Sleep and the neurodivergent brain: trouble falling asleep, broken nights, and the evidence on melatonin

In a hurry? The short version

• It is not just willpower. ADHD and autism both involve real differences in the body clock. In ADHD the whole sleep cycle tends to run late (a delayed circadian phase); in autism the brain's own melatonin signal can be lower or mistimed. Both make falling asleep, and staying asleep, genuinely harder.
• Start with the parts that are not a pill. Consistent timing, morning daylight, dimming evening light and screens, and, for persistent insomnia, cognitive behavioural therapy for insomnia (CBT-I) are the first-line, best-evidenced treatments. Melatonin works best on top of these, not instead of them.
• Melatonin is a body-clock signal, not a sedative. It nudges when you feel sleepy more than it knocks you out. Timing often matters more than dose.
• The evidence is strongest in autistic children (good-quality trials), reasonable in children with ADHD, and thinner in adults, where it tends to help the timing of sleep but relapses once stopped.
• "Microdosing" is a wellness term, not a clinical one. But the idea behind it, that a small, physiological dose (around 0.3 to 0.5 mg) can work as well as or better than the usual 3 to 10 mg, has real support, especially for shifting a delayed body clock. More is not better, and can be worse.
• Two important catches in the UK: melatonin is a prescription-only medicine here, not an over-the-counter supplement, and the unregulated products sold online vary wildly in how much they actually contain. This is a conversation to have with a doctor.

If you are autistic or have ADHD, there is a good chance that sleep is one of the quiet, grinding difficulties of your life: the hours lying awake with a mind that will not switch off, the 2 am alertness when the rest of the house is asleep, the broken nights, and the mornings that feel like wading through wet sand. It is one of the most common things people raise with me, often almost as an afterthought, as though it were simply a personal failing rather than part of the neurology.

It is not a failing, and it is not only about screens or discipline. This piece sets out what the research actually shows about why neurodivergent brains struggle with sleep, what helps most, and then looks carefully at the question I am asked about most: melatonin, and specifically the low-dose or "microdosing" approach that has become popular online. I have tried to be even-handed, to separate what is well established from what is plausible-but-unproven, and to be honest about where the evidence runs out.

Why the neurodivergent brain finds sleep so hard

Sleep problems are not an occasional add-on to ADHD and autism; they are woven through both, and the reasons are partly biological. It helps to take the two separately, because although the experience can look similar from the outside, the mechanisms differ.

ADHD: a body clock that runs late

The most consistent finding in ADHD is that the whole circadian rhythm, the internal 24-hour clock, tends to be shifted later. Studies measuring "dim-light melatonin onset", the evening moment when the brain starts releasing its own sleep-signalling melatonin, find that in adults with ADHD it arrives roughly 90 minutes later than in others.1 In plain terms, the body's "it is time to wind down" signal turns up late, so you are wide awake at midnight and then exhausted in the morning. Delayed sleep phase, the clinical name for this pattern, is strikingly common in ADHD, present in a large proportion of adults with the diagnosis.1,2

This sets up a vicious circle. Late sleep onset means too little sleep, and short or poor sleep worsens exactly the things ADHD already taxes: attention, working memory, emotional regulation, and impulse control. The relationship runs both ways, each side feeding the other.2 Stimulant medication, which helps many people enormously in the daytime, can also push sleep onset later if taken too late in the day, which is worth reviewing with a prescriber rather than simply enduring.

Autism: a melatonin signal that can be lower or mistimed

In autism the picture centres more on melatonin physiology itself. Between roughly half and four-fifths of autistic people experience significant sleep difficulties, far above the general population.3 A consistent thread in the research is altered melatonin biology: a number of studies have found lower night-time melatonin levels, and sometimes higher daytime levels, in autistic people, alongside variations in the genes that govern the final steps of melatonin production (notably the ASMT gene).4,5 The body's nightly sleep signal, in other words, can be quieter or out of step than it should be.

On top of this sit the things that are easy to overlook: sensory sensitivity that makes a "normal" bedroom feel too bright, too loud, or too itchy to settle in; anxiety and a mind that replays the day; and a need for predictability that an erratic sleep routine offends. These are not separate from the biology, they interact with it, and they matter for what helps.

For many people, of course, ADHD and autism co-occur, and so can the two patterns: a late-running clock and a weak melatonin signal together. That combination can make sleep one of the hardest things to get right, which is all the more reason to approach it methodically rather than reaching straight for a bottle of pills.

Start with the parts that are not melatonin

This is the section people skip, and it is the most important. Across the board, the best-evidenced treatment for persistent insomnia, neurodivergent or not, is not a drug at all: it is cognitive behavioural therapy for insomnia, or CBT-I, which professional guidelines recommend as the first-line treatment ahead of medication.6 It addresses the habits and the anxious associations with sleep that keep insomnia going, and for most people its effects outlast anything a pill provides. It can be adapted for neurodivergent people, and there are good digital versions where access to a therapist is limited.

Alongside that, a handful of measures genuinely move the body clock, which is the real target when sleep runs late:

• Get daylight early. Bright light in the morning is the single strongest signal for pulling a delayed clock earlier. Light is not a gentle add-on here, it is the main lever.
• Dim the evening, especially blue light. Bright light and screens late at night tell the brain it is still daytime and push melatonin even later. Lowering the lights for an hour or two before bed matters more than most people expect.
• Hold the timing steady. A consistent wake time, even at weekends, anchors the clock. For neurodivergent people this is hard and a little dull, which is exactly why it is worth building support around rather than relying on willpower.
• Treat the obvious culprits. Late caffeine, late stimulant doses, alcohol, and an over-stimulating wind-down all undermine the rest, and are worth addressing before adding anything.

Melatonin, where it has a place, works best layered on top of these foundations. Used instead of them, it tends to disappoint.

What melatonin actually is, and is not

Melatonin is a hormone your own brain makes each evening as darkness falls. Taken as a medicine, it does two rather different jobs, and confusing them is the source of most disappointment.

The first is as a mild hypnotic: a larger dose near bedtime can make you a little sleepier. The second, and arguably more useful for neurodivergent sleep, is as a chronobiotic: a small dose taken at the right time can shift the body clock itself, gently pulling a late-running rhythm earlier so that you naturally feel sleepy sooner.7,8 The crucial point is that for this clock-shifting effect, when you take it can matter more than how much you take, and the right timing is typically several hours before your current sleep onset, not at the moment your head hits the pillow.

This is why melatonin is best understood as a signal rather than a sedative. It does not switch you off the way a sleeping tablet does. It tells your brain that night has arrived, and in a brain whose own night-time signal is late or faint, that can be genuinely valuable. It also explains why people who take a large dose at bedtime and feel nothing conclude it "does not work", when a smaller, earlier dose might have done.

The evidence, told honestly

In autistic children: the strongest ground

This is where melatonin has its best evidence by some distance. A systematic review and meta-analysis found that melatonin improved both how quickly autistic children fell asleep and how long they slept, with few side effects.9 A well-conducted 13-week randomised controlled trial of a prolonged-release paediatric formulation in 125 children with autism found meaningful improvements in total sleep time and in how long it took them to fall asleep, with benefits to the children's daytime behaviour and to their parents' quality of life.10 Follow-up over two years suggested it remained effective and well tolerated, with no signal of harm to growth or development.11 On the strength of evidence like this, a prolonged-release melatonin (Slenyto) is licensed in the UK specifically for sleep problems in autistic children.

In children with ADHD: reasonable, but narrower

A randomised controlled trial in 105 medication-free children with ADHD and long-standing trouble falling asleep found that melatonin advanced sleep onset by more than half an hour in about half of those treated, and was well tolerated.12 Importantly, though, it improved the sleep, not the daytime ADHD traits: there was no measurable change in behaviour, cognition, or quality of life. That is an honest and useful finding, and not a disappointing one. ADHD is lifelong, part of how a brain is wired rather than anything to be treated away, and melatonin was only ever going to help with the sleep. Better rest can make ADHD easier to live alongside, which is worth a great deal in itself.

In adults: thinner ground

Here the evidence is more limited, and honesty requires saying so. The most relevant work comes from research on adults with ADHD and delayed sleep phase, where melatonin (often combined with bright light) successfully advanced the body clock, measured by that dim-light melatonin onset, and modestly eased symptoms. But there is a sting: when treatment stopped, the clock and the symptoms drifted back to where they started within a couple of weeks.13 The lesson is not that melatonin fails, but that it manages a tendency rather than curing it, and that the behavioural anchors above are what make any gain stick.

The "microdosing" question: does less work better?

First, a piece of plain speaking: "microdosing melatonin" is a popular wellness phrase, not a recognised clinical category. It usually means taking a small, near-physiological dose, somewhere around 0.3 to 0.5 mg, rather than the 3 to 10 mg that fills most shop shelves. So the useful question is not whether "microdosing" is a real thing, but whether low doses work as well as high ones. And on that, the science is genuinely interesting.

A healthy young adult's own night-time melatonin amounts to only a few tenths of a milligram, so a 0.3 mg dose roughly recreates the body's natural night-time level, whereas 5 mg floods the system with around ten to twenty times that. More turns out not to be better. In older adults with insomnia, a 0.3 mg dose restored sleep effectively, and did so at least as well as much larger doses, while the higher doses offered no added benefit and pushed blood levels far above the natural range.14 For shifting the body clock specifically, the effect of a 0.5 mg dose is comparable to that of a 3 mg dose, again arguing that the timing and the modest size of the dose, not its bulk, are what count.8 Larger doses can also linger into the morning, leaving a groggy "hangover" and, taken at the wrong time, can even nudge the clock the wrong way.

So the rationale for a low, well-timed dose is sound, and it fits the biology of the neurodivergent sleep problems described above, a late or faint night-time signal that a small, correctly timed top-up can help correct. What I want to be careful about is overselling it. Most of the elegant low-dose work was done in older adults and in circadian-rhythm studies, not in large trials of autistic or ADHD adults specifically. The neurodivergent trials that exist, including the strong paediatric ones, mostly used higher, prolonged-release doses. So "a small, early dose is often the smart starting point" is a reasonable, evidence-informed position; "microdosing is proven for neurodivergent adults" would be overstating what we actually know. The truthful summary is that the principle is well supported and the specific application is still under-studied.

The catches that matter

Three things deserve real attention before anyone reaches for melatonin, and they are the reason this should be a medical conversation rather than an online purchase.

In the UK, melatonin is a prescription-only medicine. Unlike in the United States, where it sits on supermarket shelves, here it is regulated as a medicine. The licensed products are Circadin, for short-term insomnia in adults over 55, and Slenyto, for sleep problems in autistic children and in Smith-Magenis syndrome; beyond those, it is prescribed off-label, including for adults, at a clinician's discretion. That regulation exists for good reasons, and it means the sensible route is through a GP or specialist who can consider your wider health, rather than around them.

The unregulated products are alarmingly inconsistent. When researchers analysed melatonin supplements of the kind sold online, the actual content ranged from 83 per cent below the label to nearly 480 per cent above it, varied between batches of the same product, and in a quarter of cases contained serotonin, a substance that should not be there at all.15 If you cannot trust the dose on the label, the careful low-dose approach becomes impossible to follow, which is another argument for a regulated, prescribed product.

It is generally safe, but not nothing. Melatonin has a reassuring short-term safety record. The common side effects are mild, daytime sleepiness, fatigue, and occasional mood changes, and, contrary to a frequent worry, taking it does not appear to switch off your own production.16 That said, long-term data in adults are still limited, it can interact with other medicines (including some antidepressants and anticoagulants), and it is not recommended casually in pregnancy. These are exactly the things a prescriber is there to weigh.

A sensible way through

Pulling the threads together, the approach I would gently steer most people towards looks like this. Treat the foundations first: morning light, a dimmed evening, steady timing, and, for entrenched insomnia, CBT-I, which remains the most effective treatment we have. Understand melatonin as a way of correcting a mistimed or faint body-clock signal rather than as a sedative, which usually points towards a small dose taken earlier in the evening rather than a large one at bedtime. And make it a conversation with a doctor, both because melatonin is prescription-only here and because a prescribed product is the only way to be sure what you are actually taking.

None of this is a reason for despair if your sleep has been difficult for years. Neurodivergent sleep problems are real, they are biological, and they are also, to a meaningful degree, workable. The aim is not a perfect night every night, which no one gets, but a body clock that is a little less at war with the life you are trying to live.

A gentle clinical note

This article is general information, not personal medical advice, and it is not a substitute for assessment by your GP or a specialist. Melatonin is a prescription medicine in the UK, and decisions about whether to use it, at what dose, and when, should be made with a clinician who knows your history. Persistent insomnia can also be a sign of something else worth looking at, including depression, anxiety, sleep apnoea, or restless legs, which is another reason not to self-treat in the dark. If poor sleep is tipping into hopelessness, or you are struggling to keep going, please reach out: you can contact your GP, call NHS 111, or speak to the

Samaritans

on 116 123, free, any time.

References and further reading

• van Andel E, Bijlenga D, Vogel SWN, Beekman ATF, Kooij JJS. Attention-deficit/hyperactivity disorder and delayed sleep phase syndrome in adults: a randomized clinical trial on the effects of chronotherapy on sleep. Journal of Biological Rhythms. 2022;37(6):673-689.

pubmed.ncbi.nlm.nih.gov/36181304

• Bijlenga D, Vollebregt MA, Kooij JJS, Arns M. The role of the circadian system in the etiology and pathophysiology of ADHD: time to redefine ADHD? ADHD Attention Deficit and Hyperactivity Disorders. 2019;11(1):5-19.

doi:10.1007/s12402-018-0271-z

• Carmassi C, Palagini L, Caruso D, et al. Systematic review of sleep disturbances and circadian sleep desynchronization in autism spectrum disorder. Frontiers in Psychiatry. 2019;10:366. (Sleep problems affect roughly 50 to 80 per cent of autistic people.)

frontiersin.org

• Melke J, Goubran Botros H, Chaste P, et al. Abnormal melatonin synthesis in autism spectrum disorders. Molecular Psychiatry. 2008;13(1):90-98.

doi:10.1038/sj.mp.4002016

• Rossignol DA, Frye RE. Melatonin in autism spectrum disorders: a systematic review and meta-analysis. Developmental Medicine & Child Neurology. 2011;53(9):783-792.

doi:10.1111/j.1469-8749.2011.03980.x

• Qaseem A, Kansagara D, Forciea MA, Cooke M, Denberg TD. Management of chronic insomnia disorder in adults: a clinical practice guideline from the American College of Physicians. Annals of Internal Medicine. 2016;165(2):125-133. (CBT-I recommended as first-line treatment.)

doi:10.7326/M15-2175

• Auld F, Maschauer EL, Morrison I, Skene DJ, Riha RL. Evidence for the efficacy of melatonin in the treatment of primary adult sleep disorders. Sleep Medicine Reviews. 2017;34:10-22. (Melatonin as a chronobiotic and mild hypnotic.)

doi:10.1016/j.smrv.2016.06.005

• Burgess HJ, Revell VL, Molina TA, Eastman CI. Human phase response curves to three days of daily melatonin: 0.5 mg versus 3.0 mg. Journal of Clinical Endocrinology & Metabolism. 2010;95(7):3325-3331.

doi:10.1210/jc.2009-2590

• Rossignol DA, Frye RE. Melatonin in autism spectrum disorders: a systematic review and meta-analysis. Developmental Medicine & Child Neurology. 2011;53(9):783-792. (Improved sleep onset latency and total sleep time.)

doi:10.1111/j.1469-8749.2011.03980.x

• Gringras P, Nir T, Breddy J, Frydman-Marom A, Findling RL. Efficacy and safety of pediatric prolonged-release melatonin for insomnia in children with autism spectrum disorder. Journal of the American Academy of Child & Adolescent Psychiatry. 2017;56(11):948-957.

doi:10.1016/j.jaac.2017.09.414

• Maras A, Schroder CM, Malow BA, et al. Long-term efficacy and safety of pediatric prolonged-release melatonin for insomnia in children with autism spectrum disorder. Journal of Child and Adolescent Psychopharmacology. 2018;28(10):699-710.

doi:10.1089/cap.2018.0020

• van der Heijden KB, Smits MG, van Someren EJW, Ridderinkhof KR, Gunning WB. Effect of melatonin on sleep, behavior, and cognition in ADHD and chronic sleep-onset insomnia. Journal of the American Academy of Child & Adolescent Psychiatry. 2007;46(2):233-241.

pubmed.ncbi.nlm.nih.gov/17242627

• van Andel E, Bijlenga D, Vogel SWN, Beekman ATF, Kooij JJS. Effects of chronotherapy on circadian rhythm and ADHD symptoms in adults with ADHD and delayed sleep phase syndrome: a randomized clinical trial. Chronobiology International. 2021;38(2):260-269. (Gains relapsed after treatment stopped.)

doi:10.1080/07420528.2020.1835943

• Zhdanova IV, Wurtman RJ, Regan MM, Taylor JA, Shi JP, Leclair OU. Melatonin treatment for age-related insomnia. Journal of Clinical Endocrinology & Metabolism. 2001;86(10):4727-4730. (A 0.3 mg physiological dose restored sleep; higher doses no better.)

doi:10.1210/jcem.86.10.7901

• Erland LAE, Saxena PK. Melatonin natural health products and supplements: presence of serotonin and significant variability of melatonin content. Journal of Clinical Sleep Medicine. 2017;13(2):275-281.

doi:10.5664/jcsm.6462

• Besag FMC, Vasey MJ, Lao KSJ, Wong ICK. Adverse events associated with melatonin for the treatment of primary or secondary sleep disorders: a systematic review. CNS Drugs. 2019;33(12):1167-1186. (Generally mild adverse effects; reassuring short-term safety.)

doi:10.1007/s40263-019-00680-w

A note on the evidence. The figures and conclusions above are drawn from peer-reviewed clinical research, cited in full. The strongest trials are in children, particularly autistic children; the adult and low-dose evidence is more limited, and I have tried to flag the difference rather than blur it. Where this article and your own doctor's advice differ, follow your doctor.