Sleep, Energy & Cognition
What Dose of Melatonin Do the Trials Actually Use?
The short answer
Across the trials pooled in recent dose-response meta-analyses, melatonin doses have ranged from roughly 1 mg up to 10 mg per night. A 2024 dose-response meta-analysis of 26 randomized controlled trials (1,689 observations, trials published 1987-2020) found melatonin's sleep-promoting effect grew with dose up to about 4 mg/day, a higher figure than the 2 mg dose commonly used in practice, and found timing — about 3 hours before bedtime — mattered as much as dose. A 2005 meta-analysis of 17 earlier trials in 284 subjects, pooling a range of doses, found melatonin cut sleep onset latency by an average of 4.0 minutes and extended total sleep time by 12.8 minutes. A Cochrane review of shift workers, where doses across the pooled trials ranged from 1 to 10 mg, explicitly found no dose-response relationship for sleep length after night shifts. This article reports the doses researchers tested — it does not recommend a dose for anyone to take.
Quick answers
- What is it?
- This reports the actual doses of melatonin that clinical trials have tested for sleep, ranging roughly from 1 mg to 10 mg per night.
- Why are people interested in it?
- Melatonin bottles on shelves vary wildly in dose, so people want to know what dose researchers actually found effective rather than guessing.
- What has actually been studied?
- A 2024 dose-response meta-analysis of 26 trials (1,689 observations), a 2025 pediatric dose-response meta-analysis of 21 studies, an earlier 2005 meta-analysis of 17 trials, and a 2014 Cochrane review of melatonin in shift workers.
- How strong is the evidence?
- Reasonably consistent across the two largest, newest dose-response analyses, which both found melatonin's sleep-promoting effect peaked around 4 mg per day — higher than the commonly used 2 mg — though a shift-worker population showed no dose-response relationship at all.
- What don't we know yet?
- Whether 4 mg is actually superior to 2 mg in a trial designed to test that directly, why shift workers didn't show a dose-response effect when other populations did, long-term effects of sustained higher-dose use, and how individual factors shift the ideal dose or timing.
- Where can I read the original research?
- Sources at the bottom of this page: Cruz-Sanabria F, Bruno S, Crippa A, et al., Journal of Pineal Research, 2024; and Bruno S, Cenerini G, Lo Giudice L, et al., Sleep Medicine Reviews, 2025.
The short answer: There is no single dose used across melatonin trials — researchers have tested doses spanning roughly 1 mg to 10 mg per night, and the two largest dose-response meta-analyses retrieved this session both found that melatonin's sleep-promoting effect increases with dose up to around 4 mg/day, beyond which added benefit levels off. That 4 mg figure is higher than the 2 mg dose the same researchers describe as most common in everyday practice. Timing appears to matter about as much as dose: giving melatonin roughly 3 hours before the desired bedtime, rather than the more typical 30 minutes before, was associated with the strongest reduction in how long it took people to fall asleep. Not every population shows this pattern — a Cochrane review of shift workers using doses from 1 to 10 mg found no dose-response relationship at all. The studies below report what dose was administered in each trial; none of this is a recommendation for what any individual should take.
What the research shows
The largest dose-response analysis to date
The most direct answer to "what dose do trials use" comes from a 2024 paper in the Journal of Pineal Research, which set out specifically to find the dose and timing of melatonin associated with the strongest sleep effect. The researchers systematically reviewed double-blind, randomized, placebo-controlled trials in people with insomnia and in healthy volunteers, and combined 26 trials published between 1987 and 2020, covering 1,689 individual observations. Using dose-response and meta-regression modeling, they found that melatonin's effect on sleep onset latency and total sleep time grew steadily as dose increased, reaching its maximum benefit at around 4 mg per day, with little additional gain beyond that. Two other variables predicted outcomes independently of dose: the time between administration and the sleep episode (earlier administration — about 3 hours before desired bedtime — was linked to a bigger reduction in sleep onset latency), and whether participants had diagnosed insomnia versus being healthy volunteers. The authors explicitly contrasted their finding with "the exogenous melatonin schedule most used in clinical practice" — 2 mg taken 30 minutes before the desired bedtime — and suggested that schedule might not be optimized for effect, pending confirmation in larger trials designed to test dose and timing directly rather than pooled after the fact.
A related 2025 meta-analysis in Sleep Medicine Reviews, focused specifically on children and adolescents with neurological and psychiatric conditions, reached a similar conclusion from a different set of 21 studies: melatonin's sleep-promoting effect reached its maximum somewhere between 2 mg and 4 mg per day, with earlier administration relative to bedtime again linked to shorter sleep onset latency, and longer treatment duration linked to greater gains in sleep efficiency and total sleep time.
What earlier, smaller pooled analyses found
Before those larger dose-response models existed, the reference meta-analysis in this field was a 2005 paper in Sleep Medicine Reviews that combined 17 trials — 284 subjects total — published between 1980 and 2003. Rather than modeling dose-response directly, this analysis pooled outcomes across the range of doses used in the underlying trials and found melatonin treatment reduced sleep onset latency by an average of 4.0 minutes (95% CI 2.5-5.4), increased sleep efficiency by 2.2 percentage points, and increased total sleep duration by 12.8 minutes, compared to placebo. Restricting the analysis to the 15 studies of healthy subjects or people without other relevant medical conditions produced similar numbers: a 3.9-minute reduction in sleep onset latency, a 3.1-point gain in sleep efficiency, and a 13.7-minute increase in total sleep duration. Because this analysis pooled trials using different doses rather than modeling dose-response, it can't tell you which specific dose drove the effect — only that, averaged across whatever doses were used, the effect was real but modest.
A population where dose didn't seem to matter
Not every context shows a clean dose-response relationship. A 2014 Cochrane systematic review examined pharmacological treatments for sleep problems in shift workers, and identified 9 placebo-controlled trials of melatonin among its 15 included studies, with doses across those trials ranging from 1 mg to 10 mg taken after the night shift. Pooling seven of those trials (263 participants), melatonin increased daytime sleep length by 24 minutes on average (95% CI 9.8 to 38.9) compared to placebo, and a separate pooling of three trials (234 participants) found a 17-minute increase in subsequent night-time sleep. But the reviewers stated explicitly: "We did not find a dose-response effect." Melatonin in this population also failed to meaningfully change sleep onset latency compared to placebo (mean difference 0.37 minutes). The Cochrane authors rated the underlying evidence as low quality throughout, reflecting small trial sizes and inconsistent methods across the shift-work literature.
Reading these numbers together
The dose range actually tested across all of these trials — roughly 1 mg to 10 mg — is wide, and the "optimal" dose identified by the two newer dose-response meta-analyses (around 2-4 mg) sits within the lower half of that range, not at the extremes. Both of the larger, more methodologically ambitious dose-response analyses agree on a similar peak-effect window, which gives that estimate more weight than any single trial. But it's worth noting these are dose-response models built from published trials that used a handful of different, fixed doses — not trials that randomized participants to a full continuum of doses within one study — so the shape of the curve between data points is estimated rather than directly observed.
What we don't know yet
Whether 4 mg is actually superior to 2 mg in a trial designed to test that comparison directly. The dose-response findings come from meta-regression across trials that used different fixed doses, not from a single trial that randomized participants between the two doses head-to-head.
Why shift workers didn't show a dose-response relationship when general insomnia and pediatric populations did. The Cochrane review flagged the absence of a dose-response effect in shift workers but didn't explain why that population behaved differently from the trials analyzed elsewhere.
Long-term effects of sustained higher-dose melatonin use. The retrieved meta-analyses pooled trials that were largely short-term; none of the sources here established what happens with months or years of continued use at any dose.
How individual factors — age, baseline melatonin production, or specific sleep disorders — shift the ideal dose or timing. The pooled meta-analyses modeled insomnia status and time-of-day as predictors, but retrieved data didn't break results down by other individual patient characteristics.
Whether the same dose-response pattern holds outside the insomnia and pediatric neuropsychiatric populations studied here — for example, in jet lag, delayed sleep phase disorder, or otherwise healthy adults without diagnosed sleep complaints.
Sources
- Cruz-Sanabria F, Bruno S, Crippa A, Frumento P, Scarselli M, Skene DJ, Faraguna U. Optimizing the Time and Dose of Melatonin as a Sleep-Promoting Drug: A Systematic Review of Randomized Controlled Trials and Dose-Response Meta-Analysis. Journal of Pineal Research, 2024;76(5):e12985. DOI
- Bruno S, Cenerini G, Lo Giudice L, et al. Optimizing timing and dose of exogenous melatonin administration in neuropsychiatric pediatric populations: a meta-analysis on sleep outcomes. Sleep Medicine Reviews, 2025;84:102158. DOI
- Brzezinski A, Vangel MG, Wurtman RJ, Norrie G, Zhdanova I, Ben-Shushan A, Ford I. Effects of exogenous melatonin on sleep: a meta-analysis. Sleep Medicine Reviews, 2005;9(1):41-50. DOI
- Liira J, Verbeek JH, Costa G, Driscoll TR, Sallinen M, Isotalo LK, Ruotsalainen JH. Pharmacological interventions for sleepiness and sleep disturbances caused by shift work. Cochrane Database of Systematic Reviews, 2014;2014(8):CD009776. DOI
Article records maintained via PubMed. Peptide Proof reports what studies found. It does not provide medical advice, dosing guidance, or treatment recommendations.
Topic: Sleep & Recovery
Sources
- Optimizing the Time and Dose of Melatonin as a Sleep-Promoting Drug: A Systematic Review of Randomized Controlled Trials and Dose-Response Meta-Analysis
- Journal of Pineal Research, 2024 · n=1689 · DOI
- Optimizing timing and dose of exogenous melatonin administration in neuropsychiatric pediatric populations: a meta-analysis on sleep outcomes
- Sleep Medicine Reviews, 2025 · DOI
- Effects of exogenous melatonin on sleep: a meta-analysis
- Sleep Medicine Reviews, 2005 · n=284 · DOI
- Pharmacological interventions for sleepiness and sleep disturbances caused by shift work
- Cochrane Database of Systematic Reviews, 2014 · n=718 · DOI
Get the evidence, not the hype.
New research briefs, plainly written and fully sourced. No dosing advice, ever.