What this is
Something to do
with the next hour.
This is for the stretch right after a meal, when the room gets loud and there's nowhere to put it. There's no right way to use it and nothing here is watching you.
Here's everything that's inside
So there are no surprises.
- Breathe. A circle that grows and shrinks. You follow it. Longer out than in.
- Draw. A blank space, or a faint pattern to fill in. Your finger, no skill needed.
- Drift. Slow moving water and light. Nothing to do.
- Cards. One sentence at a time. Two decks, pick whichever you can stand today.
You can leave any of it whenever you want. Nothing keeps count.
What this doesn't do
No numbers of any kind. No streaks, scores or progress. No logging what you ate or how you felt. No camera. No account. Nothing is saved, so tomorrow it starts blank again and you haven't lost anything.
Breathe
Breathe
In as the circle grows. Out as it shrinks, slower. Through your nose if you can, mouth is fine.
Stay sitting or lying down. Don't hold your breath at either end. If you feel lightheaded, stop and tell a staff member.
Draw
Move your finger around.
There's nothing to make. Scribbling counts. Filling in the pattern counts.
Nothing here is saved or seen by anyone. Clearing it is not undoing progress.
Drift
Just look at it.
Ten or fifteen minutes is about right. Less is fine. Looking away is fine.
Cards
Cards
That happens
This making it worse is a real thing, and it isn't yours to fix alone.
Quiet and stillness can turn the volume up instead of down. Roughly four in ten young people in one study of a guided breathing programme said it made them feel worse, not better. You didn't do it wrong.
Tell a person
If you're in a programme right now, the fastest thing is the nearest staff member. Saying "this hour is bad" is a complete sentence. You don't have to explain it well, and asking is not a setback.
If there's no one in the room
- 988 reaches the Suicide and Crisis Lifeline in the US by call or text, any hour, for any kind of crisis, not only suicide. 988lifeline.org
- ANAD runs a free peer support helpline and support groups for eating disorders.
- NEDA lists current screening and support options.
- Outside the US, Beat in the UK and Butterfly in Australia both run helplines.
Or try a different shape
Some people do better with their hands busy than with their eyes closed. Drawing and moving beat sitting still for a lot of people.
Recovery from an eating disorder happens, including for people who've been ill a long time. A bad hour is not a verdict on that.
Why this
Every choice in here comes from a study, including the choices to leave things out.
Written for the people who have to review this before it goes on a unit, and for anyone using it who wants to know what it thinks it's doing.
Why the hour after a meal
In a specialist unit, distress after lunch sat at about eight out of ten and dropped by less than a point under standard support. A structured hour moved it more than twice as far. Young people interviewed about that window said it was often harder than the meal itself, and named the thing that helped as simple interruption. In their words, it stopped you thinking about food.
Bibb, Castle & Newton 2015 · Monaghan & Doyle 2023 · Ellis et al. 2024
Why a plain circle and no heart rate sensor
A fixed expanding circle at six breaths a minute produced the same physiological result as adaptive heart rate biofeedback, with none of the hardware. That matters twice over here, because heart rate variability runs high in acute anorexia as an effect of starvation, so a biofeedback score would reward the illness and hand the user one more number to chase.
Tabor et al. 2022 · Jenkins et al. 2021 · Lehrer & Gevirtz 2014
Why seated only, and no breath holding
Blood pressure response to standing is blunted in anorexia and can stay that way after weight restoration. In the one biofeedback study run on an eating disorder unit, a participant stopped after feeling faint. So no posture changes, no held breaths, no hyperventilation phases.
Jenkins et al. 2021 · Scolnick, Mostofsky & Keane 2014
Why drawing, and why nothing about the meal
Children told to draw as a distraction improved in mood roughly twice as much as children told to draw about the upsetting thing. What predicted the lift was absorption and feeling capable, not enjoyment or how good the drawing was. Structured designs beat free-form for anxiety, though a mandala specifically is no better than any other structure, so both options are here.
Drake & Winner 2013 · Drake 2021 · Curry & Kasser 2005 · Støre et al. 2022
Why the drift scene is flat, not immersive
In a meta-analysis of simulated nature, medium immersion produced the largest gain in positive affect and full immersion produced the smallest. A calm two dimensional scene is the empirically better choice, not the budget one. Ten to fifteen minutes is where the anxiety effect peaks.
Simulated nature meta-analysis 2023 · Chen, Yan & Yu 2025 · Hagerhall et al. 2008
Why two decks of cards
Self compassion training helped people with eating disorders, but only the ones who weren't afraid of self compassion to begin with. Fear of a kind voice is measurable and common. So the warm deck has a plain twin that says the same things without the warmth, and switching decks is one tap and no explanation.
Kelly & Carter 2015 · Kelly & Tasca 2016
Why you're never asked how you feel
Difficulty identifying and describing emotions is elevated across eating disorders at medium to large effect sizes. A tool that opens with a feelings question locks those users out at the door. So entry is a picture and a tap.
Westwood et al. 2017 · Leppanen et al. 2022
Why there are no numbers anywhere
Three separate safe messaging guidelines say the same thing. No weights, no calories, no measurements, no counts of behaviour, no before and after. Scales, tape measures and empty plates are named as prohibited imagery, which rules out progress rings and gauges too, since a ring is a measuring instrument wearing a friendly face.
Mindframe, NEDC & Butterfly 2021 · Beat media guidelines · NEDA
Why there are no streaks and no scores
Diet and fitness app design produced eight distinct harms in women with eating disorder behaviours, and those harms held even when the person's stated goal was recovery. Streaks, reminders and red or green feedback were named directly. Perfectionism in anorexia runs at large effect sizes against healthy controls, and the striving kind runs highest of all, so achievement mechanics aren't motivation here, they're fuel. Gamification buys engagement and not outcomes.
Eikey 2021 · Dahlenburg et al. 2019 · Ngabo-Woods et al. 2026
Why the layout never moves and the words are short
Adults with anorexia underperform across neuropsychological testing at a moderate effect, and body mass index moderates it, so patients early in refeeding sit at the impaired end. Set shifting is impaired across eating disorders and worst in the restricting subtype. Anxiety itself eats attentional capacity. So one decision per screen, no wizards, no timed choices and controls that stay exactly where they were yesterday.
Stedal et al. 2021 · Wu et al. 2014 · Eysenck et al. 2007
Why everything is told to you up front
Intolerance of uncertainty is one of the largest measured differences between people with eating disorders and everyone else, and it's larger still in anorexia. Ambiguity is not a neutral design cost in this population, so the welcome screen lists the whole contents before you start and nothing appears that wasn't announced.
Brown et al. 2017 · WCAG 2.2.1 Timing Adjustable
Why sound is off and motion can be turned down
Sensory sensitivity is raised in anorexia and survives adjustment for body mass index, while sensation seeking is lower. In adolescent programmes, a large share of patients are autistic or have attention deficit hyperactivity disorder. Default quiet, let people add. Animation also has to be pausable and reducible under the accessibility guidelines, which a breathing circle would otherwise fail.
Saure et al. 2022 · Kinnaird et al. 2020 · ARFID at higher levels of care 2024 · WCAG 2.2.2
Why the colours are muted, and what Dim is for
Saturation drives arousal harder than hue does, with a very large effect, and it moves skin conductance and not only self report. Desaturation damps affective intensity in both directions, which is the point for a distressed user. Backgrounds stay quiet while text and controls hold full contrast. Dim lowers luminance and pulls out short wavelength light for use near lights out, because ordinary room light below two hundred lux suppressed melatonin in ninety nine per cent of people tested.
Wilms & Oberfeld 2018 · Lin, Mottaghi & Shams 2024 · Gooley et al. 2011 · WCAG 1.4.3
What this is not
It's not treatment, it's not a medical device and it isn't a substitute for a person in the room. Regulators have already ruled in a related context that a screen cannot stand in for staff presence during high risk monitoring, and that's the right rule. This sits beside supervision, it doesn't thin it out.
42 CFR 483 Subpart G · SAMHSA trauma informed principles 2014
Where it's honest about not knowing
There's no randomised trial showing that any app improves outcomes in diagnosed anorexia. Post meal relaxation has one small trial from 2008 and it found no winner among the modalities, which is why this offers a menu instead of a protocol. Enforced stillness after meals may not even be right, since supervised light activity halved ritual behaviours compared with rest. Treat this as a well grounded prototype, not a proven intervention, and measure it.