Melatonin benefits

Short answer

The evidence for melatonin is strongest in situations where the body clock is shifted (jet lag, delayed sleep phase); in primary insomnia the difference is measured in minutes; in shift work the evidence is of low certainty; for quality of life and symptoms in cancer patients, no benefit has been shown. Below, the population studied and the limit of the evidence are given for each claim; study doses are not recommendations. ^{[1] [5] [10]}

Does melatonin help insomnia?

Meta-analysis: Ferracioli-Oda and colleagues pooled 19 randomized trials (1,683 people; adults and children mixed) comparing melatonin with placebo in primary sleep disorders. Melatonin improved sleep onset time by a mean of 7.06 minutes (95% CI 4.37–9.75) and total sleep time by 8.25 minutes (1.74–14.75); overall sleep quality was also better than with placebo (standardized mean difference 0.22). The authors consider the effect modest and report that the effect does not appear to wane with continued use; they add that the absolute benefit is smaller than that of other sleep medicines. ^[1]

The guidelines differ:

• European Insomnia Guideline 2023: Prolonged-release melatonin can be used for up to 3 months in patients aged 55 and over (recommendation grade B). Immediate-release melatonin is not recommended for the treatment of insomnia (A). ^[2]
• AASM 2017 (pharmacologic treatment of chronic insomnia in adults): The guideline recommends against melatonin for difficulty falling or staying asleep (weak recommendation); the same guideline gives a use recommendation, also at the weak level, for some medicines such as zolpidem, zaleplon and ramelteon. ^[3]

Licensing data of the prolonged-release product at age 55 and over: In the EMA product information (2 mg prolonged-release), in a polysomnography study, sleep onset time after 3 weeks of treatment was 9 minutes shorter than with placebo; in an outpatient study, the rate of clinically meaningful improvement in sleep quality and morning alertness was 47% versus 27%. In a study extending to six months, the benefit was maintained up to month 3 but the primary analysis failed at month 6; at month 3, about 10% additional responders compared with placebo were seen. These are the studies used in the product’s licensing assessment and summarized in the product information; they are not an independent meta-analysis. ^[4]

Conclusion: the effect is small and depends on the population and the product; no finding supports the claim that “melatonin puts everyone to sleep.” The evidence on other sleep options such as magnesium is covered in the magnesium guide; there too the difference is small and the evidence is of low certainty.

Does melatonin work for jet lag?

The Cochrane review (Herxheimer and Petrie) evaluated 10 studies in adults traveling by air (passengers, crew, soldiers) in which oral melatonin was compared with placebo or another drug; all were comparisons with placebo, one also with zolpidem. Nine were of adequate quality for evaluation. In nine studies, melatonin taken near the target bedtime (22:00–24:00) reduced jet lag on flights crossing 5 or more time zones. Daily doses of 0.5–5 mg were found similarly effective; with 5 mg, sleep onset and sleep quality were better than with 0.5 mg, and a dose above 5 mg gave no additional benefit. The calculated NNT is 2. The benefit is larger as the number of time zones crossed increases and smaller on westward flights; if the timing is wrong (early in the day), there is a risk of sleepiness and delayed adaptation to local time. The review also notes that the 2 mg slow-release form was relatively ineffective. The authors write that the frequency of side effects is low but that there are case reports of harm in people with epilepsy and in those using warfarin. ^[5]

Limits: the review dates from 2002, the studies are small and were conducted only in adults. Timing details are on a separate page.

Detailed guideWhen is melatonin taken? Jet lag direction and timing→

Does melatonin work in delayed sleep phase?

In delayed sleep phase disorder (a condition in which sleep begins markedly later than desired), van Geijlswijk and colleagues pooled data from 91 adults in 5 studies and 226 children in 4 studies. Melatonin shifted endogenous melatonin onset earlier by a mean of 1.18 hours (95% CI 0.89–1.48) and sleep onset time by 0.67 hours (0.45–0.89); it shortened sleep onset time by 23.27 minutes (4.83–41.72). Wake time and total sleep time did not change significantly. ^[6]

This result is the most consistent example showing that melatonin works here not as a sedative but through an effect that shifts the clock. The meta-analysis is limited by the small size of the studies and covers only people with this diagnosis; this effect cannot be assumed in people without delayed sleep phase.

Does melatonin work in shift work?

The Cochrane review (Liira et al.) covered 15 placebo-controlled studies (718 participants) evaluating drug interventions in shift workers; nine were conducted with melatonin. Melatonin (1–10 mg) taken after the night shift may lengthen daytime sleep by a mean of 24 minutes (95% CI 9.8–38.9; 7 studies, 263 participants; low certainty) and night sleep by 17 minutes (3.71–30.22; 3 studies, 234 participants; low certainty). There was no difference from placebo in sleep onset time (mean difference 0.37 min; 5 studies, 74 participants) and no dose-response relationship was found. The authors conclude that there is low-certainty evidence that melatonin increases sleep duration after a shift, and that no benefit was seen in other measures of sleep quality. ^[7]

Does melatonin improve sleep in children?

Abdelgadir and colleagues examined 13 RCTs (682 children) comparing melatonin with placebo or another intervention in children under 18 with neurodevelopmental disorders. In the meta-analysis of nine studies (541 children), total sleep time increased by 48.26 minutes compared with placebo (95% CI 36.78–59.73); in 11 studies (581 children), sleep onset time shortened by 28.97 minutes; there was no difference in the frequency of night waking. No serious drug-related side effects were reported; the authors state that the quality of the evidence is limited because of heterogeneity and inconsistency between studies. ^[8]

This result is for children with neurodevelopmental disorders; it cannot be carried over to sleep problems in healthy children. Use, the safety discussion and the Türkiye context are on a separate page.

Detailed guideMelatonin in children: evidence, safety and physician decision→

Does melatonin reduce preoperative anxiety?

The Cochrane review (Madsen et al., search date July 10, 2020) covered 27 RCTs (2,319 participants) in adults aged 15–90 given melatonin before surgery; 24 compared with placebo and 11 with a benzodiazepine. No study was at low risk of bias in all domains. Compared with placebo, melatonin reduced preoperative anxiety by 11.69 mm on a 0–100 mm visual analog scale (95% CI 9.59–13.80; 18 studies, 1,264 people) (moderate certainty); measurement was made 50–120 minutes after administration. The reduction in anxiety measured immediately after surgery was much smaller (5.04 mm; low certainty) and its clinical importance was considered uncertain. Compared with midazolam and alprazolam, there was little or no difference (mean difference 0.78 mm; 7 studies, 409 people; moderate certainty). Nausea, sleepiness, dizziness and headache were reported as side effects, with no serious events; melatonin also caused sedation compared with placebo, and because 14 studies did not report side effects at all, it cannot be stated with certainty that melatonin is better tolerated than benzodiazepines. ^[9]

This result is for melatonin given as preoperative premedication; it contains no conclusion about anxiety in daily life or long-term use.

Does melatonin work in cancer?

The meta-analysis by Fan and colleagues covered 19 RCTs (2,101 patients) examining the effect of melatonin on quality of life and symptoms in cancer patients. Melatonin showed no significant effect on quality of life (standardized mean difference −0.01; 95% CI −0.14 to 0.11), sleep quality (−0.18; −0.62 to 0.26), fatigue (−0.34; −0.73 to 0.06), pain (−0.34; −0.70 to 0.02) or stomatitis severity (RR 0.78; 0.47–1.30). A reduction in the frequency of stomatitis was reported (RR 0.47; 0.26–0.88); this reduction was not present in people with head and neck cancer (RR 1.09; 0.92–1.29). A limited reduction in depression was reported in two subgroups: in those using it for longer than 14 days (standardized mean difference −0.14; −0.27 to −0.01) and in those who had surgery (−0.17; −0.32 to −0.03). The authors state that larger studies are needed because of the heterogeneity of treatment, duration and cancer type. ^[10]

This meta-analysis measures quality of life and symptoms; it does not evaluate the effect on tumor growth or survival.

Detailed guideMelatonin: overview, status in Türkiye and evidence summary→ Detailed guideMelatonin harms and side effects→ Detailed guideWhen is melatonin taken?→

This guide does not recommend personal doses; doses used in studies only describe study conditions.