Melatonin for Sleep: What the Evidence Shows

Melatonin is the default answer for anyone who can’t fall asleep — a gummy on every nightstand, sold as the natural off-switch for a busy brain.

Here’s the honest version. Melatonin works, but modestly: across the trials it shaves only about 7 minutes off the time it takes to fall asleep. It is most useful for a specific job — resetting a body clock that’s in the wrong time zone or running late — and it is genuinely powerful for jet lag. But it is a chronobiotic (a clock-setter), not a sedative. It won’t knock out a racing mind, timing matters more than the milligrams on the label, and what’s in the bottle often isn’t what the label says.

Does melatonin actually help you fall asleep faster?

The most-cited answer comes from a 2013 meta-analysis in PLOS ONE (Ferracioli-Oda et al.) that pooled 19 randomized placebo-controlled trials, 1,683 subjects. Melatonin significantly reduced sleep-onset latency — but by a weighted mean of just 7.06 minutes (95% CI 4.37 to 9.75). It also increased total sleep time and improved overall sleep quality, and, usefully, the effect did not fade with continued use.

The authors’ own framing is the part the packaging skips: the effects are “modest,” and the absolute benefit is smaller than other pharmacological treatments for insomnia. What earns melatonin its place is not potency — it’s the benign side-effect profile. Seven minutes is real. It is also not the knockout most people expect from the dose they’re taking.

If you have chronic, lie-awake-for-hours insomnia, the specialists have a clear position. The American Academy of Sleep Medicine’s 2017 clinical practice guideline for the pharmacologic treatment of chronic insomnia (Sateia et al.) reviewed the evidence drug by drug and concluded: “We suggest that clinicians not use melatonin as a treatment for sleep onset or sleep maintenance insomnia.”

That’s a recommendation against — the same verdict the guideline gave valerian, tryptophan, and diphenhydramine (the antihistamine in most “PM” sleep aids). The reason is simply that the evidence for a meaningful effect on insomnia didn’t clear the bar. This is the single most important thing to understand about melatonin: the condition most people buy it for — racing-mind, can’t-switch-off insomnia — is the one it’s least supported for.

The distinction that changes everything: chronobiotic, not sedative

Melatonin isn’t a drug that makes you drowsy so much as a signal that tells your brain it’s night. Your pineal gland releases it as darkness falls; taking a pill mimics that “it’s bedtime” cue and can nudge your internal clock earlier. That’s why it excels at problems of timing rather than problems of arousal.

Where it shines is jet lag. A Cochrane systematic review (Herxheimer & Petrie) found melatonin “remarkably effective in preventing or reducing jet-lag” in 9 of 10 trials for flights crossing five or more time zones — with a number needed to treat of 2, an unusually strong result in sleep medicine. The benefit is larger the more time zones you cross and for eastward travel. The same logic applies to delayed sleep phase — the classic night-owl whose clock is simply set too late.

Who it likely won’t help: the person lying awake at 1 a.m. with a mind that won’t stop. That’s an arousal problem, not a clock problem, and a clock-setter is the wrong tool. For that pattern, the racing-mind arousal loop is the thing to address, and the first-line treatment with the strongest evidence is CBT-I, which outperforms sleeping pills long-term.

Timing and dose: why more milligrams backfire

Here’s where the supplement aisle gets it backwards. The Cochrane jet-lag review found daily doses between 0.5 and 5 mg were similarly effective, and doses above 5 mg were no more effective. Yet gummies routinely pack 5, 10, even 12 mg — far past the point of added benefit.

Timing is the real lever. Melatonin taken close to the target bedtime works; taken at the wrong time — early in the day — it “is liable to cause sleepiness and delay adaptation.” A clock-setter shifts your clock in whichever direction you dose it, so a mistimed dose can push your rhythm the wrong way and leave you groggy. Because the physiological nighttime signal is tiny, a low dose taken at the right time often does more than a mega-dose taken carelessly. More milligrams mostly buy you more next-morning grogginess, not more sleep.

What’s actually in the bottle

Then there’s the quality problem, which most listicles never mention. A 2017 analysis in the Journal of Clinical Sleep Medicine (Erland & Saxena) chemically tested 31 commercial melatonin supplements and found the actual melatonin content ranged from 83% below to 478% above the labeled dose. More than 71% of products missed their own label by more than 10%, and lot-to-lot variation within a single product reached 465%.

The unsettling finding: serotonin was detected in 8 of the supplements — a related, regulated compound — at 1 to 75 micrograms. So “5 mg” on the label can mean anywhere from a near-placebo to nearly six times that, plus an unlabeled compound you didn’t sign up for. It’s an argument for choosing pharmaceutical-grade or third-party-tested products, and for not assuming the number on the front is the number you’re taking.

The takeaway

Does melatonin work? Yes — modestly, and for the right job. It trims sleep onset by a handful of minutes, and it’s genuinely effective for jet lag and shifting a clock that’s set to the wrong time. It is a chronobiotic, not a sedative: use a low dose (0.5–1 mg is plenty for most), take it at a consistent target bedtime, and don’t expect it to quiet a racing mind. If your problem is nighttime arousal rather than timing, melatonin is the wrong tool — as is often the case with magnesium and the rest of the supplement tier, and even the gentler mechanism of glycine. For the system these all sit inside — actually winding a nervous system down — see our anxiety regulation and sleep restoration work. None of this is medical advice; if sleep problems are persistent, talk to a clinician before reaching for a bottle.