The Bedtime Routine That Beat the Sound Machine

Two conversations with a clinician moved more than any sound ever has — here is what was in them

We sell a sleep app, so take this in the spirit it is offered: the best-evidenced thing you can do for an ADHD child's sleep is not a device, a supplement or a sound. It is a routine, built deliberately, and kept boringly the same.

We know that because someone ran the trial.

What the Trial Found

In 2015 the BMJ published a randomised controlled trial by Hiscock and colleagues: 244 children aged 5–12 with ADHD, recruited across 21 paediatric practices in Australia, most of them already on stimulant medication.

Half the families received two consultations about sleep, two weeks apart, plus a follow-up phone call. The other half carried on with usual care. That was the entire intervention.

Sleep problems. At three months, moderate-to-severe sleep problems were present in 30% of the intervention group against 56% of the control group (odds ratio 0.30, p<0.001).
ADHD symptoms. Symptom severity fell further in the intervention group at three months (effect size −0.3) and at six months (effect size −0.4).
Everything else. Behaviour, quality of life and daily functioning improved too, and most of the gains were still there six months later.

Two conversations. For context, a randomised trial of weighted blankets in children with ADHD — a far more popular intervention — produced an average of 7.7 extra minutes of sleep and no change at all in how long it took to fall asleep.

What Was Actually In Those Sessions

The programme is published, so this is not guesswork.

Session one — understanding. How sleep actually works: sleep cycles, sleepiness cues, and what is normal for the child's age. Parents leave knowing what they are aiming at.
Session one — a plan for this child. Not a generic checklist. The clinician works out what kind of sleep problem this is, and picks strategies to match.
Between sessions — a sleep diary. Two weeks of writing down when the child got into bed, when they actually fell asleep, and what happened in between.
Session two — read the diary, keep what worked. Reinforce what is working, troubleshoot what is not, adjust.
Follow-up call. One more chance to ask the questions that only appear once you have tried something.

Notice what is doing the work there. It is not novelty — it is diagnosis, a tailored plan, measurement, and one round of correction. That is a shape you can copy.

The Techniques They Used

Bedtime fading

For a child whose body clock runs late. Move bedtime later, to roughly the time they genuinely fall asleep, so that going to bed is quickly followed by sleeping instead of by an hour of failure. Once they are falling asleep within about 15 minutes for several nights, walk bedtime back 10–15 minutes at a time.

Graduated withdrawal, or "camping out"

For a child who needs a parent present. You stay — but each few nights you sit slightly further from the bed, then by the door, then outside it. The change is gradual enough that no single night is a shock.

The bedtime pass

For a child who keeps getting up. One card, good for one legitimate trip out of the room. Used or unused, it is handed in for the night — and an unused pass earns a small reward in the morning. It has its own randomised trial: fewer exits, less crying out, faster settling, still working at three months. Full instructions here.

How to Run This at Home

Pick the wake-up time and defend it. Same time every day, weekends within about half an hour. A drifting wake-up time undoes everything else you do.
Keep a diary for one week before changing anything. Time into bed, estimated time asleep, number of exits, wake time. One week of honest data will tell you whether you have a "won't sleep" problem or a "can't sleep yet" problem.
Choose the strategy from the diary. Late but calm sleep onset means bedtime fading. Fast asleep once settled, but chaos getting there, means the resistance techniques.
Build a 30–45 minute wind-down and never reorder it. The order of the steps is the signal, more than any individual step.
Put the sound at the same point every night — last. Music starts when the lights go down, not before. It is the final item in a sequence, so it comes to mean "the sequence is finished".
Change one thing at a time, and give it a week. This is the step everyone skips, and it is why nobody knows what worked.

Where the App Fits

Honestly? It is step five, and step five only works if steps one to four are in place.

What it does well is remove friction from the parts of the routine that fail at 3 a.m. or on the twelfth consecutive night. Same as last time is one button that restarts exactly what ran last night — the same track, the same ambient mix, the same volumes, the same timer — so repeating the identical evening costs one tap rather than five decisions; the same thing sits on the home-screen widget, usable without unlocking the phone. Auto Start begins that music at the same time every evening, which makes "same time, same signal" something the phone remembers instead of something you have to. The sleep timer fades out instead of stopping dead, because sudden silence is a change and a change wakes a light sleeper — and a shrinking ring shows how much is left, so the end is visible rather than imagined. And the screen goes dark on its own, announcing it ten seconds beforehand, because a bedtime aid that requires a lit phone in the bedroom is arguing with itself.

That is a real contribution to a routine. It is not a substitute for one — the trial above is the reason.

Common Questions

How long before I see anything?

The trial measured at three months, but families typically notice something within two to three weeks. What you are building is a conditioned association, and associations need repetitions. If nothing at all has shifted after a month of genuine consistency, that is a good moment to talk to your doctor rather than to try a sixth sound.

Does this work for teenagers?

A modified version does. A 2025 randomised trial of a cognitive-behavioural sleep programme for adolescents with ADHD improved sleep habits, sleep-related behaviour and depressive symptoms — although wrist-monitor measures did not shift. With teenagers the routine has to be negotiated with them rather than imposed on them.

We are already doing all of this and it is not working.

Then the next step is a conversation, not another technique. Restless legs, sleep-disordered breathing, anxiety and medication timing all cause sleep problems that no routine will fix, and all of them are things a doctor can act on. A one-week sleep diary is the most useful thing you can bring to that appointment.

This page summarises published studies and clinical guidance. It is not medical advice — for anything about your child's sleep or medication, talk to your paediatrician.