Melatonin is one of the most-used sleep supplements in the world, and also one of the most mistimed, misdosed, and mislabeled. Most people take a small amount 30 minutes before bed, from a bottle that may not contain what its label claims. The trial data suggests the dose and timing most people use are off, one specific travel-related use case has unusually strong evidence behind it, and separate lab-testing evidence raises real questions about whether what's on the label matches what's actually in the capsule. All three threads point toward the same conclusion: melatonin's science is more settled than the products sold under its name.
The dose-response evidence
A 2024 systematic review and dose-response meta-analysis of double-blind randomised controlled trials in the Journal of Pineal Research — 26 trials published between 1987 and 2020, contributing 1,689 observations in insomnia patients and healthy volunteers — found melatonin's effect on sleep onset increases progressively with dose, peaking at 4mg/day: higher than the 1–3mg most consumer products contain.1 Just as notably, the same analysis found that advancing the timing of administration to roughly 3 hours before the target bedtime outperformed the far more common practice of taking it 30 minutes before bed (time between administration and the sleep episode was a significant predictor of sleep onset latency, β = −0.16, p = 0.023). Melatonin's job is to shift your circadian signal, not to sedate you on contact — timing it earlier lines up with how it actually works, which is the same logic behind the timing advice in our guide to what actually resets your sleep clock.
The timing effect makes sense once you separate melatonin's two roles. Your body already produces its own melatonin on a schedule, with levels normally starting to rise a few hours before habitual bedtime as light exposure fades — a process sleep researchers call dim-light melatonin onset. A supplemental dose taken right before lights-out arrives too late to do that signaling job; it's acting more like a sedative than a circadian cue. Taken hours earlier, it nudges the body's internal clock forward instead, which is closer to what the hormone does naturally. That distinction also explains the limit of what melatonin is studied for at all: its trial evidence is about getting to sleep, not about reshaping the balance of NREM and REM sleep once you're under.
Where the evidence is strongest: jet lag and shifted schedules
Away from nightly use, melatonin's best-established application is travel across time zones. A Cochrane systematic review that identified 10 randomised trials (nine of adequate quality to contribute) found melatonin "remarkably effective" at preventing or reducing jet lag, with a number needed to treat of 2 — meaning roughly one traveler benefited for every two who took it, for trips crossing five or more time zones, particularly flying east.4 That review found daily doses between 0.5mg and 5mg similarly effective, though people fell asleep faster and slept better after 5mg than 0.5mg, and doses above 5mg appeared to be no more effective at all — which lines up with the dose-response curve above flattening rather than climbing indefinitely.
One honest tension worth flagging: the Cochrane review found melatonin worked when taken close to the target bedtime at the destination (10pm to midnight), while the 2024 dose-response analysis favours taking it around three hours before the desired bedtime for ordinary sleep-onset problems. Those are different problems — resetting a clock you have just dragged across five time zones is not the same job as nudging a clock that is merely running late — and the two findings are best read as applying to their own use cases rather than as one overriding the other. The same logic likely extends to shift work, where circadian misalignment is the core problem rather than an occasional inconvenience, though the trial base there is thinner and more mixed than the jet lag evidence.
Is higher-dose melatonin safe?
A 2022 systematic review and meta-analysis in the same journal pooled 79 randomized controlled trials (3,861 participants) specifically examining the safety of higher doses (defined as 10mg or more) in adults over 30.2 The important caveat first: only four of those 79 trials met the review's pre-specified low risk-of-bias criteria for meta-analysis, so the pooled numbers rest on a very small subset. Within that subset, melatonin was associated with a real increase in mild adverse events — drowsiness, headache, and dizziness (rate ratio 1.40, 95% CI 1.15 to 1.69, p < 0.001) — but no detectable increase in serious adverse events (0.88, 0.52 to 1.50, p = 0.64) or withdrawals due to adverse events (0.93, 0.24 to 3.56, p = 0.92). The bigger finding may be the gap in the data itself: 29 of the 79 trials (37%) made no mention of the presence or absence of adverse events at all, and the review's own authors said their confidence limits were very wide due to the paucity of suitable data, calling explicitly for better reporting in future long-term trials.
That gap matters more once you notice who's actually buying melatonin. U.S. poison control centers have logged a sharp rise in pediatric melatonin exposures over the past decade, most involving unsupervised children getting into gummies left within reach — a pattern that's less about melatonin's inherent toxicity and more about how casually it's sold and stored as a candy-like product. It's a separate issue from dose-response efficacy, but it's part of why "just take more" isn't a responsible answer to the 4mg finding above without better packaging, dosing clarity, and quality control across the category.
The bottle doesn't always match the label
Even if you know the effective range sits around 4mg taken a few hours before bed, that's only useful information if the product you buy actually contains what its label says. It often doesn't. An analysis of 31 melatonin supplements published in the Journal of Clinical Sleep Medicine measured actual melatonin content against label claims and found it ranged from 83% less to 478% more than what the label declared.3 More than 71% of the products tested fell outside a 10% margin of their stated dose, and the same product tested across different lots varied by as much as 465% from batch to batch. The same analysis also detected serotonin — an unlabeled, pharmacologically active compound — in eight of the 31 products, at levels of 1 to 75 micrograms. If you want to know what does let you check a claim like this, our guide to what a certificate of analysis actually tells you walks through it.
| What was measured | What the testing found | Why it matters |
|---|---|---|
| Actual vs. labeled content, 31 products | Ranged from 83% under to 478% over the labeled amount | Two bottles labeled identically could deliver nearly 7-fold different doses |
| Products within 10% of label claim | Fewer than 29% met that margin | Landing near the 4mg effective dose from a label alone is mostly guesswork |
| Same product, different production lots | Up to 465% variation lot-to-lot | A trusted brand can't guarantee this bottle matches the last one you bought |
That study is from 2017, and dosage forms have shifted toward gummies since then, but the underlying problem — melatonin manufactured and tested with far less rigor than a regulated drug — hasn't gone away, and more recent independent testing of gummies has kept finding mismatches between label and contents. Combine that with the pediatric exposure pattern above and you get a category where the science on dose and timing is more settled than the science on whether any given bottle delivers that dose at all.
Where melatonin sits in the range today
The evidence for melatonin's core effect — helping people fall asleep faster, at the right dose and timing, and helping travelers reset a shifted clock — is genuinely solid. What's less settled is long-term safety data at the higher doses that actually show the strongest effect.
Melatonin is in two products in our range, and neither is dosed to match the 4mg figure above. Sleep Formula uses 2mg — within the 1–3mg range most consumer products use, but below the roughly 4mg where the dose-response data shows the effect on sleep onset peaking. We're not making a "dosed to the studied peak" claim for it: 2mg is a reasonable, well-tolerated starting dose, nothing more. Sleep Support goes the other way, at 10mg — above the peak rather than below it, which is not unsafe in itself but is not a "stronger is better" argument either, given that the curve flattens past 4mg and the long-term safety reporting at 10mg and up is thin. Both doses are printed on their labels. What is not disclosed, and cannot be, is the per-ingredient breakdown of Sleep Support's 905mg botanical blend: we asked our manufacturer directly in September 2026 and those individual amounts are not available to publish, so we are not going to describe that as a gap with a closing date. If you are choosing between the two, our head-to-head comparison lays out the trade-off, and the one thing not to do is take both the same night — that is 12mg combined, three times the studied peak, with no research behind that specific stack. If you'd rather avoid melatonin entirely, we've covered the non-melatonin options with real trial support separately.
Bottom line
If you use melatonin, the evidence points toward taking it several hours before bed rather than right before, and suggests the commonly sold low doses may be underdosed relative to what the trials actually tested. Its strongest, most consistent evidence base is arguably jet lag, not nightly insomnia. The safety data at higher doses is reassuring on serious harms but genuinely thin on long-term reporting — and separate testing shows you often can't be confident the bottle in front of you contains what its label says in the first place. And if the problem is chronic insomnia rather than the occasional bad night, no dose of melatonin is the first-line answer: CBT-I has substantially better evidence than any supplement, including this one. Worth knowing before treating any dose as risk-free, or any label as accurate by default.
This article is for informational purposes only and has not been evaluated by the FDA. It is not intended to diagnose, treat, cure, or prevent any disease. Speak with a healthcare provider before starting any new supplement, especially if you take medication or have an existing health condition.
References
- Cruz-Sanabria F, et al. Optimizing the Time and Dose of Melatonin as a Sleep-Promoting Drug: Systematic Review of RCTs and Dose-Response Meta-Analysis. Journal of Pineal Research, 2024. onlinelibrary.wiley.com
- Safety of higher doses of melatonin in adults: A systematic review and meta-analysis. Journal of Pineal Research, 2022. onlinelibrary.wiley.com
- Erland LA, Saxena PK. Melatonin Natural Health Products and Supplements: Presence of Serotonin and Significant Variability of Melatonin Content. Journal of Clinical Sleep Medicine, 2017. pubmed.ncbi.nlm.nih.gov
- Herxheimer A, Petrie KJ. Melatonin for the prevention and treatment of jet lag. Cochrane Database of Systematic Reviews, 2002. cochranelibrary.com
Related reading: The Science of Deep Sleep, Sleep Hygiene Tips vs Real Treatment: What the Evidence Hierarchy Actually Looks Like, CBT-I and Sleep Difficulties in Autism: What the Melatonin Research Shows.
Want the evidence behind every number on our labels, as we re-check it? Join the list — no spam, just it straight.