Hard eating moments
How the app supports a patient through a loss-of-control eating moment with compassion first, how forbidden-food rules get gently dismantled instead of avoided, and the one toggle you hold over behavioral experiments.
Some of your patients live with binge-spectrum eating. The instinct, and the thing a lot of apps do, is to count the moment, name it, and coach the patient to steer clear of the food that "set them off." The evidence says that backfires: restriction and forbidden-food rules are what manufacture the next episode (restraint theory, the abstinence-violation effect). So this feature is built the other way around. The app never uses the word "binge" to the patient, never keys on how much they ate, and works toward planned inclusion of feared foods rather than avoidance. The Hard eating moments card is your window into that work, and it holds the one switch you may want to keep for yourself.
What the patient never sees
The single most important thing to understand about this feature is what it withholds. The app keys on the experience of losing control, never on amount or calories, and it never hands the patient a verdict.
- It never shows the word "binge." When a hard moment is recognized, the patient describes it in their own words. The app mirrors their language back; it never labels the moment for them.
- It never keys on how much they ate. Recognition comes from the experience, a "kept eating past comfort" urge plus how it felt, or a tight cluster of logs in one evening, not from a number.
- It never counts calories at the patient. No amount talk appears anywhere in these flows. That is a hard rule for this population, where calorie-facing tools measurably worsen outcomes.
The support moment (compassion first)
When the app recognizes a hard eating moment, the patient does not get a lecture or an immediate interrogation. They get a short, warm flow, and the order is deliberate: self-compassion comes first (Kelly & Carter found self-compassion training halves binge days; self-criticism is the engine of the cycle). Only then does the app offer anything else.
The anti-compensation line is doing quiet clinical work: it is both the abstinence-violation interceptor and the purge guard. If the patient does report an urge to compensate, the app routes them to the existing coping tools and, where relevant, the same support resources your safety plan uses. That question is always optional, never required, and the exit is warm no matter what happened.
The next day, never the same evening, the app can offer a gentle guided review journal, an "after a hard food moment" reflection that walks the patient through what the day was like beforehand, whether they were alone, the feeling before and after, and whether a food rule was in the mix. It lives in the patient's private, local-only journal. The patient can also author an alternative-activities list, a few things to reach for when an urge crests, since urges tend to pass in twenty to thirty minutes if they are not fed.
Foods with rules, and why we include them
The founding idea here, a patient's "association" with a food like ice cream, is real, but the evidence reframes it. A food that keeps showing up in hard moments is almost always a food the patient has a rule against. The rule causes the episode (deprivation, rule-break, "what the hell," episode, shame, a stronger rule), so the fix is planned, unremarkable inclusion, the CBT-E avoided-food reintroduction move, not avoidance.
Telling a patient to keep a feared food out of the house is exactly the intervention that manufactures the next binge. Restraint is the cause, not the cure. So the app moves toward including the food on purpose, in daylight, planned, until it becomes unremarkable, rather than flagging it as dangerous.
The patient keeps a private "foods I have rules about" list. It syncs, so when a patient is attached to you it is visible on your dashboard and you can co-plan from it. Each food carries a tension level, how charged it feels, which is what the card renders as badges.
| Tension | What it signals |
|---|---|
| a little | A mild rule. Often a good first candidate for a planned inclusion. |
| some | A meaningfully charged food. Worth including with a little scaffolding. |
| a lot | A high-anxiety food. Consider supervising the work, or starting elsewhere. |
Behavioral experiments, and the toggle you hold
From that list the patient can run a behavioral experiment, the evidence-based planned-inclusion technique (Fairburn's guided self-help, NICE first-line for BED). The structure is belief, then prediction, then a small planned inclusion, then compare.
The planned inclusion writes a gentle, unremarkable line into that day's meal-plan slot. The patient's normal log and check-in are the experiment record, there is no special tracking. The next day they compare their prediction against what actually happened and re-rate the belief. An episode during an experiment is treated as data, never as failure.
This is the one place you have direct control. The Hard eating moments card carries a Behavioral experiments toggle.
- It is on by default when a patient is attached to you. For most binge-spectrum patients, planned inclusion is exactly the right work, and NICE puts guided self-help first-line.
- You turn it off when feared-food work needs your supervision, most importantly for underweight or anorexia-spectrum patients, where Fairburn's own gate says self-help is not appropriate.
Turning experiments off does not touch the rest: the support moment, the review journal, the alternative-activities list, and the rule-foods list all remain available, because those are safe self-help across the board. Only the experiment composer disappears.
If a patient logs several weeks of hard-moment reviews with no easing, the app gently suggests connecting with a professional and points to the connect code flow. It mirrors the NICE step-up checkpoint, so the app knows when it has reached the edge of what self-help can do.
Where else this shows up
The card is not the only surface. Restrict-then-binge patterns also raise flags in the roster (a restriction-then-episode coupling reads as higher concern, recurring urges as medium), and the underlying logs, the evening clusters, the earlier gaps, the backfilled morning entries, are all readable in the Food Log. The card header and its language respect patient view, so nothing a patient could see during a screen-share ever reads as a "binge" verdict.
ED-safety rails
- Never the word "binge." The patient names the moment; the app mirrors their words and never hands down a label.
- Never amount-based. Recognition keys on the experience of losing control, not on calories or quantity.
- Compassion first. Self-compassion opens the support flow, before any tool, question, or reframe.
- Inclusion, not avoidance. The app moves toward planned inclusion of feared foods and never suggests removing or avoiding one.
- Experiments are gated. They are yours to disable, and a "see a professional" checkpoint is built in for patients who are not easing.
- Off in ARFID mode. Binge-support features are disabled entirely when a patient is in ARFID mode, where the eating challenge is a different one.