Melatonin for sleep: a circadian signal that gets sold as a sedative

What melatonin actually is (and is not)

Melatonin is a hormone your own body makes. It is produced by the pineal gland, a pea-sized structure deep in the brain, and it is released on a strict schedule: levels stay low through daylight, begin rising a couple of hours before your habitual bedtime, peak in the middle of the night, and fall away toward dawn. That rise is triggered by darkness and suppressed by light — which is why bright screens late at night blunt it. In plain terms, melatonin is the chemical announcement that the sun has gone down. It is the body’s darkness signal.

This is the crucial thing to hold onto, because it reframes everything that follows. A classic sleeping pill — a benzodiazepine, a “Z-drug” like zolpidem, an antihistamine — works by pushing on the brain’s sedation machinery. It dampens neural activity and produces drowsiness whether or not your body clock agrees. Melatonin does nothing of the sort. It does not bind the sedative receptors those drugs target. It carries information, not force. When you swallow a melatonin tablet, you are not sedating your brain; you are telling it, chemically, that night has arrived and the sleep window is open. Whether your brain acts on that message depends heavily on whether the message matches the rest of your circadian state.

That single mechanistic fact predicts the entire evidence base. If melatonin is a timing signal rather than a sedative, then it should be most powerful exactly where timing is the problem — a body clock stuck in the wrong time zone, drifted too late, or fighting a night shift — and comparatively weak where the problem is not timing but the sheer inability to fall or stay asleep in an otherwise normal schedule. That is precisely the pattern the research shows. Melatonin is not a failed sleeping pill. It is a successful circadian tool being asked, most of the time, to do a sleeping pill’s job.

The evidence: circadian wins, insomnia is modest

Split the question in two, because the literature does. There is melatonin for circadian misalignment, and there is melatonin for ordinary insomnia. They earn different grades.

Start with its home turf. For jet lag, the evidence is genuinely good. A Cochrane systematic review by Herxheimer and Petrie pooled ten randomized trials and concluded that melatonin, taken close to the target bedtime at the destination, is “remarkably effective” at preventing or reducing jet lag, with the benefit clearest when crossing five or more time zones and travelling east.³ Cochrane reviews are cautious documents; that is unusually strong language from them. The reason it works is exactly the mechanism above — jet lag is a mismatch between your internal clock and local time, and a well-timed darkness signal is a tool built for that job.

The same logic extends to delayed sleep phase — the “night owl” disorder where the body clock runs hours late and sleep will not arrive until the small hours — and to shift work. The American Academy of Sleep Medicine’s 2015 clinical practice guideline for circadian rhythm sleep-wake disorders reviewed this body of evidence and issued a recommendation for strategically timed melatonin in delayed sleep-wake phase disorder, and for melatonin to promote daytime sleep in shift workers.⁵ When a specialty guideline actually recommends a supplement for a specific indication, that is a meaningfully higher bar than the usual “may help” hedging. The signal melatonin pulls — realign the clock — is the one these disorders need.

Now the harder, more commercially important question: does melatonin help plain insomnia, the ordinary difficulty falling asleep on a normal schedule that sells most of the bottles? Here the honest answer is: yes, a little. The Ferracioli-Oda 2013 meta-analysis pooled 19 randomized trials in 1,683 people with primary sleep disorders and found melatonin significantly reduced the time to fall asleep — by an average of about seven minutes — while modestly increasing total sleep time and improving overall sleep quality.¹ An earlier and deliberately conservative meta-analysis by Buscemi and colleagues reached a similar quantitative conclusion: a statistically significant but small reduction in sleep-onset latency, on the order of a few minutes, which the authors judged unlikely to be clinically meaningful for most patients.² The Auld 2017 review in Sleep Medicine Reviews put it plainly — the evidence supports melatonin mainly for reducing sleep-onset latency and for circadian disorders, not as a broad hypnotic.⁴

Seven minutes is the number worth memorizing, because it is the number the marketing never mentions. It is a real effect — consistent across trials, statistically robust — and for some people, especially older adults with genuinely low nighttime melatonin, it is more than that.⁶ But it is not the effect of a sleeping pill. If you fall asleep in forty-five minutes tonight, melatonin might get you there in thirty-eight. That is help. It is not oblivion.

The fair reading of the whole table: melatonin is a strong tool pointed at the wrong problem most of the time. Aimed at a misaligned clock, it earns a guideline recommendation. Aimed at garden-variety insomnia, it buys you a handful of minutes. Both statements are true, and the confusion between them is the reason melatonin is simultaneously underrated by clinicians and overrated by consumers.

Melatonin does not push you into sleep. It tells your brain the sun has set — and how loudly that message lands depends entirely on whether the rest of your clock agrees.

Dose: why lower is often better

Here is where melatonin defies supplement intuition. With most things, more is more, or at least more is not worse. With melatonin, the physiological dose — the amount that roughly matches what your body would make on its own — is often as effective as, and sometimes better than, the megadose on the shelf.

The doses sold over the counter, typically 3, 5, or 10 mg, are pharmacological: they can push blood melatonin to many times the natural nighttime peak. The amount the body actually uses to signal night is far smaller — in the range of a few tenths of a milligram. Zhdanova and colleagues demonstrated this directly. In older adults with insomnia and low nocturnal melatonin, a physiological dose of about 0.3 mg restored normal nighttime blood levels and improved sleep efficiency, while a higher 3 mg dose was no better for the outcome that mattered and pushed melatonin to unnaturally high, sustained levels — the kind that can spill into the morning and leave you groggy.⁶ More drug, more residue, no more benefit.

Two problems compound at the high end. The first is next-day grogginess: melatonin has a real half-life, and a 10 mg dose can leave meaningful levels circulating well past your alarm, producing the “melatonin hangover” people describe. The second is more subtle — chronically flooding the system with supraphysiological melatonin may blunt the responsiveness of its own receptors, so the signal a big dose is trying to shout can actually land softer over time. The counterintuitive takeaway that runs through the sleep-medicine literature is that for circadian timing you generally want the smallest dose that does the job, not the biggest. This is one of the rare cases where the value-size bottle is the worse buy.

Timing: the phase response curve

If dose is counterintuitive, timing is the part almost nobody gets told about — and it can matter more than the dose. Because melatonin is a clock signal, when you take it does not just change how fast it acts. It changes the direction in which it moves your body clock.

This is described by what chronobiologists call the phase response curve (PRC): a map of how a given cue shifts the clock depending on the circadian time it is delivered. Burgess and colleagues built a detailed human PRC for melatonin, giving it at different clock times and measuring which way the internal clock moved.⁷ The finding, simplified: melatonin taken in the afternoon and early evening, before your natural melatonin rise, advances the clock — it pulls your whole rhythm earlier, which is what a night owl wanting to fall asleep sooner actually needs. Melatonin taken in the morning, on the other hand, delays the clock, pushing sleep later — the opposite of what most people want, and a genuine way to make a schedule worse.

The practical implication is large. Someone with delayed sleep phase who wants to shift earlier is generally served by a small dose taken several hours before target bedtime, not a big dose swallowed at lights-out. Take melatonin at the wrong circadian time and you can, in principle, shove your clock in the wrong direction. This is also why “melatonin didn’t work for me” is so common: the person took a hypnotic-sized dose at bedtime expecting sedation, when the tool’s real leverage was a small, earlier dose aimed at the clock. Burgess’s work also reinforced the dose point — a 0.5 mg dose produced phase shifts comparable to 3 mg, again arguing that for shifting the clock, low and well-timed beats high and late.⁷

The quality problem: what is really in the bottle

Everything above assumes the tablet contains the dose it claims. For over-the-counter melatonin in North America, that assumption is shaky, and this is not a fringe concern — it is one of the best-documented quality failures in the supplement aisle.

Erland and Saxena analyzed the actual melatonin content of a batch of commercial supplements and found the numbers alarming. Measured content ranged from 83% below to 478% above the amount stated on the label. Lot-to-lot variability within a single product reached as high as 465% — meaning two bottles of the same brand could differ several-fold. More than 71% of products failed to fall within a 10% margin of their label claim, and about a quarter of products also contained serotonin, a related compound that is not benign to dose blindly. Gummies, the format most likely to be handed to a child, were among the most variable.⁸

Put those two facts together — the right dose is often a few tenths of a milligram, and the bottle might contain five times its label — and you see why quality is not a footnote. A product labelled 5 mg that actually delivers 24 mg is not just wasteful; it guarantees the grogginess and supraphysiological flooding the low-dose research warns against. The defensible move is to buy from a manufacturer that submits to independent third-party verification (look for a USP, NSF, or equivalent seal), and to treat the printed dose as a starting hypothesis rather than a fact.

Grey areas: safety, tolerance, and the sedative myth

Melatonin is, for most healthy adults, safe in short-term use. The common side effects are mild and predictable: next-morning grogginess (usually a dose-and-timing problem), headache, dizziness, and vivid or unusually intense dreams, which some people find unsettling. It does not produce the classic dependence or the abuse potential of sedative-hypnotics, and physiological tolerance appears low. A genuine rebound — a night or two of worse-than-baseline sleep after stopping — is possible, particularly after high doses, but this is a milder phenomenon than the withdrawal seen with prescription sleep drugs.

The cautions that do matter cluster in specific groups. Because melatonin is a hormone with signalling roles beyond sleep, its use in children should be a clinician’s decision, not a default parenting tool — long-term effects on development are not fully characterized, and the dosing chaos above is most dangerous in small bodies. It is generally avoided in pregnancy and breastfeeding for lack of safety data. People with autoimmune conditions warrant caution because melatonin can modulate immune activity, and those with epilepsy or a seizure history should involve a clinician, as the evidence on seizure threshold is mixed. There are also real drug interactions — melatonin can add to the effect of sedatives, interact with anticoagulants and immunosuppressants, and its levels are strongly raised by some medications such as the antidepressant fluvoxamine. None of this makes melatonin dangerous for the average adult; it makes it a hormone that deserves the respect of a conversation with a professional when any of these apply.

Which brings the argument back to its center, and to the one claim in this article graded HYPE. The pervasive belief that melatonin is a “natural sleeping pill” that will knock you out is not merely an overstatement — it is a category error. Melatonin is not a hypnotic. It does not sedate. It signals. Selling it, or taking it, as a sedative sets up the disappointment that makes people think it “doesn’t work,” when in fact they aimed a precise circadian instrument at a problem it was never built to hit. As a researcher, I read melatonin as a genuinely excellent, under-used tool for the narrow set of problems it fits — and a genuinely oversold one everywhere else.

Open questions

The honest gaps are specific. First, long-term safety of nightly use over years, especially in children, is not well characterized — most trials run weeks to months, and melatonin’s hormonal reach means “probably fine” is not the same as “studied.” Second, the optimal dose and timing for ordinary insomnia (as opposed to circadian disorders) remain genuinely unsettled; the low-dose, PRC-guided logic is strongest for clock realignment and only partly transfers to someone with a normal schedule who simply cannot sleep.⁶⁷ Third, individual variability is large and poorly predicted — the same dose can do nothing for one person and cause a full “hangover” in another, and we cannot yet say why in advance. Fourth, the field still needs more large, modern trials of the physiological low doses; a great deal of the OTC market runs on doses the evidence does not actually endorse.¹ None of these gaps overturn the core picture; they mark its edges.

The verdict

Melatonin is a good tool with a bad reputation — bad in both directions. It is oversold as a nightly knockout drop and, partly because of that, dismissed by people who tried it as a sedative and got seven minutes. The accurate view sits between. For circadian problems — jet lag, delayed sleep phase, shift work — melatonin is genuinely effective and guideline-recommended, and that earns those claims a legitimate MODERATE grade.³⁵ For ordinary insomnia it delivers a small, real, meta-analytically confirmed benefit — help, not transformation.¹² The idea that it is a strong sedative is HYPE, and the assumption that the bottle contains its labelled dose is, on the published testing, WEAK.⁸

So how to use it well, if you and a clinician decide to. Treat it as a clock cue, not a sedative. Start low — the evidence favours a fraction of a milligram to 1 mg over the 5–10 mg megadoses. Mind the timing — for shifting a delayed clock earlier, small and several hours before target bedtime; for jet lag, timed to the destination’s night. Buy a third-party-verified product and distrust the number on any untested gummy. And keep the expectation honest: melatonin can realign a clock beautifully and nudge sleep-onset gently, but it will not overpower a mind that will not switch off, and it is no substitute for the foundations of good sleep. Judged as what it actually is — the body’s darkness signal, in a pill — melatonin is one of the more useful things on the shelf, provided you stop asking it to be a sleeping pill. That is the whole trick, and almost nobody is told it.

For the levers that do more of the sedative-style work melatonin is wrongly asked to do, our reads on glycine for sleep quality, apigenin, and magnesium glycinate sit next to this one. And because melatonin is a light-and-dark story at heart, the pieces on blue-light-blocking glasses and cooling mattresses cover the environmental side of the same clock.