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. During Phase 1 of family-based treatment, the meal is the treatment. Negotiation, bargaining, and "just this once" exceptions are not neutral — each one is a structural concession to the illness that compounds across subsequent meals (Lock & Le Grange, 2013). The research-backed alternative is a small, repeatable set of scripted responses delivered calmly, paired with structural choices about who plates the food, where the meal happens, and when it ends.
The Lock and Le Grange treatment manual is explicit that in Phase 1 of FBT, the act of completing the prescribed meal is the active ingredient — not the conversation around it, not the child's stated motivation, and not the emotional tone of the room. Lock et al.'s 2010 randomised trial established that parent-led refeeding outperforms adolescent-focused individual therapy on weight restoration during active starvation, and the manual attributes much of that effect to parents' willingness to hold the meal plan against resistance (Lock et al., 2010).
This reframes the question parents most often ask. The question is not "how do I make this meal less awful?" — most meals will feel awful for months. The question is "how do I make this meal end with the food consumed?" Those are different problems with different research-backed answers.
Eating disorders in adolescents are notable for what clinicians call cognitive rigidity around food: any opening — a substitution offered, a portion adjusted at the table, a time limit extended — is read by the illness as a precedent. Once a precedent is set, the illness will test it at every subsequent meal (Lock & Le Grange, 2013). Parents who report "the meals are getting harder, not easier" after several months of treatment are usually parents who, with the best intentions, have been negotiating in real time.
The Maudsley parent-coaching tradition frames this as a feature of the illness, not a failure of the parent. The child cannot, in Phase 1, reliably tell the illness from their own preferences. Parental consistency is the external scaffold that compensates for that inability — not because the parent knows better than the child, but because the illness exploits any inconsistency it can find (Eisler et al., 2016).
The research and the FBT clinical tradition converge on four structural decisions parents can make before a meal that meaningfully change how it goes.
Watching the food being served is a recognised trigger for refusal. Plating in the kitchen, then bringing the plate to the table, removes one negotiation point (what goes on, how much) and converts it into a fait accompli. The child can refuse the plate, but they cannot renegotiate its construction.
When two parents both speak during a difficult meal, the child can address the softer voice and avoid the firmer one. The FBT manual recommends one parent leads each meal (with the other as silent backup), pre-agreed in advance. Roles can rotate by meal or by week — what matters is that they are decided before the meal starts (Lock & Le Grange, 2013).
Open-ended meals invite the illness to wait the parent out. A clear time limit (commonly 30–45 minutes), agreed with the treatment team and held consistently, ends the negotiation about negotiation. Food not completed within the window is typically replaced with a calorically equivalent supplement, again per the treatment plan — not as punishment but as the structural answer to "what happens if I don't eat this?"
Parents who try to use the meal to also address the illness, the day, or the child's feelings often find both conversations fail. Lock and Le Grange recommend deliberately small, neutral table talk during Phase 1 meals — weather, a TV show, a sibling's day — anything that does not invite a debate about the food. The processing happens later, with the treatment team or with the parent in a non-meal moment.
Improvising responses during a refusal is one of the most exhausting and least effective patterns in early FBT. Scripted responses, repeated across meals, do two things at once: they reduce the parent's cognitive load and they remove the conversational openings the illness uses to negotiate.
On refusal: "I know this is hard. We're going to stay here with you until this is finished. You don't have to want it. You just have to eat it."
On bargaining ("if I eat half, can I skip dinner?"): "We're not making deals about food. The plan came from the team. We're sticking to it."
On "I'm full": "Your body is still learning what full means. The doctors said this is what getting better looks like right now."
On "you don't understand": "You're right that I haven't lived this. I know the doctors said this is the part of getting better that we have to do together."
On hostility ("I hate you"): said internally, not aloud — "that's the anorexia talking, not my child". Aloud: "I love you. We're going to keep going."
These are not magic words. They work because they remove the conversational openings the illness uses to negotiate, and because they free the parent from inventing new arguments mid-meal — which the illness will always meet with a new counter-argument.
Most parental work in Phase 1 happens between meals, not during them. Specifically: pre-meal alignment with the co-parent (problem 4 of the overview), post-meal recovery for the parent's own state, and weekly review with the treatment team. Parents who only think about meals during meals tend to enter each one already depleted.
The research-backed move is to treat the day around a meal as part of the meal. A 10-minute pre-meal check-in with the co-parent (what's the plan, who leads, what do we say if she refuses), a 10-minute post-meal walk for the parent who led, and a weekly debrief with the treatment team are not optional add-ons — they are the structure that lets the in-meal scripts stay calm.
---
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.