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. Externalising the illness — talking about "the anorexia" as a separate entity from the adolescent — is a core technique in family-based treatment, articulated explicitly in the Lock and Le Grange manual (Lock & Le Grange, 2013). It is not a metaphor for parents to feel better; it is a clinical move with two measurable effects: it protects the parent-child bond during refeeding, and it stabilises parental persistence by reframing whose anger the parent is sitting through. The research suggests doing it deliberately, in language, and across both meal and non-meal contexts.
The technique was articulated in the original Maudsley protocol (Dare & Eisler, in the 1980s and 1990s) and elaborated in Lock and Le Grange's manualised FBT. The core claim is that anorexia is not an aspect of the adolescent's personality or values but an ego-dystonic illness that hijacks the adolescent's machinery for choice around food. Treatment is therefore positioned as parents and child together fighting the illness — not as parents fighting the child, and not as the child fighting herself (Lock & Le Grange, 2013).
Eisler and colleagues' work on multi-family therapy in adolescent anorexia uses externalisation as a structuring device across families: when several families in treatment compare notes, the shared frame of "what the anorexia is doing" makes the illness's tactics visible across cases (Eisler et al., 2016). Children in recovery commonly report afterwards that they did not experience themselves as angry during refeeding — they experienced the anorexia as angry, which is the felt-state version of the same separation the technique formalises.
Clinical and qualitative research on FBT identifies two consistent effects of consistent externalisation by parents.
Without externalisation, the adolescent's hostility at the table reads to the parent as my child is angry at me, my child wants to hurt me, my child no longer trusts me. With externalisation, the same hostility reads as the illness is angry, my daughter is the hostage. The first frame produces guilt and withdrawal; the second produces persistence and tenderness. Parents who hold the externalised frame report being able to remain warm with their child while holding the meal plan firm — which is the FBT target state.
Lock and Le Grange describe the technique as a tool for parental endurance, not just child wellbeing. Sitting through three to six daily episodes of resistance for months is a different psychological task when the resistance is framed as symptom rather than judgment. Parents who have not been coached on externalisation tend to wear down faster; parents who use it consistently report being able to sit through the same intensity of refusal at month 4 that would have broken them at month 1.
The FBT clinical tradition gives several concrete patterns parents can use across the day.
Some families name it formally — "the anorexia", "the ED", or a chosen name (often a name the child suggests, sometimes a deliberately mocking one). Some keep it generic. The clinical research is agnostic on which form is used; what matters is that the entity is named and used consistently across the family.
This is the use parents most often miss. Externalisation is also a tool for the parent's own state, said silently. "That glare across the table is the anorexia, not my daughter." "That sentence I just heard is something the illness gave her." Parents who do this report that the same minute-to-minute experience becomes more bearable, and that they recover faster between meals.
Sometimes externalisation feels false to a particular family — culturally, generationally, or because of the specific child's temperament. The Lock and Le Grange manual is permissive about this: the separation between illness and child matters, the specific words do not. Some families never name "the anorexia" but consistently treat the illness as separate through behaviour — "I know this is not what you actually want", "this is the part of getting better that's hard for everyone" — and achieve the same effect.
The research-relevant point is that some form of separation should be operating, in language or in stance. Families in which the parent treats the eating-disorder behaviours as the child's preferences, values, or character tend to get more rigid and more guilt-laden over time, regardless of whether they use the formal vocabulary.
Externalisation works best when both parents (or carers) use it consistently. A household in which one parent externalises and the other treats the eating-disorder behaviours as the child's choices creates a split the illness can exploit — the child can address the non-externalising parent and bypass the frame. Co-parent alignment on externalising language is one of the small structural choices that has outsize effects across months of treatment (problem 4 of the overview).
---
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.