Melatonin is a hormone the pineal gland releases in darkness, telling the rest of the body it is night. As a supplement it is one of the most used sleep products in the world, available on a supermarket shelf in the United States and only on prescription in most of Europe.

Two different things get called "melatonin for sleep". One is shifting the body clock: jet lag, shift work, a sleep phase that has drifted late. For that, timed melatonin has good evidence and a clear mechanism. The other is treating insomnia, where the clock is fine and sleep still will not come. For that, the evidence is modest at best and the better-designed reviews are skeptical.

Melatonin acts on receptors in the brain's master clock, the suprachiasmatic nucleus, and on receptors elsewhere. Taken in the evening it advances the clock; taken in the morning it delays it. That is a robust, dose-dependent effect and the basis for its use in circadian disorders.

Its direct sleep-promoting effect is separate and weaker. It lowers core body temperature and alertness modestly, which shortens the time to fall asleep by minutes. It is not a sedative in the way prescription hypnotics are, which is both its safety advantage and the reason its effect on insomnia is small.

It shortens sleep onset and lengthens sleep, slightly. A 2013 meta-analysis of 19 placebo-controlled trials in 1,683 people with primary sleep disorders found melatonin reduced the time to fall asleep by about 7 minutes, increased total sleep by about 8 minutes, and improved sleep quality by a small amount. Larger doses and longer trials showed larger effects on onset and duration. Those are the honest numbers: real, and small.

Chronic insomnia in adults: not clearly effective. A 2022 set of systematic reviews in Sleep Medicine Reviews covering 24 randomized trials found melatonin significantly improved sleep onset and total sleep time in children and adolescents, but in adults with chronic insomnia it did not significantly improve onset, total sleep or efficiency. The authors conclude it did not appear to be effective in adults.

Dose and timing may explain the disagreement. A 2024 dose-response meta-analysis of 26 trials found effects on onset and duration increased with dose up to about 4 mg and were larger when melatonin was taken around three hours before the intended bedtime, both at odds with the common practice of a smaller dose half an hour before bed. If that holds up, many null trials tested the wrong schedule.

Safety at higher doses. A 2022 systematic review of 79 trials using high doses found no detectable increase in serious adverse events or withdrawals, but a higher rate of drowsiness, headache and dizziness, and noted that more than a third of trials did not report adverse events at all.

At usual doses melatonin is well tolerated. Next-day drowsiness, headache and dizziness are the common complaints, and they rise with dose. Long-term safety data in adults are thin because trials are short, and there are almost none in children despite widespread use.

Two practical issues outweigh the pharmacology. First, melatonin suppresses alertness, so timing it before driving or operating machinery matters. Second, in countries where it is sold as a supplement, independent testing has repeatedly found products containing far more or far less than the label states, sometimes several-fold. A prescription product has a verified dose; a gummy may not.

Melatonin's status is split more sharply than almost any other compound here. In the United States it is a dietary supplement, sold without prescription and without regulatory review of dose or claims. In the United Kingdom, the European Union, Australia and much of the rest of the world it is a prescription medicine, licensed mainly for insomnia in people over 55 and for sleep problems in children with neurodevelopmental conditions.

The same molecule, then, is either a gummy for a toddler or a doctor's decision, depending on the country. For a reader, the relevant fact is that the prescription products have a verified content and the supplements often do not.