hard to have foods touching on a plate, or who don’t like a wet texture with a dry one — preferences that, in the clinician’s words, were there “well before the eating disorder”. Is making room for that accommodating the disorder, or accommodating the person? The rule does not say.

A study published in the European Eating Disorders Review, accepted on 16 July 2026, asked clinicians inside UK eating disorder services how they are handling that question. Emy Nimbley and Amelia Austin, with Kate Tchanturia, Kai Thomas, Catherine Jones, Karri Gillespie-Smith and Fiona Duffy, interviewed 12 clinicians — nurses, dietitians, psychological therapists, a psychiatrist, a clinical psychologist and an occupational therapist, across adult and child and adolescent services — whose services were beginning to adopt autism-affirming care.

The fear of feeding the disorder

The study’s overarching barrier was what the authors call a longstanding belief about the risks of accommodating an eating disorder. Clinicians described anxiety in both directions: that failing to adapt care for autistic patients could be aversive and cause harm, and that adapting it too far could accommodate the eating disorder and make it more entrenched. Some described suspicion about the intentions behind requested adaptations. One put it plainly:

we need to make sure that this isn’t a convenient door for the eating disorder to keep its foot jammed in for any longer than is necessary.

That suspicion is not irrational. The authors link it to the egosyntonic nature of some eating disorders — the illness can feel like part of the self. But it creates a problem that autism makes acute. Autistic traits may be misinterpreted as eating disorder pathology, and clinicians described uncertainty about how to challenge the eating disorder without challenging adaptive autistic traits in a harmful way. Four clinicians said that trying to separate the autism from the eating disorder was itself a barrier.

When “evidence-based” means “manualised”

The second barrier was quieter and, arguably, more structural: an assumption that evidence-based practice means the manualised treatment and nothing else. One clinician said that going off the model “makes it not FBT [Family Based Treatment], makes it not an evidence-based intervention, and that’s risky.” Another described the opposite risk — Family Based Treatment practitioners who knew a young person was autistic, were clearly struggling to deliver the intervention, and still did nothing to adapt it.

The authors are careful here. They note that randomised controlled trials have been treated as the gold standard for evidence-based practice, while also having limits in how representative their patients are, and that the original definition of evidence-based practice included clinical expertise and patient choice as well as research evidence. The manual is evidence. It is not the whole of it. What most of the clinicians interviewed said helped was reflective practice and supervision: a space in which a team can try an adaptation, admit when it has gone wrong, and try something else.

What helped

The enablers the study found read like an alternative model of care rather than a list of fixes:

• Individualised formulation — understanding the person’s needs rather than untangling two diagnoses. Six clinicians favoured screening and developmental interviews over waiting for a formal autism diagnosis; the authors suggest this preference for needs over diagnosis may partly reflect barriers to formal assessment.
• A “return to before” approach to food — asking what a person ate, and how, before they became unwell.
• A diverse multidisciplinary team — occupational therapists were the enabler named most often, for their skills in sensory assessment, followed by dietitians, autism leads and, for two participants, autistic colleagues.

And the barriers at system level were the familiar ones: autism training that was helpful but not about eating disorders, autism teams that, some said, had little appreciation of how dangerous eating disorders are, and, for some, commissioning that restricts how far they can go into autism work. “We are not commissioned to be an autism supporting service,” one clinician said.

The protocol and the person

The study’s conclusion is measured: autism-affirming adaptations “appear to challenge some core underpinning messages historically embedded in siloed ED services”. Of the core messages it names — what counts as evidence, and the risks of accommodating disordered eating — the second is the one this piece began with. Much of traditional care rests on that belief, delivered in services organised separately from the ones that understand autism. The rule was not built for a patient whose non-negotiable needs can look, from inside the protocol, like the illness’s defences.

What the clinicians who are making it work describe is not a new manual. It is a change in where the evidence starts: from the person, and what they were like before they became unwell, rather than from the protocol and what it permits. That is harder to commission, harder to train and harder to audit. It is also what, on these clinicians’ account, made autism-affirming care possible. The fear of feeding the disorder is legitimate. A service that treats a person’s needs as the disorder’s defences is feeding something else — and it will keep doing so for as long as autism and eating disorders are treated as two diagnoses to untangle rather than one person to understand.

Citations

Nimbley, E., Austin, A., Tchanturia, K. et al. (2026) — Clinician Perspectives on Barriers and Enablers to Autism-Affirming Eating Disorder Care — European Eating Disorders Review