Does Melatonin Work, and Who Actually Needs It?

Melatonin is not a sedative. It is a hormonal timing signal that tells your internal clock biological night has started, and it shifts the phase of your circadian rhythm. That is why it performs well when the clock is misaligned — jet lag, delayed sleep phase, night shifts, blindness — and poorly as a general sleeping pill, where meta-analysis puts the average gain at about seven minutes of faster sleep onset. The lever that matters is not the size of the dose but when you take it: 0.5 to 1 mg several hours before bed moves the clock better than 10 mg swallowed at lights-out.

What melatonin actually is#

Melatonin is produced by the pineal gland under the control of the suprachiasmatic nucleus of the hypothalamus — the body's master clock. Secretion begins in the evening, roughly two hours before habitual sleep onset, peaks in the middle of the night and collapses with morning light. The controlling input is light exposure, not tiredness: bright evening light suppresses the rise, darkness permits it.

Functionally, this is a time stamp rather than a sleep substance. Melatonin does not switch off the cortex. It tells every system in the body — thermoregulation, metabolism, hormone release — that night has begun. The clearest proof is comparative: nocturnal animals secrete melatonin during exactly the same dark hours as we do, even though they are awake and hunting then. The molecule encodes darkness, not sleepiness.

This is the source of nearly every disappointment with it. Someone buys a 10 mg tablet expecting the blunt force of a prescription hypnotic, feels nothing dramatic, and concludes it does not work. In fact it did precisely what it does: it announced night. A weak subjective effect is normal pharmacology, not a product failure.

The honest effect size in insomnia#

The most cited quantitative estimate comes from Ferracioli-Oda and colleagues in 2013, pooling 19 randomised trials and 1,683 participants with primary sleep disorders. Melatonin shortened time to fall asleep by roughly 7 minutes, increased total sleep time by about 8 minutes, and produced a modest improvement in subjective sleep quality.

Those numbers deserve to be said out loud. Seven minutes is a genuine average difference, and it is also nothing like what someone lying awake for an hour is hoping for. Prescription hypnotics shorten sleep onset by roughly 15 to 20 minutes; a full course of cognitive behavioural therapy for insomnia produces gains in the same range and, unlike the drugs, keeps them after treatment ends.

That is why the 2017 American Academy of Sleep Medicine guideline on pharmacological treatment of chronic insomnia issues a recommendation against melatonin for both sleep-onset and sleep-maintenance insomnia. The wording is cautious — a weak recommendation based on low-quality evidence — but the direction is not ambiguous: melatonin is not part of the treatment toolkit for ordinary insomnia.

The overall shape of the evidence is consistent. Melatonin looks reasonable in circadian rhythm sleep disorders and delayed sleep phase, modest in older adults whose own secretion has declined, and weak in primary insomnia in middle-aged adults.

Читайте также: Sleep Hygiene: 12 Rules That Actually Work

Where melatonin genuinely earns its place#

Jet lag#

This is its strongest indication. The Cochrane review by Herxheimer and Petrie, published in 2002, pooled ten randomised trials and concluded that melatonin is remarkably effective at preventing or reducing jet lag when crossing five or more time zones, with the benefit largest for eastward travel. The regimen tested was evening dosing at destination local time, starting on the day of arrival and continuing for a few days.

One detail from that review is worth carrying everywhere else in this article: 0.5 mg was almost as effective as 5 mg, and fast-release formulations outperformed slow-release ones. Even in the setting where melatonin works best, more is not better.

Delayed sleep-wake phase disorder#

Here the internal clock is stably shifted several hours late. The person physiologically cannot fall asleep before two or three in the morning and, left alone, wakes near midday. This is not poor discipline and not insomnia: allowed to keep their own schedule, they sleep normally in both duration and architecture.

Melatonin is being used for its actual mechanism in this case, and the protocol is completely different from bedtime dosing. A low dose of 0.5 to 1 mg is taken 4 to 6 hours before current sleep onset. If you currently fall asleep at 3 a.m., that means roughly 9 to 11 p.m. — hours before bed. Light is not optional here: bright light immediately on waking and dim light in the evening roughly doubles what the supplement can achieve on its own, and the phase shift accumulates by 30 to 60 minutes per day at best.

Shift work#

The 2014 Cochrane review by Liira and colleagues on pharmacological interventions for shift work found that melatonin increased daytime sleep after a night shift by about 24 minutes compared with placebo, but did not shorten time to fall asleep. The authors graded the evidence as low quality. The practical reading: melatonin may buy you a little extra daytime sleep, does nothing for alertness during the night shift itself, and is no substitute for deliberate light management and roster design.

Non-24-hour rhythm in blind people#

Mechanistically the strongest case, clinically the rarest. People with no light perception lose the main synchroniser, so their intrinsic period — slightly longer than 24 hours — drifts freely. Weeks of normal sleep alternate with weeks of severe insomnia and daytime collapse as the internal night rotates out of phase with the calendar. Daily melatonin at a fixed clock time acts as an artificial dusk and can entrain the rhythm. This is a treatment a sleep physician supervises, not a self-care experiment.

Situation	Dose	Timing	Strength of evidence
Jet lag, 5+ time zones	0.5–5 mg, fast release	Destination local evening, 2–5 days	Strong (Cochrane, 2002)
Delayed sleep-wake phase disorder	0.5–1 mg	4–6 h before current sleep onset, plus morning light	Moderate to strong
Non-24 rhythm in blind adults	0.5–3 mg	Same clock time daily, under medical supervision	Moderate to strong
Daytime sleep after a night shift	1–3 mg	Before the daytime sleep period	Low: about +24 min (Cochrane, 2014)
Insomnia in adults	not recommended	—	Low; AASM 2017 recommends against
Insomnia in adults aged 55+	2 mg prolonged release	1–2 h before bed, as a course	Modest, small effect size
Children and adolescents	prescriber-directed only	—	Limited; long-term safety not established

Dose and timing: the part everyone gets wrong#

Retail shelves are dominated by 5 and 10 mg tablets, with 20 mg appearing in "extra strength" lines. Physiological night-time blood levels are reproduced by something in the region of 0.3 mg. A 10 mg dose produces concentrations tens of times higher than anything the body makes and keeps them elevated well into the following morning.

Those doses buy nothing. In phase-shifting studies low doses move the clock at least as well as high ones and sometimes better, because a large, long-lasting dose spills across the phase response curve into the region that shifts the clock the opposite way, partially cancelling its own effect. Side effects, by contrast, scale straightforwardly: morning grogginess, headache, vivid or unsettling dreams, a drop in core temperature.

The practical rules are short.

• You want a mild sleep-onset nudge. 0.5 to 1 mg, 30 to 60 minutes before bed. If it does not work, raising the dose is not the answer — dose is not the limiting factor.
• You want a phase shift. 0.5 mg, 4 to 6 hours before your current sleep onset, daily, moving your target gradually and pairing it with bright morning light.
• Keep courses short. A few days for jet lag, two to four weeks for a phase shift. Melatonin was conceived as a resetting tool, not a nightly habit for years.

Formulation matters too. Fast-release acts as a clean "night has started" signal and suits jet lag and phase shifting. Prolonged release — a prescription 2 mg product in many countries, licensed for insomnia in people over 55 — mimics the overnight secretion profile and targets sleep maintenance rather than sleep onset.

The supplement quality problem#

The main risk here comes from the industry, not the molecule. Erland and Saxena analysed 31 commercial melatonin supplements sold in Canada in 2017. Actual melatonin content ranged from 83% below to 478% above the label claim. Between different lots of the same product, variability reached 465%. Most troubling, serotonin — a controlled substance that has no business being in an over-the-counter product and that interacts with antidepressants — was detected in 8 of the 31 samples.

The consequence is concrete. When you buy "3 mg", the tablet may contain 0.5 mg or 17 mg. That makes careful dose selection largely theoretical and explains a good share of the contradictory experiences people report. Gummies compound the problem: uneven distribution of active ingredient across a batch is harder to control in a confectionery matrix, and they are also the format children reach for.

What to do about it. Prefer products with independent third-party verification (USP, NSF, Informed Choice), or products registered as medicines rather than supplements, where content is held to pharmaceutical specification. And do not assume a low number on the label protects you from an unexpectedly large dose — the label may simply be wrong in that direction.

Safety and who should stay away#

In the short term melatonin is well tolerated. The common adverse effects are daytime sleepiness, headache, dizziness, nausea and unusually vivid dreams. All are dose-dependent and usually resolve when the dose comes down to 0.5 or 1 mg.

One thing melatonin does not do is produce rebound insomnia. Stopping a Z-drug or a benzodiazepine after regular use makes sleep worse than baseline for three to seven days, because the GABA system has adapted to the drug. Melatonin does not act on that system, no withdrawal syndrome has been described, and tolerance has not been demonstrated. If sleep deteriorates when you stop taking it, the likeliest explanation is that the underlying insomnia was there all along and was never treated — which is a reason to look at CBT-I, not to resume the supplement.

Читайте также: Insomnia: What to Try Before Reaching for Pills

The real gap is long-term use. Trials running for years essentially do not exist, and this is not a formality: melatonin participates in reproductive and immune regulation, so taking supraphysiological doses nightly for a decade is genuinely uncharted territory. The honest position is that we do not know, which is an argument for short courses rather than for alarm.

Читайте также: Magnesium: Forms, Doses and Who Actually Needs It

How this differs from CBT-I#

If difficulty falling or staying asleep occurs three or more nights a week for longer than three months and affects your daytime functioning, that is chronic insomnia — a diagnosis, not a supplement problem. The 2021 AASM guideline names cognitive behavioural therapy for insomnia as first-line care: a structured four- to eight-week programme built on sleep restriction, stimulus control, cognitive work on catastrophic beliefs about sleep, and relaxation training.

The distinction is mechanistic. Chronic insomnia is maintained by hyperarousal: worry about sleep raises activation, activation prevents sleep, and the failure feeds the worry. Melatonin does not touch that loop. It reports that night has begun, which the brain of a person lying awake in a dark room already knows. CBT-I attacks the loop directly, matches hypnotics in the short term, beats them in the long term, and persists after treatment ends. Digital versions with demonstrated efficacy exist, which removes the usual excuse about therapist availability.

There is a hidden cost to substituting one for the other. A year spent cycling through melatonin brands and dosages is a year in which the conditioning between bed and anxiety keeps strengthening.

Читайте также: How to Manage Stress: What Actually Works

What to do in practice#

1. Name the problem first. A flight, a stubbornly late body clock or a rotating roster are melatonin problems. Months of broken sleep with a normal schedule are not.
2. Start low. 0.5 to 1 mg. If only 3 or 5 mg tablets are available, split them.
3. Time it against the goal, not against bedtime. For a phase shift, the dose belongs in the early evening, hours before you intend to sleep.
4. Add light. Bright light on waking and dim light in the evening will do more than any increase in dose.
5. Keep the course short and judge it after three to seven days for jet lag, two to three weeks for a phase shift.
6. Check your medicines and conditions before starting, especially anticoagulants and antidepressants.
7. If it does not work, do not escalate. Take the insomnia itself to a clinician and ask about CBT-I.

Melatonin is an unusual supplement in that its mechanism is well understood and its niche is real — just much narrower than the shelf space it occupies. Stop expecting a sedative, start using it as what it is, a clock adjustment, and it becomes a genuinely useful tool.