Melatonin

What it is

Melatonin is a hormone made by the pineal gland in response to darkness, and also manufactured synthetically; most supplements are synthetic. Its legal status differs by country, and that difference matters for quality and dose. In the United States the FDA regulates melatonin as a dietary supplement sold over the counter. In the European Union prolonged-release melatonin is a prescription medicine: Circadin 2 mg for primary insomnia in people aged 55 and over, and Slenyto for insomnia in children with autism spectrum disorder, neurogenetic disorders or ADHD. Neither EU product can be obtained without a prescription.

What the research says

Melatonin is a body-clock signal, not a sedative, and the measured effects on sleep are small. A meta-analysis of 19 randomised placebo-controlled trials in 1,683 people found people fell asleep about 7 minutes sooner and slept about 8 minutes longer; when sleep was measured objectively rather than by diary, the extra sleep time vanished (0.33 minutes, p = 0.95). The same authors noted the effect is smaller than that of benzodiazepines or Z-drugs. It appears more useful where the problem is timing - jet lag, delayed sleep timing - and in children with autism or neurogenetic disorders, where the EU licensed a prescription product. It did not help sleep in Alzheimer's disease in a Cochrane review. Two guideline bodies decline to recommend it for chronic insomnia. The most substantial harms in the literature are not pharmacological but regulatory: a 530% rise in paediatric poison-centre calls in the US over a decade with two deaths, and analyses showing supplement content from 83% below to 478% above the label, with serotonin in a quarter of products.

Evidence grade: Limited evidence.

How it works

Drug class: Pineal indoleamine hormone and MT1/MT2 melatonin receptor agonist; a dietary supplement in the United States and a prescription prolonged-release medicine in the European Union

Melatonin is released by the pineal gland when it gets dark, rising in the evening and peaking in the middle of the night. It acts on receptors in specific brain regions involved in sleep and in the 24-hour clock, helping to bring on sleep and to set its timing. Taking it by mouth adds to that signal; prolonged-release tablets release it over several hours to mimic the natural overnight pattern. (Source 1)

What it is used for

• Approved in the European Union as the prescription medicine Circadin, for up to 13 weeks. In the three registration studies 32% of patients reported significant improvement at three weeks against 19% on placebo. This use is not an approved drug indication in the United States, where melatonin is sold as a supplement. Evidence: limited. (Source 1)
• Approved in the European Union as the prescription medicine Slenyto, used after sleep-routine measures have failed and reviewed at least every six months. The supporting long-term data are open-label and industry-sponsored. Not an approved drug indication in the United States. Evidence: limited. (Source 2)
• In the United States this is supplement use, not an approved drug indication: there is no FDA-approved melatonin drug product for insomnia. Trials show a small benefit - about 7 minutes faster to sleep - and the American Academy of Sleep Medicine and American College of Physicians both decline to recommend it for chronic insomnia. Evidence: limited. (Source 3)
• Supplement use in the United States; in the United Kingdom a melatonin product has been licensed for short-term jet lag treatment. The reviewed trials are small and favour melatonin for overall jet lag symptoms, with low-quality evidence for sleep quality. Evidence: limited. (Source 4)
• Supplement use. The American Academy of Sleep Medicine made a weak 2015 recommendation for timed melatonin, explicitly uncertain whether the benefits outweigh the harms. Evidence: limited. (Source 5)
• A Cochrane review found low-certainty evidence that melatonin up to 10 mg may have little or no effect on any major sleep outcome in Alzheimer's disease over eight to ten weeks. Evidence: not-supported. (Source 6)
• Two 2014 reviews found the studies generally small or inconclusive; the possible 24-minute gain in daytime sleep rested on low-quality evidence and no recommendation was made. Evidence: not-supported. (Source 7)

Interactions

• Fluvoxamine (and other CYP1A2-inhibiting antidepressants) (pharmacokinetic study): Fluvoxamine blocks the liver enzyme that clears melatonin, so the same melatonin dose produces many times the blood level. In five volunteers the area under the curve was 17 times higher. (Source 8)
• Caffeine (coffee, tea, energy drinks) (pharmacokinetic study): Caffeine competes for the same liver enzyme (CYP1A2), so evening caffeine raises night-time melatonin levels. A randomised crossover study in 12 people found 200 mg of caffeine - about two cups of coffee - increased the overnight melatonin area under the curve by 32%. (Source 9)
• Alcohol (clinical trial): Evening alcohol suppresses the body's own melatonin for the first half of the night, dose-dependently, in amounts people drink socially. (Source 10)
• Nifedipine and other calcium channel blockers used for blood pressure (clinical trial): In a randomised crossover trial in 47 people whose hypertension was controlled on nifedipine, adding evening melatonin raised 24-hour blood pressure by 6.5/4.9 mmHg and heart rate by 3.9 beats per minute. The authors concluded melatonin cannot be treated as merely a dietary supplement in people on these drugs. (Source 11)
• Blood-thinning medicines, and epilepsy medicines (theoretical): NCCIH states that people on blood thinners and people with epilepsy need medical supervision when taking melatonin. This is a cautionary position rather than a quantified pharmacokinetic finding. (Source 12)
• Serotonin found as a contaminant in melatonin supplements, and serotonergic drugs (pharmacokinetic study): This is a product-quality interaction rather than a drug one: serotonin was detected in about a quarter of melatonin supplements analysed, at 1 to 75 micrograms, and serotonin is itself a controlled substance used to treat neurological disorders. (Source 13)

Stopping it

• There is no described withdrawal syndrome. Adverse events in the randomised trials either resolved by themselves or immediately on stopping melatonin. (Source 14)
• Tolerance does not appear to develop: in meta-regression, longer trials and higher doses produced larger rather than smaller effects, which the authors contrast with benzodiazepines. (Source 15)
• The European regulators nevertheless build in stopping points rather than indefinite use: Circadin is limited to 13 weeks, and for the paediatric product treatment is to be reviewed at least every six months and continued only if the child is benefiting. (Source 2)
• Long-term safety of continuous use is simply not established: the systematic review of adverse events says the scarcity of long-term randomised trials limits any conclusion about extended use. (Source 16)

What goes wrong

Accidental and intentional melatonin ingestions by children reported to US poison centres rose more than five-fold in a decade, with thousands of hospitalisations and two deaths. (Source 17)

• Survey study, Moderate certainty.
• Size: 260,435 single-substance paediatric melatonin ingestions reported to the US National Poison Data System.
• Who: Children, adolescents and young adults aged 19 years or under, United States.
• How long: 1 January 2012 to 31 December 2021.
• Result: 260,435 ingestions, 94.3% unintentional, 82.8% asymptomatic. Of 27,795 who attended a health facility, 4,097 (14.7%) were hospitalised and 287 (1.0%) needed intensive care; 4,555 (1.6%) of all ingestions had moderate, major or fatal outcomes; five children required mechanical ventilation and two died, both aged under two years. Annual ingestions rose 530%, from 8,337 in 2012 to 52,563 in 2021.
• Funding: independent (CDC / academic paediatric authors)

Among 27,795 patients who received care at a health care facility, 19,892 (71.6%) were discharged, 4,097 (14.7%) were hospitalized, and 287 (1.0%) required intensive care.

The amount of melatonin in a supplement often does not match the label, and some products contain serotonin. (Source 13)

• Survey study, Low certainty.
• Size: 31 commercial supplements analysed (30 quantified for melatonin)
• Who: Melatonin natural health products and supplements sold in Ontario, Canada, across brands and formulations.
• How long: Single analysis, with lot-to-lot comparison.
• Result: Melatonin content ranged from 83% below to 478% above the labelled amount; lot-to-lot variation within one product reached 465%; more than 71% of supplements were outside a 10% margin of label claim; serotonin was found in eight supplements at 1 to 75 micrograms, and in 26% of products overall.
• Funding: not stated.

Melatonin content was found to range from -83% to +478% of the labelled content. Additionally, lot-to-lot variable within a particular product varied by as much as 465%.

In randomised trials the commonest side effects are daytime sleepiness and headache, at low single-figure percentages, but long-term safety data barely exist. (Source 14)

• Systematic review, Low certainty.
• Size: 37 randomised placebo-controlled trials.
• Who: Adults and children taking melatonin for primary or secondary sleep disorders; trials of pre-term infants excluded.
• How long: Up to 29 weeks, most four weeks or less.
• Result: Daytime sleepiness 1.66%, headache 0.74%, other sleep-related adverse events 0.74%, dizziness 0.74%, hypothermia 0.62%. Daily doses ranged from 0.15 mg to 12 mg. Only eight of 37 studies were judged good quality on Cochrane risk-of-bias criteria; 16 fair and 13 poor.
• Funding: not stated.

The most frequently reported AEs were daytime sleepiness (1.66%), headache (0.74%), other sleep-related AEs (0.74%), dizziness (0.74%) and hypothermia (0.62%).

In people whose blood pressure was already controlled on nifedipine, taking melatonin in the evening pushed blood pressure and heart rate back up. (Source 11)

• Randomized trial, Moderate certainty.
• Size: 47 mild to moderate essential hypertensive outpatients.
• Who: Patients with hypertension well controlled on nifedipine GITS 30 or 60 mg monotherapy for at least three months.
• How long: Four weeks per arm, double-blind crossover.
• Result: Melatonin 5 mg at 22.30 h raised 24-hour systolic blood pressure by 6.5 mmHg (P < 0.001), diastolic by 4.9 mmHg (P < 0.01) and heart rate by 3.9 beats per minute (P < 0.01) compared with placebo.
• Funding: not stated.

The evening administration of melatonin induced an increase of blood pressure and heart rate throughout the 24 h period (DeltaSBP = + 6.5 mmHg, P < 0.001; DeltaDBP = + 4.9 mmHg, P < 0.01; DeltaHR = + 3.9 beats min-1, P < 0.01).

What the evidence supports

Across 19 randomised placebo-controlled trials, melatonin shortened the time to fall asleep by about seven minutes and lengthened sleep by about eight minutes. (Source 3)

• Meta-analysis, Moderate certainty.
• Size: 19 studies involving 1,683 subjects.
• Who: Adults and children with primary sleep disorders in randomised placebo-controlled trials.
• How long: Varied by trial; most short-term.
• Result: Sleep latency weighted mean difference 7.06 minutes (95% CI 4.37 to 9.75), Z = 5.15, p<0.001; total sleep time 8.25 minutes (95% CI 1.74 to 14.75), p = 0.013; sleep quality standardised mean difference 0.22 (0.12 to 0.32), p<0.001. Heterogeneity for sleep latency Q = 31.9, I-squared = 56%.
• Funding: independent (academic authors; no industry sponsor stated)

Melatonin demonstrated significant efficacy in reducing sleep latency (weighted mean difference (WMD) = 7.06 minutes [95% CI 4.37 to 9.75]

As a licensed prescription medicine in the European Union, prolonged-release melatonin 2 mg improved sleep quality and next-day functioning in about a third of people aged over 55, against a fifth on placebo. (Source 18)

• Official position, Moderate certainty.
• Size: Three main studies with a total of 681 patients aged over 55.
• Who: Patients aged 55 years or over with primary insomnia.
• How long: Three weeks of treatment, with one additional study to six months.
• Result: 86 of 265 (32%) on Circadin reported significant symptom improvement after three weeks versus 51 of 272 (19%) on placebo, an absolute difference of about 13 percentage points, i.e. about eight people treated for one extra responder. The regulator's summary, not a trial report.
• Funding: not applicable (regulatory assessment of industry-sponsored trials)

When the results of all three studies were looked at together, 32% of the patients taking Circadin (86 out of 265) reported a significant improvement in symptoms after three weeks, compared with 19% of those taking placebo (51 out of 272).

For jet lag the reviewed trials favour melatonin, but the trials are small and the evidence for sleep quality is graded low quality. (Source 4)

• Official position, Low certainty.
• Size: 142 travellers across four studies for eastward flights, 234 in one sleep-quality study, 90 across two studies for westward flights.
• Who: Air travellers crossing multiple time zones.
• How long: Short-term, around travel.
• Result: Melatonin may be better than placebo for overall jet lag symptoms after eastward flights (four studies, 142 travellers) and westward flights (two studies, 90 travellers); low-quality evidence for improved sleep quality in a 234-traveller study. Based on medium-sized reviews from 2010 and 2014.
• Funding: not applicable (US government agency summary)

Four studies that included a total of 142 travelers showed that melatonin may be better than a placebo (an inactive substance) in reducing overall symptoms of jet lag after eastward flights.

Higher doses and longer trials produced larger, not smaller, effects, which argues against tolerance developing. (Source 15)

• Meta-analysis, Low certainty.
• Size: 19 studies, 1,683 subjects (meta-regression)
• Who: Adults and children with primary sleep disorders.
• How long: Trial durations varied; used as the moderator.
• Result: Meta-regression: longer trials gave greater effects on sleep latency (parameter estimate 0.53, 95% CI 0.21 to 0.86, p = 0.001) and total sleep time (1.60, 0.50 to 2.69, p = 0.004); higher doses also gave greater effects on total sleep time (7.25, 1.94 to 12.56, p = 0.007). The authors read this as no evidence of tolerance.
• Funding: independent (academic authors; no industry sponsor stated)

These findings suggest that there is no evidence of the development of tolerance with melatonin use.

What the evidence does not support

When sleep was measured objectively rather than by self-report, melatonin did not lengthen total sleep time at all. (Source 19)

• Meta-analysis, Moderate certainty.
• Size: 13 trials contributing total sleep time data within the 19-study meta-analysis.
• Who: Adults and children with primary sleep disorders.
• How long: Varied by trial.
• Result: Subjective total sleep time increased by 11.93 minutes (95% CI 4.06 to 19.81, p = 0.002) but objective total sleep time by 0.33 minutes (95% CI -11.19 to 11.87, Z = 0.05, p = 0.95). In the random-effects model the overall total sleep time gain of 8.48 minutes was not significant (p = 0.184)
• Funding: independent (academic authors; no industry sponsor stated)

Melatonin significantly increased total sleep time on subjective measures (WMD = 11.93 minutes [95% CI: 4.06 to 19.81], Z = 2.91, p = 0.002) but did not on objective measures (WMD = 0.33 minutes [95% CI: −11.19 to 11.87]

For sleep problems in people with Alzheimer's disease, melatonin up to 10 mg may do nothing at all. (Source 6)

• Systematic review, Low certainty.
• Size: Five melatonin trials, 222 participants, of which two contributed data to meta-analysis (n = 184)
• Who: People with mostly moderate-to-severe dementia due to Alzheimer's disease and an identified sleep disturbance.
• How long: Eight to ten weeks.
• Result: Total nocturnal sleep time mean difference 10.68 minutes (95% CI -16.22 to 37.59; 2 studies, n = 184); ratio of daytime to night-time sleep -0.13 (-0.29 to 0.03). No effect on sleep efficiency, time awake after sleep onset, number of awakenings or duration of sleep bouts. GRADE certainty: low. No serious adverse effects of melatonin reported.
• Funding: independent (Cochrane review)

We found low-certainty evidence that melatonin doses up to 10 mg may have little or no effect on any major sleep outcome over eight to 10 weeks in people with AD and sleep disturbances.

For shift workers the trials are too small or inconclusive to show a benefit. (Source 7)

• Official position, Very low certainty.
• Size: Seven studies with 263 participants, and eight studies with 300 participants (five overlapping)
• Who: Night-shift and rotating-shift workers.
• How long: Short-term.
• Result: People taking melatonin may sleep about 24 minutes longer during the daytime, but other aspects of sleep such as time to fall asleep may not change, and the evidence was considered low quality. The second review's six high-quality studies had inconclusive results and it made no recommendation.
• Funding: not applicable (US government agency summary of 2014 reviews)

The results suggested that (1) people taking melatonin may sleep about 24 minutes longer during the daytime, but (2) other aspects of sleep, such as time needed to fall asleep, may not change.

Two professional bodies say the evidence is not strong enough to recommend melatonin for chronic insomnia. (Source 12)

• Official position, Certainty not rated.
• Size: Not a study: two clinical practice guidelines.
• Who: Adults with chronic insomnia.
• How long: Not applicable.
• Result: The American Academy of Sleep Medicine (2017) and the American College of Physicians (2016) both decline to recommend melatonin for chronic insomnia; the ACP strongly recommends cognitive behavioural therapy for insomnia as initial treatment instead.
• Funding: not applicable (US government agency summary of guidelines)

According to practice guidelines from the American Academy of Sleep Medicine (2017) and the American College of Physicians (2016), there’s not enough strong evidence on the effectiveness or safety of melatonin supplementation for chronic insomnia to recommend its use.

Where the evidence is mixed

Melatonin's effect on falling asleep is smaller than that of prescription sleeping tablets. (Source 20)

• Meta-analysis, Moderate certainty.
• Size: 19 studies, 1,683 subjects, compared against published hypnotic meta-analyses.
• Who: Adults and children with primary sleep disorders.
• How long: Varied by trial.
• Result: Melatonin reduced sleep latency by slightly less than 7 minutes, against 10.0 to 19.6 minutes for benzodiazepines and 12.8 to 17 minutes for non-benzodiazepine hypnotics in an earlier meta-analysis. No head-to-head trials.
• Funding: independent (academic authors; no industry sponsor stated)

A previous meta-analysis demonstrated a significant benefit of benzodiazepines (10.0 to 19.6 minutes) and non-benzodiazepine sleep medicines (12.8 to 17 minutes) in reducing sleep-onset latency for primary sleep disorders ²¹.

In children with autism spectrum disorder and neurogenetic disorders, prescription prolonged-release melatonin improved sleep over a year, but the follow-up had no control group and was industry-sponsored. (Source 21)

• Randomized trial, Low certainty.
• Size: 95 children and adolescents who completed a 13-week double-blind trial (51 melatonin, 44 placebo) entered a 39-week open-label follow-up.
• Who: Children aged 2 to 17.5 years with autism spectrum disorder or neurogenetic disorders, with or without ADHD, who had not responded to behavioural intervention.
• How long: 52 weeks of continuous treatment in the originally randomised group.
• Result: After 52 weeks the originally randomised group slept 62.08 minutes longer (p = 0.007), fell asleep 48.6 minutes faster (p < 0.001), had 89.1 minutes longer uninterrupted sleep (p = 0.001) and 0.41 fewer awakenings per night (p = 0.001) than baseline. 55 of 72 completers (76%) gained at least one hour of sleep or latency improvement. Commonest treatment-related adverse events: fatigue 5.3%, mood swings 3.2%. All comparisons were against baseline, not placebo.
• Funding: industry-funded (sponsored by Neurim Pharmaceuticals; several authors are Neurim employees)

PedPRM was generally safe; most frequent treatment-related adverse events were fatigue (5.3%) and mood swings (3.2% of patients).

Adults who ate more melatonin-containing food were less likely to be obese or depressed, in a one-off survey that cannot show which came first. (Source 22)

• Survey study, Very low certainty.
• Size: 8,320 adults.
• Who: Adult participants in an open cohort of university graduates, mean age 35.9 years, 67.7% women.
• How long: Baseline (single time point)
• Result: Compared with the lowest fifth of dietary melatonin intake, obesity was less common in Q2 (IRR 0.81, 95% CI 0.69-0.96), Q3 (IRR 0.72, 0.60-0.86) and Q4 (IRR 0.79, 0.67-0.94), and depression in Q3 (IRR 0.79, 0.67-0.93) and Q4 (IRR 0.79, 0.67-0.94). Mean intake was 25554.7 ng a day. No association was found with hypertension, diabetes, metabolic syndrome, dyslipidaemia, sleep apnoea or sleep duration.
• Funding: independent (Brazilian public research agencies CAPES, FAPEMIG and CNPq)

Limit of this finding: This is dietary melatonin from ordinary food, in nanogram amounts, not a supplement dose, and everything was measured on the same day, so it is impossible to tell whether a melatonin-rich diet came before the lower weight and better mood or simply goes with other habits. Nothing here shows that eating more melatonin changes body weight or depression.

In multivariate models, obesity was inversely associated with melatonin intake in Q2 (IRR 0.81, 95%CI 0.69–0.96); Q3 (IRR 0.72, 95%CI 0.60–0.86) and Q4 (IRR 0.79, 95%CI 0.67–0.94). Depression was inversely associated in Q3 (IRR 0.79, 95%CI 0.67–0.93) and Q4 (IRR 0.79, 95%CI 0.67–0.94),

Where the research disagrees

Whether melatonin should be recommended for insomnia at all

• Ferracioli-Oda, Qawasmi and Bloch (PLoS ONE, 2013), meta-analysis of 19 randomised placebo-controlled trials, 1,683 subjects: "However, melatonin should be considered in clinical practice due to its benign side-effect profile, cost and limited evidence of habituation and tolerance." (Source 20)
• American Academy of Sleep Medicine (2017) and American College of Physicians (2016), as summarised by NCCIH, clinical practice guidelines: "there’s not enough strong evidence on the effectiveness or safety of melatonin supplementation for chronic insomnia to recommend its use" (Source 12)

Whether melatonin is a supplement or a medicine

• US Food and Drug Administration, as recorded by CDC authors, regulatory classification: "Melatonin is regulated by the Food and Drug Administration (FDA) as a dietary supplement." - sold over the counter in tablet, capsule, liquid and gummy form (Source 23)
• European Medicines Agency / Circadin, regulatory marketing authorisation: "The medicine can only be obtained with a prescription." (Source 1)
• Lusardi and colleagues (British Journal of Clinical Pharmacology, 2000), randomised double-blind crossover trial, 47 hypertensive patients: "As the pineal hormone might interfere with calcium channel blocker therapy, it cannot be considered simply a dietary supplement." (Source 11)

How much

• Reference intake: Melatonin is a hormone, not a nutrient, so no body sets an RDA or adequate intake for it. Where it is a prescription medicine the dose is set by the prescriber: the European Medicines Agency records one 2 mg prolonged-release tablet a day, one to two hours before bedtime and after food, for up to 13 weeks (Circadin medicine overview). (Source 1)
• Upper limit: No tolerable upper intake level or acceptable daily intake has been set for melatonin by any body we could reach. NCCIH states plainly that at doses above what the body makes there is not enough information to have a clear picture of overall safety, and that long-term safety data are lacking. (Source 12)
• Studied: The 19-trial meta-analysis pooled trials in which higher doses produced larger effects, and found no evidence of tolerance across the dose range studied. (Source 15)
• Studied: The adverse-event systematic review of 37 randomised trials covered daily doses from 0.15 mg to 12 mg, monitored for up to 29 weeks. (Source 14)
• Studied: The EU prescription product for adults over 55 is 2 mg prolonged-release once daily for up to 13 weeks. (Source 1)
• Studied: The paediatric prolonged-release trial used nightly 2, 5 or 10 mg in children aged 2 to 17.5 years with autism spectrum disorder or neurogenetic disorders. (Source 21)
• Studied: The nifedipine interaction trial used melatonin 5 mg at 22.30 h for four weeks. (Source 11)
• Studied: The fluvoxamine interaction study used a single 5 mg melatonin dose with and without 50 mg fluvoxamine. (Source 8)

A common belief, and what the research shows

The belief: Melatonin is a natural sleeping pill, so a bigger dose works better and gummies are a safe way to give it to children.

What the research shows: Melatonin is a timing signal, and the measured effect is small: "Subjects randomly assigned to melatonin fell asleep 7 minutes earlier on average than subjects receiving placebo", and on objective measures total sleep time did not increase at all. It is also not reliably dosed: "Melatonin content did not meet label within a 10% margin of the label claim in more than 71% of supplements and an additional 26% were found to contain serotonin." And for children the paediatric exposure data are substantial - "During 2012–2021, a total of 260,435 pediatric melatonin ingestions were reported to poison control centers", with 4,097 hospitalisations, five children ventilated and two deaths.

Questions and answers

What is it?

Melatonin is a hormone the pineal gland in the brain releases when it gets dark; it is also made synthetically and sold in tablets, capsules, liquids and gummies. Its regulatory status depends on where you are. In the United States the Food and Drug Administration regulates it as a dietary supplement, available over the counter. In the European Union the prolonged-release forms are prescription medicines. (Source 23)

What does it do in the body?

Melatonin is a timing signal rather than a sedative. Its level rises after dark and peaks in the middle of the night, and it acts on specific brain areas to help bring on sleep and to keep the 24-hour body clock aligned. Taking it shifts that signal earlier or later, which is why it is used for jet lag and delayed sleep timing as well as insomnia. (Source 1)

Is it good or bad for you?

It depends on the dose, the person and the product. In trials it modestly helps: about seven minutes faster to sleep, less than prescription sleeping tablets. It did not help sleep in Alzheimer's disease. The harms are mostly about access and quality rather than the molecule: 260,435 children were reported to US poison centres for melatonin ingestion in a decade, and supplement content ranged from 83% under to 478% over the label. (Source 20)

How do you get more of it?

In the United States it is bought over the counter as a supplement; in the European Union prolonged-release melatonin needs a prescription, at 2 mg once a day one to two hours before bed after food, for up to 13 weeks. Trials have used anything from 0.15 mg to 12 mg daily. The body also makes its own, and keeping evening light low allows that to happen. (Source 1)

If it is harmful, what reduces it?

Melatonin is not something that needs removing in healthy people, and the adverse events seen in trials mostly cleared up by themselves or as soon as the melatonin was stopped. The body's own melatonin production is suppressed by light: being exposed to light at night blocks it. Alcohol also lowers night-time melatonin, and melatonin is broken down by the liver enzyme CYP1A2. (Source 14)

Why might someone be low in it or missing it?

Night-time melatonin can be low because of age, light exposure or alcohol. Older people may produce less melatonin, which is the stated rationale for the prescription prolonged-release product in people over 55. Evening light at night blocks production. In a randomised crossover study, evening alcohol cut nocturnal melatonin by 41% at midnight. (Source 1)

Which whole foods contain it or feed it?

Melatonin occurs naturally in many plant foods, in amounts far smaller than a supplement dose. A 2026 analysis estimated the melatonin in 119 food items from a validated food frequency questionnaire and found coffee, beans and lentils, and rice were the main contributors to what people actually ate. Average intake among the 8320 adults surveyed was about 25,500 ng a day, roughly 25 micrograms, around a fortieth of a single 1 mg tablet. Higher dietary melatonin went with less obesity and less depression in that survey, but a one-off survey of this kind cannot establish cause. (Source 24)

What happens if you do not have it?

Low or badly-timed melatonin is linked to insomnia rather than to a deficiency disease; there is no melatonin deficiency syndrome in the literature we searched. The regulator's rationale for prescribing it to people over 55 is that reduced production leads to insomnia. In children with autism and neurogenetic disorders, where melatonin production can be abnormal, insomnia is common and treatment options are limited. (Source 2)

How can you test for it?

Melatonin can be measured in blood or saliva, and the usual research measure is the dim light melatonin onset (DLMO), the time the evening rise begins under dim light. It is the standard tool for assessing body-clock phase and assay method makes little difference to the result, but saliva levels are only about 30% of blood levels, sampling must be done under 30 lux, and the paper that assembled reference ranges found most patients diagnosed with delayed sleep-wake phase disorder actually had a DLMO inside the normal range. (Source 25)

References

1. European Medicines Agency. Circadin (melatonin) : EPAR - Medicine overview. 2026. Read the source
2. European Medicines Agency. Slenyto (melatonin) : EPAR - Medicine overview. 2026. Read the source
3. PLoS ONE. Meta-Analysis: Melatonin for the Treatment of Primary Sleep Disorders. 2013. PMID 23691095, DOI 10.1371/journal.pone.0063773. Read the source
4. National Center for Complementary and Integrative Health (NIH). Melatonin: What You Need To Know (jet lag and children's sleep disorders). 2026. Read the source
5. National Center for Complementary and Integrative Health (NIH). Melatonin: What You Need To Know. 2026. Read the source
6. Cochrane Database of Systematic Reviews. Pharmacotherapies for sleep disturbances in dementia.. 2020. PMID 33189083, DOI 10.1002/14651858.cd009178.pub4. Read the source
7. National Center for Complementary and Integrative Health (NIH). Melatonin: What You Need To Know (shift workers). 2026. Read the source
8. Clinical Pharmacology and Therapeutics. Increased bioavailability of oral melatonin after fluvoxamine coadministration.. 2000. PMID 10668847, DOI 10.1067/mcp.2000.104071. Read the source
9. Journal of Endocrinological Investigation. Caffeine raises the serum melatonin level in healthy subjects: an indication of melatonin metabolism by cytochrome P450(CYP)1A2.. 2003. PMID 12906366, DOI 10.1007/bf03345194. Read the source
10. The Journal of Clinical Endocrinology and Metabolism. Ethanol inhibits melatonin secretion in healthy volunteers in a dose-dependent randomized double blind cross-over study.. 1993. PMID 8370699, DOI 10.1210/jcem.77.3.8370699. Read the source
11. British Journal of Clinical Pharmacology. Cardiovascular effects of melatonin in hypertensive patients well controlled by nifedipine: a 24-hour study.. 2000. PMID 10792199, DOI 10.1046/j.1365-2125.2000.00195.x. Read the source
12. National Center for Complementary and Integrative Health (NIH). Melatonin: What You Need To Know (safety and guideline positions). 2026. Read the source
13. Journal of Clinical Sleep Medicine. Melatonin Natural Health Products and Supplements: Presence of Serotonin and Significant Variability of Melatonin Content.. 2017. PMID 27855744, DOI 10.5664/jcsm.6462. Read the source
14. CNS Drugs. Adverse Events Associated with Melatonin for the Treatment of Primary or Secondary Sleep Disorders: A Systematic Review.. 2019. PMID 31722088, DOI 10.1007/s40263-019-00680-w. Read the source
15. PLoS ONE. Meta-Analysis: Melatonin for the Treatment of Primary Sleep Disorders (tolerance). 2013. PMID 23691095, DOI 10.1371/journal.pone.0063773. Read the source
16. CNS Drugs. Adverse Events Associated with Melatonin for the Treatment of Primary or Secondary Sleep Disorders: A Systematic Review (conclusion).. 2019. PMID 31722088, DOI 10.1007/s40263-019-00680-w. Read the source
17. MMWR Morbidity and Mortality Weekly Report (CDC). Pediatric Melatonin Ingestions — United States, 2012–2021. 2022. PMID 35653284, DOI 10.15585/mmwr.mm7122a1. Read the source
18. European Medicines Agency. Circadin (melatonin) : EPAR - Medicine overview (how it was studied, and the benefit shown). 2026. Read the source
19. PLoS ONE. Meta-Analysis: Melatonin for the Treatment of Primary Sleep Disorders (total sleep time results). 2013. PMID 23691095, DOI 10.1371/journal.pone.0063773. Read the source
20. PLoS ONE. Meta-Analysis: Melatonin for the Treatment of Primary Sleep Disorders (discussion and comparison with hypnotics). 2013. PMID 23691095, DOI 10.1371/journal.pone.0063773. Read the source
21. Journal of Child and Adolescent Psychopharmacology. Long-Term Efficacy and Safety of Pediatric Prolonged-Release Melatonin for Insomnia in Children with Autism Spectrum Disorder. 2018. PMID 30132686, DOI 10.1089/cap.2018.0020. Read the source
22. Journal of Human Nutrition and Dietetics. The Content of Dietary Melatonin in 119 Food Items and Its Relationship With Chronic Diseases: Results of the CUME+ Study. 2026. PMID 41588582, DOI 10.1111/jhn.70193. Read the source
23. MMWR Morbidity and Mortality Weekly Report (CDC). Pediatric Melatonin Ingestions — United States, 2012–2021 (regulatory status and trend). 2022. PMID 35653284, DOI 10.15585/mmwr.mm7122a1. Read the source
24. Journal of Human Nutrition and Dietetics. The Content of Dietary Melatonin in 119 Food Items and Its Relationship With Chronic Diseases: Results of the CUME+ Study (Summary key points). 2026. PMID 41588582, DOI 10.1111/jhn.70193. Read the source
25. SLEEP. The dim light melatonin onset across ages, methodologies, and sex and its relationship with morningness/eveningness. 2023. PMID 36799668, DOI 10.1093/sleep/zsad033. Read the source