The win is circadian, not sedative. A 0.3 to 0.5 mg dose timed 5 to 7 hours before bed shifts your body clock earlier in a few days; the same tiny dose at the destination bedtime cuts a jet-lagged trip's fogged days from four to two. The catch is honesty about magnitude — for sleep onset alone, the effect is about five minutes, smaller than people expect, and the 5 to 10 mg gummies you'll find at the pharmacy are working against you, not for you.
Your brain has a clock. It sits behind your eyes, in a little knot of cells called the SCN, and it runs roughly a 24-hour cycle of "be alert" and "wind down." It's not metaphor — neurons in there literally fire faster during your subjective day and slower at night. Melatonin is the chemical signal that night has arrived. Your pineal gland releases it about two to three hours before you normally fall asleep, peaks it around 2 AM, and shuts it off by morning.
So a melatonin pill isn't telling your body to sleep. It's telling your body it's later than it thinks. That distinction is the whole article. When you take it matters more than how much you take.
The clock can be moved. Take melatonin in the late afternoon or early evening — before your own would normally rise — and your clock shifts earlier the next day. Take it in the deep middle of the night and you shift later. The biggest forward shift, about 90 minutes per night, comes from a dose 5 to 7 hours before your usual bedtime Burgess et al. 2010. That's the chronobiotic effect — the one the evidence base is built on.
The reason small doses are enough: the receptors that move the clock — called MT1 and MT2 — saturate at concentrations close to what your own body normally makes. A 0.3 mg pill puts blood levels at about two or three times your natural nighttime peak. A 3 mg pill puts them at 30 times. A 10 mg gummy goes higher still. None of that extra signal does more clock-shifting — the receptors are already maxed. What it does do is leave a tail of melatonin in your blood the next morning, when your body has decided it's day. That tail is the grogginess Zhdanova et al. 1995.
What it actually does — and doesn't
Two separate questions, two separate answers. The first: does it shift your body clock? Clearly yes, replicated for thirty years, endorsed in formal practice guidelines for four named circadian disorders AASM 2015. The second: does it knock you out at bedtime like a sleeping pill? Modestly, by a handful of minutes — far less than people expect.
Pooled across 19 controlled trials and nearly 1,700 people, melatonin shortens the time it takes to fall asleep by about seven minutes and adds about eight minutes to total sleep time Ferracioli-Oda et al. 2013. An earlier meta-analysis put the sleep-onset effect closer to four minutes Brzezinski et al. 2005. Both effects are statistically real and clinically small. If you're expecting the kind of hit a benzodiazepine delivers, you'll conclude melatonin doesn't work and you'll be empirically correct — for that purpose.
The story flips for circadian problems. For jet lag across five or more time zones, a Cochrane review of ten randomized trials found melatonin reliably reduced the lag, particularly for eastward travel, with the 0.5 mg dose nearly as effective as the 5 mg arm and doses above 5 mg conferring no extra benefit Herxheimer and Petrie 2002. For people with delayed sleep phase — chronically falling asleep at 3 AM and unable to shift earlier — a small dose at the right hour pulls onset earlier within days Mundey et al. 2005. And for totally blind adults, whose internal clock has no light to entrain it and drifts a little later every day, nightly melatonin is the standard of care; it can lock the clock to a 24-hour day Sack et al. 2000.
Older adults are a special case. Endogenous nighttime melatonin output drops steadily from your 20s; by your 60s and 70s it's a fraction of what it was. A nightly 0.3 mg dose in that group improves sleep continuity in a way it doesn't in healthy 25-year-olds Zhdanova et al. 2001. The prolonged-release prescription version used in Europe shows the same pattern over six months without rebound on stopping Wade et al. 2010.
The three things almost everyone gets wrong
It's not a sleeping pill. Most people who take melatonin take it at bedtime, expecting sedation, and get a few minutes of mild drowsiness. That's the whole hypnotic effect. What it actually does well is shift the clock — but that requires taking it hours before the bedtime you're aiming for, not at the bedtime you currently have.
More is not more. The receptors that move the clock saturate near 0.3 mg. Take 5 mg and you get the same clock shift plus a morning hangover. Take 10 mg and you get the same clock shift plus a worse one. The Burgess head-to-head proved it directly: the bigger dose did not move the clock further Burgess et al. 2010.
The dose on the bottle isn't the dose in the bottle. In the US, melatonin is a dietary supplement — exempt from the FDA pre-market review that prescription drugs face. When researchers actually measured the melatonin content of products on the shelf, they found a horror show: across 31 supplements, content ranged from 83% less than label to nearly five times more, with most products off by more than 10% Erland and Saxena 2017. A follow-up sampling 25 melatonin gummies in 2023 found most contained more than labeled — one had 347% of the stated dose, and several contained CBD that wasn't on the label at all Cohen et al. 2023. The "low dose" gummy you're carefully measuring out may not be low.
Living on the wrong time zone of your own life
Picture the person whose body thinks it's 1 AM when their alarm goes at 7. Not the once-a-decade insomnia attack — the chronic version. The Sunday-night dread that they can't fall asleep before 2. The morning where coffee is the difference between functioning and not. The mid-afternoon crash they explain away as a bad lunch. The Friday where a friend says "you look tired" and they're not sure when they last didn't.
That's the felt version of clock drift. It's most common in people who've gradually pushed their bedtime later — late-shift jobs, college years that never fully ended, screens in bed. It compounds. The clock isn't broken; it's set wrong. And the thing the body needs to reset it is a clean, dim, dark evening followed by a bright morning — which most people's evening no longer is, with overhead light and devices keeping the melatonin signal suppressed until they finally pass out.
For the frequent eastward traveler, it's the four days of the conference where the productive version of them never shows up. Meetings they can't track. Dinners where they're polite but not really there. A flight back to the same problem in reverse. It's not laziness or weakness — their suprachiasmatic nucleus is running on the time zone they left.
Without intervention, the standard alternative is one of two things: a sleeping pill (which sedates but doesn't fix the clock and brings its own dependence problem) or chronic poor sleep that the person stops noticing until something larger breaks. Misalignment is associated with worse mood, worse metabolism, worse cognitive performance — not from any single night, but from the years of being slightly off.
How to actually take it
There are three protocols, and they're not interchangeable. Pick the one that matches your problem.
The dosing problem is real and worth a paragraph. Almost no US bottle comes in 0.3 mg. The cleanest workaround is a 1 mg tablet split into thirds, or a liquid that lets you measure small drops. If you can find a USP-Verified or NSF-Certified product, that mostly solves the "is the dose actually the dose" question. Avoid gummies — both because the content varies most wildly and because gummies are why poison-control calls for children have multiplied.
When not to take it
Why "I tried it and it didn't work"
Almost always one of four things.
Wrong time. Bedtime is too late to start the chronobiotic signal — your own body is already releasing melatonin by then. Take the pill 5 to 7 hours earlier and the same dose does something completely different.
Wrong amount. A 5 mg dose is not "stronger" — it's worse. It buys you supraphysiological levels that linger into morning. People then chase the lingering grogginess with more caffeine, attribute the wakefulness problem to "needing a bigger dose," and escalate further. The fix is to go down.
Wrong problem. If you can't fall asleep because of an anxious mind or sleep apnea or restless legs, melatonin is the wrong tool. It won't quiet a racing head and it won't open an airway. The clock isn't the bottleneck.
Wrong product. If the bottle says 1 mg and actually contains 3.5 mg, you don't have a low-dose experiment — you have a high-dose one labeled wrong Erland and Saxena 2017.
What else moves the same lever
Light is the bigger zeitgeber, by far. Fifteen minutes of bright morning sun pulls the clock earlier more reliably than any pill, and a dim, dark evening lets your own melatonin actually rise on schedule. For most cases of "I can't fall asleep before 2 AM," the first move is the light environment, not the supplement. The pill helps once the light protocol is in place.
For sleep onset specifically — the actual "I want to fall asleep faster tonight" goal — cognitive behavioral therapy for insomnia (CBT-I) outperforms every pill over the long term and doesn't carry the dependence that benzodiazepines and Z-drugs do. The catch: it requires four to eight sessions and the work of actually doing it.
For non-24-hour circadian disorder in blind adults, there's also a prescription melatonin-receptor drug called tasimelteon that hits the same receptors with a longer half-life; it works, and it costs many thousands of dollars a year compared to a few dollars for OTC melatonin.
For pure short-term sedation — a wedding tomorrow, a flight tonight, a single hard night — the honest answer is that melatonin is not the right tool. The Z-drugs (zolpidem, eszopiclone) and benzodiazepines do that work better, with a clinician supervising the prescription. The tradeoff is dependence risk and morning impairment. What you don't want as a recurring fix is the over-the-counter antihistamine sleep aid — the diphenhydramine in Benadryl and Tylenol PM — whose anticholinergic load is the kind worth keeping off your nightly tally; for anything regular, low-dose melatonin is the cleaner default.
Where to get it, what to pay, what to avoid
In the United States, melatonin is on the supplement shelf at any drugstore. A bottle of 100 to 300 tablets costs five to fifteen dollars, lasts most users a year or more at the as-needed cadence this article recommends, and is functionally free at scale. Look for: USP-Verified or NSF-Certified labeling, tablets (not gummies), 0.5 or 1 mg strength so a third or a half gets you to the target dose, and a brand that's been tested by ConsumerLab or Labdoor.
In most of Europe, the UK, Canada, Australia, and Japan, melatonin is prescription-only or pharmacist-controlled — usually as a 2 mg prolonged-release tablet called Circadin, indicated for adults 55 and over with primary insomnia. The dose is higher than what this article recommends for jet lag, but the formulation releases slowly across the night, which matches the endogenous secretion curve and minimizes morning carryover.
One workflow worth knowing: if you travel a lot, keep a small pill cutter and a bottle of 1 mg tablets in your dopp kit. Splitting on the plane is fine. The pill doesn't need to be perfect — it's a tiny range of doses that all work.
What changes if you use it right
The chronobiotic version, three nights in: the eyes-open moment moves from 1:30 AM to closer to midnight, and stays there. Mornings stop feeling like the wrong end of a tunnel. The body becomes willing to be tired at the hour it's supposed to be tired, instead of staging a rebellion every Sunday night.
The jet-lag version, on day two of a trip to London: instead of four fogged days that you spend pretending to be present, you get one and a half. The dinner on day three is one you actually remember. The flight back doesn't feel like a fresh injury.
The older-adult version, after a few weeks: less time spent staring at the ceiling at 3 AM. Fewer night wakings that don't return to sleep. Mornings where the alertness shows up sooner than the second cup of coffee Wade et al. 2010.
None of these are dramatic. Be honest: you're not going to feel transformed. You'll feel like the version of you whose clock fits the day you're actually living. Which, if your clock has been off, is a lot.
Related things to look at
Morning sunlight exposure is the bigger lever for the same problem — read that one first. A dark bedroom at night is the third leg of the same protocol; the cleaner your dark, the better your own melatonin works. If the underlying issue is "I can't fall asleep" rather than "my clock is off," look at sleep apnea and CBT-I before any pill. And if you wake tired despite eight hours, look at upper airway resistance.
- — Low-dose melatonin is a clock-shifting tool, most useful when your timing is off, not your sleep length.
- — Reach for this instead of Benadryl or Tylenol PM — those nighttime antihistamines pile onto a dementia-linked drug burden.
- — Low-dose melatonin shifts your clock from the evening side; morning light does it from the other side, often more powerfully.
- — The right melatonin dose is a fraction of what sleep-supplement gummies pack — it resets the clock, not knocks you out.
- — The dose printed on a melatonin bottle is often wildly off; a USP or NSF seal is how you get the 0.3 mg you actually want.
- — Bright evening light blunts the melatonin a small dose tops up — true amber lenses before bed protect the same body clock.
- — One common side effect of melatonin is unusually vivid dreams. If yours have got intense, a bedside notebook turns that into something worth keeping.
- — Melatonin resets the clock from inside; dimming the evening lights does it from outside — same lever.
- — Before reaching for melatonin, kill the light leaks — darkness restores the hormone you're trying to supplement.
- — Jet lag is melatonin's best use — a tiny dose at the destination bedtime cuts the fogged days, so pack it for time-zone trips.
Substance + claimed effects
Melatonin is an indoleamine hormone secreted by the pineal gland in response to darkness; it is the body's primary chemical signal of biological night. Low-dose melatonin here means 0.1–0.5 mg taken orally, immediate-release — doses that approximate or slightly exceed the physiologic nighttime plasma peak (~60–80 pg/mL in healthy young adults). This entry covers the three established functions: (1) chronobiotic phase-shifting of the circadian clock (advancing or delaying biological night), (2) modest sleep-onset facilitation, especially in older adults with reduced endogenous output, and (3) jet-lag mitigation, particularly for eastward travel. Out-of-scope claims sometimes attached to melatonin (antioxidant longevity, antitumour effects, COVID adjunct, anti-aging skin) are not addressed — they are either preclinical, mechanism-only, or load-bearing on doses 10–100× the physiologic range. The entry also covers the OTC-dosing problem in the US, where typical retail products contain 1–10 mg with measured-content variability of −83% to +478% of label Erland and Saxena 2017.
Evidence by addressing question
Mechanism
Melatonin acts via two G-protein-coupled receptors, MT1 and MT2, concentrated in the suprachiasmatic nucleus (SCN) of the hypothalamus — the circadian master clock. MT1 binding attenuates SCN firing and is the proximate driver of the mild sleep-onset (soporific) effect; MT2 binding shifts the phase of SCN rhythms and is the proximate driver of the chronobiotic effect. Endogenous secretion rises after dim-light melatonin onset (DLMO), typically 2–3 hours before habitual bedtime, peaks at 2–4 AM, and falls to undetectable by morning. The phase-response curve (PRC) — a map of clock shifts produced by melatonin given at each point in the 24-hour cycle — is approximately 12 hours out of phase with the light PRC Lewy et al. 1998. Afternoon and early-evening doses (before DLMO) advance the clock (earlier sleep / earlier wake); doses in the late biological night delay it. The PRC's largest advance zone sits 6–8 hours before habitual sleep onset, with maximum advance of ~1.5 hours per dose Burgess et al. 2010.
Pharmacokinetics: oral bioavailability is low (~15%) and variable due to first-pass metabolism (CYP1A2). A 0.3 mg dose produces plasma peaks of ~150–200 pg/mL within 30–60 minutes — roughly 2–3× the physiologic nighttime peak. A 3 mg dose produces peaks ~10–60× physiologic, and elimination tail keeps levels supraphysiologic into morning. Half-life is short (~30–50 min for immediate-release), but the elimination tail of higher doses crosses the wake threshold and is the mechanistic basis of next-morning grogginess and the receptor-desensitization concern with chronic high doses Zhdanova et al. 1995.
Evidence
Phase-shifting (the strongest evidence base): Burgess et al. 2010 directly compared 0.5 mg versus 3.0 mg for three consecutive evenings in healthy adults under controlled conditions, measuring DLMO before and after. Both doses produced robust phase advances; the 0.5 mg dose was not inferior — phase advance averaged ~1.5 h when timed 5–7 hours before habitual sleep time Burgess et al. 2010. This finding undergirds the AASM 2015 conditional recommendation for melatonin in delayed sleep-wake phase disorder, advanced sleep-wake phase disorder, and non-24-hour sleep-wake disorder in blind adults AASM 2015.
Mundey et al. 2005 randomized delayed sleep phase patients to placebo, 0.3 mg, or 3 mg melatonin given 1.5 or 6.5 hours before DLMO. Both active doses advanced DLMO; the 0.3 mg dose performed comparably to 3 mg, with the 6.5-hour timing producing the largest advance Mundey et al. 2005.
Sleep-onset (weaker, dose-dependent): Zhdanova et al. 1995 administered 0.3 mg, 1.0 mg, and pharmacologic 10 mg melatonin to healthy young men in the early evening; all doses shortened sleep-onset latency by ~5–10 min vs placebo, but only the 0.3 mg dose produced plasma levels in the physiologic range and did so without next-day residual elevation Zhdanova et al. 1995. Zhdanova et al. 2001 found 0.3 mg nightly improved sleep efficiency in older insomniacs with low endogenous melatonin Zhdanova et al. 2001. The Brzezinski 2005 meta-analysis (17 trials, 284 subjects) found a pooled sleep-onset latency reduction of ~4 minutes — statistically real, clinically modest Brzezinski et al. 2005. Ferracioli-Oda 2013 (19 RCTs, 1683 subjects) found a 7-minute reduction in sleep-onset latency and an 8-minute increase in total sleep time, again statistically robust but small in magnitude Ferracioli-Oda et al. 2013.
Jet lag: Herxheimer & Petrie 2002 Cochrane review (10 RCTs) found melatonin 0.5–5 mg taken close to the target bedtime at the destination was effective in reducing jet lag from flights crossing five or more time zones, with eastward travel responding more strongly. Doses above 5 mg conferred no additional benefit. The 0.5 mg arms were nearly as effective as 5 mg on jet lag scores Herxheimer and Petrie 2002. AASM 2015 conditionally recommends melatonin for jet lag disorder AASM 2015.
Non-24 in blind subjects (strongest indication): Sack et al. 2000 entrained 6 of 7 totally blind free-running subjects with 10 mg nightly, later replicated at 0.5 mg with similar or better results — establishing melatonin as a true chronobiotic capable of substituting for the missing light entrainment signal Sack et al. 2000. Tasimelteon (Hetlioz) FDA-approved in 2014 for the same indication is a melatonin receptor agonist.
Prolonged-release 2 mg (Circadin, EU prescription) in adults ≥55: Wade et al. 2010 found improved sleep quality and morning alertness at 3 weeks and 6 months, sustained without rebound on discontinuation Wade et al. 2010. Note: this is not "low-dose" per the entry's frame, but it informs the dose-response shape — efficacy persists at modest doses when the formulation matches the endogenous secretion curve.
Protocol
Three distinct protocols by goal:
- Phase advance (earlier sleep):
0.3–0.5 mgimmediate-release, taken 5–7 hours before habitual bedtime. Repeat for 3–7 nights to achieve a sustained shift. Pair with morning light exposure on the advance side (light 6–8 hours after DLMO advances the clock) and dim-light protocol in the evening Burgess et al. 2010. - Jet lag (eastward, e.g., NYC → London):
0.3–0.5 mgtaken at the destination's local bedtime starting the night of arrival (or starting one day before for those who can plan), for 3–5 nights. Westward travel is generally easier and may not require melatonin; if used, take at the destination bedtime Herxheimer and Petrie 2002. - Sleep-onset (age-related low endogenous output):
0.3 mg30–60 minutes before desired bedtime. Useful primarily in adults ≥55 with documented or suspected reduced endogenous secretion Zhdanova et al. 2001.
Formulation: immediate-release tablets or sublingual liquid. Sustained-release products mimic the endogenous secretion curve better for sleep-maintenance use but are less common in low-dose retail products. Pill-splitting a 1 mg tablet is a workable workaround for the US-OTC dose problem.
Contraindications
Pregnancy and breastfeeding: data are limited; standard guidance is to avoid AASM 2015. Warfarin and other anticoagulants: case reports of altered INR; monitor or avoid. Immunosuppressants (post-transplant) and autoimmune disease: melatonin has Th1-skewing effects in vitro; clinical relevance is unclear but caution advised. Antiepileptic drugs: mixed effects on seizure threshold reported. Children: pediatric use has expanded rapidly but should be physician-supervised; see the poisoning data in §3b under practicalities. Antihypertensives and antidiabetics: melatonin lowers nocturnal blood pressure slightly and may interact with timing of glucose tolerance.
Misconceptions
Three core misconceptions:
- "Melatonin is a sleeping pill." It is a chronobiotic with mild hypnotic properties. Compared with benzodiazepine receptor agonists (zolpidem reduces sleep-onset latency by ~15–20 min), melatonin's 4–7 minute effect is small Ferracioli-Oda et al. 2013. People expecting sedation are disappointed and conclude it "doesn't work" — when its main value is the next several days of circadian alignment, not the next 30 minutes.
- "More is better." The PRC saturates near
0.3–0.5 mg;5 mgand10 mgdoses do not produce larger phase shifts and do produce supraphysiologic plasma levels that persist into morning Burgess et al. 2010, Zhdanova et al. 1995. - "OTC dose = label dose." Erland & Saxena 2017 analyzed 31 Canadian melatonin products and found measured content of −83% to +478% of label, with 71% deviating by more than 10%. Cohen et al. 2023 sampled 25 melatonin gummy products in the US: 22 contained more than labeled (one had 347%), and several contained CBD as an undeclared ingredient Erland and Saxena 2017, Cohen et al. 2023. The "low dose" in a labelled
1 mggummy may be3 mg.
Failure modes
The most common failure mode is wrong timing. Taken at bedtime, immediate-release melatonin barely catches the falling edge of the natural rise — minimal phase effect, minimal soporific effect. The same dose 5–6 hours earlier produces the largest advance. Second failure mode: dose escalation in response to no perceived effect — readers double, then quadruple, then take 10 mg, which is more sedating but less chronobiotic and produces dependence-like patterns where users feel they "need" it. Third: using melatonin to treat a sleep problem driven by sleep restriction or sleep apnea — the underlying condition continues unaddressed. Fourth: chronic nightly use in the absence of a phase-shift or low-endogenous-secretion indication; in young adults with normal endogenous secretion, nightly exogenous melatonin offers little benefit beyond the first dose and may slightly suppress endogenous output (small and reversible per available data).
Practicalities
Cost and access: in the US, melatonin is sold OTC as a dietary supplement; bottles of 100–300 immediate-release tablets cost $5–15. In most of Europe, the UK, Canada, Australia, Japan, melatonin is prescription-only (or pharmacist-controlled), typically as the prolonged-release 2 mg formulation Circadin. Quality control: USP-Verified or NSF-Certified products mitigate the labelling-variance problem; ConsumerLab and Labdoor test reports identify outliers. Pill-splitting a 1 mg tablet into thirds approximates 0.3 mg; liquid forms allow finer dosing.
Pediatric exposures: US poison center calls for pediatric melatonin ingestions rose 530% from 2012 to 2021, from 8,337 to 52,563 annual calls. 4,097 children were hospitalized over the decade; 287 required ICU admission; 5 required mechanical ventilation; 2 died. The driver is the combination of OTC availability, gummy formulations attractive to children, and the OTC-dosing problem (a child eating a "1 mg" gummy bottle ingests much more) Lelak et al. 2022.
Audience
Two populations with the largest signal: (1) older adults with reduced endogenous melatonin and primary insomnia — the Wade trial and Zhdanova 2001 data are strongest here Wade et al. 2010, Zhdanova et al. 2001; (2) totally blind individuals with non-24 sleep-wake disorder — the indication where chronic nightly melatonin is most clearly indicated Sack et al. 2000, AASM 2015. Shift workers and adolescents with delayed sleep phase disorder are also responsive but require careful timing protocols.
Alternatives
For circadian alignment without supplementation: bright morning light (2,500–10,000 lux for 15–30 min after wake) is the most potent zeitgeber and the first-line non-pharmacologic intervention. For sleep onset in low-endogenous populations: behavioral interventions (CBT-I) outperform pharmacotherapy long-term. Prescription melatonin receptor agonists (ramelteon, tasimelteon) target the same MT1/MT2 receptors with longer half-lives and FDA approval for specific indications; they cost 100–1000× more than OTC melatonin. Benzodiazepines and Z-drugs (zolpidem, eszopiclone) act on a different pathway (GABA-A) and produce true sedation but carry dependence and morning-hangover risks.
Stakes
Stakes for the entry's primary reader (someone with a misaligned clock — chronic late sleeper, frequent eastward traveler, shift worker, retiree with early-evening sleepiness): a misaligned clock is associated with metabolic dysfunction, mood disorders, and impaired daytime cognitive performance. Not from a single missed night — from the cumulative drift. Without the chronobiotic intervention, the standard alternative is hypnotic medication or chronic poor sleep.
Payoff
Realistic payoffs by indication: jet lag — 1–2 fewer fogged days at destination; phase-advance protocol — sleep onset 30–90 min earlier within 3–7 days of protocol initiation; older adult low endogenous — modest improvement in sleep efficiency, less night waking. These are small wins, not transformations.
The credibility range
Optimist case
Low-dose melatonin is the cleanest intervention in the catalogue for circadian phase shifting. The phase response curve is mapped, the mechanism is known to receptor level, the dose-response shows saturation near 0.3–0.5 mg, and the AASM endorses it for four named circadian disorders. The trials are small but consistent and have been replicated across multiple labs (Lewy/Sack at OHSU, Burgess/Eastman at Rush, Wurtman/Zhdanova at MIT). For blind non-24 patients it is the standard of care. For jet lag it is the only effective intervention beyond strategic light exposure. The 0.3 mg dose produces plasma levels approximating natural physiology and washes out by morning — a near-ideal pharmacologic profile. The intervention is cheap, has decades of post-marketing safety data, and lacks the dependence pattern of true hypnotics.
Skeptic case
The sleep-onset meta-analytic effect is 4–7 minutes — clinically trivial. Most readers buying melatonin OTC believe they are buying a sleeping pill and are disappointed; their disappointment is empirically justified. The OTC products they buy are not low-dose (3–10 mg typical) and contain unknown actual quantities, sometimes ten-fold higher than labelled. Long-term safety data beyond 6 months is thin, especially in children where exposure has 6×'d in a decade. The endogenous-suppression concern, while small in studied populations, has not been rigorously studied at the doses most people actually take. Nightly use in young adults with normal endogenous secretion has no evidence base. Industry-funded prolonged-release trials show modest effects in narrow populations; generalizing those to general use is unsupported.
The author's call
Low-dose melatonin (0.3–0.5 mg) is a real chronobiotic and a marginal hypnotic. The article should frame it as a circadian tool — for jet lag, phase advance, blind non-24, age-related low endogenous output — and explicitly debunk the "OTC sleeping pill at 5 mg" frame as both pharmacologically inferior and not what the evidence supports. Sleep score is 3 (clear, not dominant); evidence is 4 (multiple replicated RCTs and a major clinical guideline, but most trials are small and the effect-size on sleep-onset is modest). Controversy is modest (level 2): the chronobiotic story is settled; debate concentrates on the magnitude of the hypnotic effect and on the OTC-dose problem. Most informed sleep clinicians agree on these contours.
Stakeholder + incentive map
- Commercial: the US OTC supplement industry has strong incentive to sell
5 mgand10 mgproducts in gummy form — higher margin per unit, broader appeal, no prescription friction. This has shaped consumer perception of melatonin as a sleeping pill. - Professional: AASM and the AAP have both issued cautionary statements about pediatric use; sleep medicine specialists broadly endorse the low-dose chronobiotic frame but are not the loudest voice in the consumer market.
- Regulatory: EMA and most non-US regulators classify melatonin as a medicine, restricting OTC sale and standardizing dose. FDA classifies it as a dietary supplement under DSHEA 1994, exempting it from premarket review.
- Cultural: melatonin became one of the top supplements during the COVID era (2020–2022 sales doubled in the US); the gummy format and "natural sleep" positioning shape ongoing demand even as poison-control data accumulates.
Population variability
Age: endogenous melatonin secretion declines progressively from young adulthood; older adults (≥55) and especially those ≥65 show the largest sleep-quality response to supplementation. Sex: minor differences in pharmacokinetics; estrogen modestly slows CYP1A2-mediated metabolism (oral contraceptives prolong melatonin elimination by ~2×). Genetics: CYP1A2 polymorphisms (rapid vs slow metabolizers) explain part of inter-individual variability in plasma levels and morning carryover. Caffeine: heavy caffeine users metabolize melatonin faster via CYP1A2 induction. Chronotype: extreme evening types (DSPD phenotype) respond strongly to evening melatonin; morning types may not benefit and may even worsen if dose is mistimed. Smokers: faster clearance. Blind: when light entrainment is absent, melatonin becomes the dominant zeitgeber.
Knowledge gaps
Long-term safety beyond 6 months in adults remains under-characterized; the bulk of long-term data is from prolonged-release Circadin in older adults. Pediatric long-term safety is essentially unstudied despite massive uptake. Optimal dosing for shift workers across rotating schedules is unsettled. The interaction between exogenous melatonin and endogenous secretion at chronic supraphysiologic doses (the OTC 5–10 mg case) has not been studied in randomized trials of sufficient duration. The relationship between melatonin supplementation and pubertal timing in adolescents — flagged because melatonin influences gonadotropin pulsatility in animal studies — has not been settled in humans. Whether the OTC dosing variability identified by Erland 2017 and Cohen 2023 has changed since publication is unknown; FDA has not acted.
Scope kept tight to the chronobiotic frame. The brief named circadian phase shifting, sleep onset, jet lag, morning grogginess, and OTC dosing concerns — all covered. Out of scope: melatonin's antioxidant claims, putative anti-tumor effects, COVID-adjunct trials, and high-dose oncology research. Those rely on doses 10–100× the chronobiotic range and would dilute the entry's main message, which is that the saturating low dose is the right dose.
Sleep score sat between 3 and 4. Settled at 3. The chronobiotic effect on circadian alignment is real and named, but the sleep-onset latency reduction (4–7 min meta-analytic) is too modest to call a sleep transformation. A 4 would oversell.
Evidence at 4, not 5. Three independent labs replicate the chronobiotic effect, AASM endorses for four indications, Cochrane review supports jet-lag use. Held below 5 because trials are small (often n < 50) and the headline meta-analyses on sleep onset show small absolute effects. Not Cochrane-grade dominance.
Cadence as-needed, not daily. The strongest indications (jet lag, occasional phase reset) are situational. Chronic daily use applies to a minority — older adults with low endogenous output and blind adults with non-24 — and those audiences are flagged in the body. Calling cadence "daily" would mis-signal nightly use as the default.
Pediatric use deliberately handled in the warning callout, not given its own section. The MMWR poisoning data are striking enough that flagging them in contraindications is sufficient; a deeper treatment would warrant its own entry on pediatric melatonin, which should land in the backlog.
Contraindication tokens used: pregnancy, breastfeeding, blood-thinners, autoimmune. The closed vocabulary doesn't have a pediatric or immunosuppressant token; both are handled in prose inside the warning callout. The autoimmune token is conservative — the clinical signal is real but small.
Future-link candidates: morning-sunlight (the bigger zeitgeber), dark-bedroom, cbt-i, sleep-apnea, uars, jet-lag-protocol (could split off if this entry gets too long), pediatric-melatonin (separate entry candidate driven by the MMWR data).
Hard call on the "sleeping pill" debunking. The OTC market has trained readers to expect sedation. Leaning hard on the "not a sleeping pill" framing risks losing the reader who actually wants the placebo of taking something at bedtime. Decision: lean in anyway. The friend-test reader is better served by the correct mental model, even at the cost of feeling oversold to in past purchases.
Low-Dose Melatonin
OTC in US at ~$5–15 for a 100–300 tablet bottle (lasts a year+ at as-needed use). Prescription in EU/UK is also low-cost. Trivial.
One small pill at the right time. Effort is in remembering the timing window (5–7 h before bed for phase shift, destination bedtime for jet lag), not in the act itself.
Multiple replicated RCTs across independent labs (Lewy/Sack at OHSU, Burgess/Eastman at Rush, Wurtman/Zhdanova at MIT), a Cochrane review on jet lag (Herxheimer 2002), and a formal AASM 2015 clinical practice guideline. Two large meta-analyses (Brzezinski 2005, Ferracioli-Oda 2013) confirm small but consistent effects. Stops short of 5 because trials are small and the sleep-onset effect size is modest.
Chronobiotic action is well established: 0.5 mg advances DLMO ~1.5 h when timed 5–7 h before bedtime (Burgess 2010), with replicated effect at 0.3 mg (Mundey 2005). Sleep-onset reduction is modest (4–7 min meta-analytic) but real; clinical guideline endorses for four named circadian disorders (AASM 2015). Meaningful but not dominant — a clear named effect, not a sleep transformation.
Improvements come indirectly through better-aligned circadian rhythm and modest sleep-onset reduction. Brzezinski 2005 meta-analysis pooled ~4-min reduction in sleep latency; Ferracioli-Oda 2013 ~7 min. Real but small daily-life lift; larger when correcting a jet-lag or phase-disorder problem.
Energy effect is mediated through reduced jet-lag fog (Cochrane 2002: fewer impaired days at destination) and morning alertness in older adults on prolonged-release 2 mg (Wade 2010). No direct stimulant action. Score reflects real but bounded daily-vitality gain in the specific use cases.
Cognitive effects are indirect, via improved sleep continuity and circadian alignment. No direct nootropic mechanism. Marginal lift in jet-lagged or circadian-misaligned states only.
No direct mood pathway. Marginal lift only via improved sleep / reduced jet-lag dysphoria. Not the reason to use it.