• Print

Exogenous melatonin in the treatment of sleep disorders: efficacy and safety

AQuAS report.

Insomnia is the most common sleep disorder. It is defined as a dissatisfaction with the amount or quality of sleep, associated with difficulties in initiating or maintaining sleep or early waking with the inability to go back to sleep, which disturbs daytime well-being and the subjective ability and functioning of the individual. The prevalence of primary insomnia in the adult population ranges from 1 to 10%, reaching 25% among the elderly. In children with disabilities and developmental disorders, it can range from 25 to 86%.

Depending on the type of symptom that predominates, 4 types of insomnia are differentiated: 1) delay in falling asleep or sleep-onset insomnia; 2) disturbances in the continuity of sleep or maintenance insomnia; 3) early waking or early waking accompanied by 4) nonrestorative sleep. Depending on the duration, insomnia may be either transient (less than one week), short-term (1 to 3 weeks) or chronic (over 4 weeks). Although the etiological classification of insomnia has been questioned, primary insomnia is commonly considered an exclusion diagnosis, whereas secondary insomnia is deemed to be associated with other disorders.

The first and most important measures in the therapeutic approach to insomnia are those related to sleep hygiene, followed by psychological and pharmacotherapeutic interventions. The evaluation of pharmacological interventions was addressed in a guide published by the European Medicines Agency (EMA) in 2011. Several tools are used to typify insomnia symptoms including subjective measures like questionnaires, diaries, and symptom grading scales, as well as objective measures like polysomnography and actigraphy.

The measurement of melatonin levels, either from blood or saliva samples, less frequently used, allows determining the dim light melatonin onset (DLMO) and categorizing the different circadian sleep disorders. Recently, the American National Sleep Foundation has made recommendations regarding the indicators for the evaluation of sleep, including those based on sleep structure.

Download PDF

Available in Catalan.


Melatonin (N-acetyl-5-methoxytriptamine) is an endogenous neurohormone that is largely, but not exclusively, produced by the pineal gland. It operates through the MT1, MT2 and MT3 receptors of the suprachiasmatic nucleus of the anterior hypothalamus, our true biological clock. Melatonin can be obtained from natural sources but in most cases is synthetic. The natural secretion of melatonin declines with age.

Since its discovery in 1958, melatonin has been subject to an ever-increasing number of studies. The findings of its multiple physiological effects almost seem to endow melatonin with the properties of a "miracle molecule" due to its ubiquity and the sheer number of processes it’s involved in (antioxidant, neuroprotective, anticancer, immunomodulatory, etc.).

In 2007 the EMA authorized the use of melatonin (Circadin®) as medication for the treatment of primary insomnia in adults over 55, but melatonin is also used as a dietary supplement; in this case, it is subject to food security agencies in these cases. The Spanish Agency of Medicines and Medical Devices (AEMPS, as per the acronym in Spanish) treats melatonin as a drug when it is used in doses of ≥ 2 mg; below this, it is classified as a dietary supplement. Overall there have been no reports of serious melatonin side effects and the lethal dose (LD) 50 for mice by oral administration is 1250 mg/kg. However, long-term safety studies on the use of melatonin on the child-adolescent population and pregnant or breastfeeding women are scarce.

Melatonin regulation is rather contradictory and variable between countries. In the USA and Canada, it is only regulated as a dietary supplement, while in Europe it is regulated in two different ways.

Melatonin as medication

The EMA authorized the marketing of Circadin® for the treatment of primary insomnia in patients aged ≥ 55 with poor sleep quality. Based on the registry studies, the EMA concluded that melatonin is effective and has a small effect size on a small fraction of patients. Regarding its safety, no increase in the frequency of side effects is recorded. The most common side effects were headache, pharyngitis, lower back pain, and asthenia, with similar frequencies in the melatonin and placebo groups. Neither withdrawal nor rebound symptoms are observed when dosages are stopped.

Melatonin as a dietary supplement

Many products that contain melatonin are marketed in Spain. A recent study evaluating the purity and dosage in some of these dietary or nutritional supplements in a representative sample of products available in Spain and the USA found many pollutants that were structurally related to melatonin. All supplements acquired in Spain were correctly labeled with this potential health effect: "could alleviate the subjective sensation of jet-lag”. However, only 5 out of 8 products that claimed to "contribute to reducing the time before falling asleep" contained the amount of melatonin required for this effect (1 mg).


This report was written at the request of the Exceptional Pharmacy Commission of the Catalonia Health Service (Comissió de Farmàcia Excepcional of CatSalut) which has detected an increasing demand for reimbursement for products with an active melatonin ingredient related to various pathologies affecting sleep. For this reason, the Agency for Health Quality and Assessment of Catalonia (AQuAS, as per the acronym in Catalan) was asked to produce a report on the effectiveness and safety of this product or active ingredient.


A selective search has been carried out using literature reviews – either with quantitative syntheses (meta-analysis) or without them – of clinical studies that evaluate the effects of melatonin on sleep disorders. Research has also been carried out into clinical practice guides and recommendations from scientific societies and other prestigious entities. However, only the sections that deal with sleep disorders, particularly insomnia, were taken into account, whereas those related to other pathologies were excluded.

The use of melatonin as a drug has also been quantified, and the dispensation data have been analyzed. However, it is worth mentioning that, as melatonin is not a publicly funded drug, there are no accurate registries regarding its use. Data on the consumption of melatonin as a dietary supplement could not be gathered.


Meta-analysis and reviews of clinical studies

The effect of exogenous melatonin on primary sleep disorders

Buscemi’s meta-analysis includes 14 clinical trials with 279 participants. Although sleep latency (the main outcome) was statistically significant, it was not considered clinically relevant. Conversely, clinically relevance arises when delayed sleep phase syndrome is considered. In terms of sleep efficiency, the results favor melatonin, but not in a significant way and with high heterogeneity. Despite the popularity of melatonin, the authors describe melatonin as being of limited utility in primary insomnia. Brzezinsbski’s meta-analysis, which refers to 17 studies, supports the use of melatonin for sleep disorders but acknowledges the need to conduct more studies to better define the dose and the best time to administer it.

Delayed sleep phase disorder is examined in van Geijlswijk’s meta-analysis, which includes 9 trials involving adults and children. The results are favorable in terms of latency, but not in terms of total sleep time and are moreover dependent on whether the measure used is subjective or based on actigraphic recordings. Ferracioli-Oda’s meta-analysis is based on 19 studies involving a total of 1683 participants. In a statistically significant way, melatonin reduces sleep latency by 7.08 min and increases the total sleep time by 8.25 min. These improvements are found to be lower in objective measures and melatonin does not appear to generate tolerance or habituation.

The Cochrane Collaboration assessed the pharmacological therapies for sleep disorders arising from changes in work shifts. There are 9 melatonin trials that show that after a night shift, for doses of 1 to 10 mg, daytime sleep duration is increased (by 24 min) and so is nighttime sleep duration (by 17 min), but the evidence is considered to be of low quality. Winkler’s meta-analysis focuses on randomized controlled trials using polysomnography, but there is only one study (of 31) that involves 2 mg of prolonged-release melatonin, and the results are negative.

The objective of the Vural’s review was to determine the optimum dose for melatonin supplementation in people aged ≥ 55. Endogenous melatonin is shown to increase after exogenous melatonin was administered, even at low doses, and therefore the review recommends the lowest possible doses (0.3 mg to 1-2 mg), taken preferably 1 hour before going to sleep. The last meta-analysis of melatonin in primary sleep disorders examines two specific populations: those affected by delayed sleep phase syndrome and blind people. The results reveal improvements in sleep latency for people with delayed sleep phase syndrome, but not for the blind, who, nevertheless obtained improvements in sleep structure.

The effect of exogenous melatonin on secondary sleep disorders

The first review, conducted by Phillips in children and adolescents with neurodevelopmental disorders; suggests that, in the short term, melatonin can be effective in reducing sleep latency, but not in improving total sleep duration or night awakenings. Buscemi carried out another meta-analysis of melatonin in secondary sleep disorders in which sleep restriction disorders are also included. The results, contrary to other reviews, do not favor melatonin in jet-lag, and in cases of a favorable effect on secondary disorders, this effect is small and has no clinical significance. However, the short-term safety of the product is recognized.

Braam’s meta-analysis focuses on people with sleep disorders and intellectual disabilities and shows melatonin to have favorable results on sleep onset latency and the number of awakenings. The side effects are minor and similar in both groups. The authors suggest that for patients with intellectual disabilities circadian rhythm sleep disorders may predominate. The favorable effects of melatonin for sleep disorders in patients with Autism Spectrum Disorders (ASD) were first reported in the Guénolé’s review. However, the authors suggest that longer-term studies are needed to establish the safety and effectiveness of exogenous melatonin, which would need to be adjusted to the lowest effective doses. Rossignol’s review and meta-analysis also focuses on ASD. This study shows a statistically significant improvement in sleep duration and sleep latency, but not in the time spent awake at night. Side effects are minimal or nonexistent. Another review published in 2011 examines the pharmacological treatment of sleep disorders in children and adolescents with disabilities caused by developmental disorders. The authors consider melatonin to be the product that is supported by the best empirical evidence and that contains the best pharmacokinetic and pharmacodynamic properties. Melatonin is seen to be more effective for the treatment of sleep initiation and sleep maintenance disorders, and the side effects are mild and do not differ greatly between the experimental and control groups.

In one meta-analysis, Wright examines the role of melatonin in the discontinuation of benzodiazepines and the quality of sleep. In the 4 trials that allow discontinuation to be analyzed, melatonin did not have a significant effect, although heterogeneity is high. In 2015 Xu published a meta-analysis on the effect of melatonin on sleep disorders and cognition in patients with dementia, showing favorable results regarding the effectiveness and total sleep time, with no effect on cognition. Additionally, pharmacotherapy for sleep disorders in patients with dementia was assessed on a meta-analysis from the Cochrane Collaboration, published in 2016. The meta-analysis included 4 melatonin trials involving 222 participants. Compared to the placebo group, at doses of up to 10 mg for 8 to 10 weeks, no improvement is found in the total sleep time, nor in the ratio of day to night sleep, nor in cognition or quotidian activities. On the other hand, Zhang performs a meta-analysis on the use of exogenous melatonin in sleep disorders in people with neurodegenerative diseases. The results show that melatonin has a positive effect on the quality of sleep on Alzheimer's disease and Parkinson's when evaluated with a questionnaire, but not when objective measures are used. There are no significant differences in side effects.

Cuomo's meta-synthesis (review of reviews) on sleep disorders in children and adolescents, published in 2017, includes 3 reviews focused on melatonin. This is one of the studies with the best results, particularly for latency and duration of sleep, despite the fact that behavioral interventions and parenthood education programs also cover most of the disorder types.

Recommendations and clinical practice guidelines

Adults and the elderly

Although 5 years have passed since its publication, the Clinical Practice Guide for the Management of Patients with Insomnia in Primary Health Care (translated from the Spanish: Guía de Práctica Clínica para el manejo de pacientes con insomnio en atención primaria), carried out by the now defunct Agencia Laín Entralgo in Madrid, must be considered. In it, melatonin is described as a drug and approved for patients aged ≥ 55 with primary insomnia, and its superior effects on delayed sleep syndrome are noted. Also, the 2010 Alberta CPG Program 2010 discusses pharmacotherapy as an adjuvant treatment to cognitive and behavioral therapy, using the lowest dose for a short period of time. It is thought that in the case of melatonin, the evidence is variable.

There is a consensus statement, published in 2010 by the British Association for Psychopharmacology, which considers benzodiazepines and the so-called “Z drugs” to be effective and safe, and recommends the use of those with the shortest half-life. Regarding melatonin, the prolonged-release formulation is said to improve latency and sleep quality in patients over 55. In children, cognitive-behavioral therapy is useful and, subsequent to this, melatonin is said to reduce latency and modify the delayed sleep phase syndrome associated to disorders such as ASD and Attention-Deficit/Hyperactivity Disorder (ADHD). Most studies focused on children used supraphysiological and rapid-release doses. When it comes to circadian rhythm disorders, melatonin is effective for jet-lag, for delayed sleep phase syndrome where light therapy can also be effective, and in free-running disorders.

The Italian sleep study group on dementia issued recommendations in 2014 about how the initial strategy must not involve medication and ought to differentiate between sleep onset and sleep maintenance difficulties. Regarding pharmacological treatment, hypnotics such as benzodiazepines with short half-lives and Z drugs are examined. Non-benzodiazepine (Z) hypnotics and melatonin receptor agonists are the safest and most effective drugs currently available. Although melatonin has been seen to have favorable results in the elderly, the studies are small and of short duration.

The Good Clinical Practice Guide for the treatment of insomnia in geriatric patients published in 2015 by the Spanish Society of Geriatrics and Gerontology (Sociedad Española de Geriatría y Gerontología) mentions the underdiagnoses of insomnia in the elderly and that, nevertheless, drugs are overused in treatment. With respect to melatonin, the CPG charts the reduction of the endogenous secretion of melatonin with age and lists the different products available in Spain. It recalls melatonin’s classification as a drug and the fact that, unlike other hypnotics, melatonin is not restricted to severe insomnia and can be used for up to 13 weeks.

In 2015 the American Academy of Sleep Medicine updated their recommendations for the treatment of intrinsic (primary) disorders of the circadian sleep-wake rhythm. The recommendations regarding melatonin are weak in most situations (for children and adults, with or without depression, with or without dementia, and with or without neurological disorders). In the 2015 insomnia guide, the Spanish Society of Sleep (Sociedad Española del Sueño) along with the General Council of Medical Colleges of Spain (Consejo general de Colegios Médicos de España), describe melatonin as a 2mg prolonged-release drug advisable for adults aged 55 or over, while for children of up to 18 there were insufficient data. In 2016 the American College of Physicians published a guide for the management of chronic insomnia in adults in which it is stated that there is insufficient evidence for the use of melatonin for the general population and the elderly. The evidence on the comparative effectiveness of different pharmacological treatments is also insufficient.

The 2017 GPC of the American Academy of Sleep Medicine is about chronic adult insomnia. In it, melatonin is mentioned: it is said to have little effect, but that most patients prefer it over any other treatment.

Children and adolescents

A document published over 10 years ago which had the support of several scientific societies (relating to sleep, pediatrics, outpatient pediatrics and primary health care) advised against melatonin being considered a nutritional supplement or being placed outside of the control of pediatricians or sleep specialists. It stated that any product containing melatonin must be accompanied by an informative sheet explaining its uses, recommendations, doses and the limitations in knowledge about it. In 2011 the Agencia Laín Entralgo published a CPG on sleep disorders in children and adolescents and those in primary health care. Melatonin use is not recommended for children aged under 6. For those aged 6 to 12 the document reports favorable results where there is no response to educational and psychological interventions, although it is stated that its use must first be officially approved. In 2013 a summary of the evidence on melatonin (as a delayed-release drug) for sleep disorders in children and adolescents with ADHD was published by the National Institute for Health and Care Excellence (NICE). Evidence is limited and there are no high-quality studies, but melatonin is seen to improve the latency and duration of sleep. Discontinuing its administration as a supplement gives rise to relapse.

In 2014 an expert consensus statement was published in Anales de Pediatría (Barcelona) on the use of melatonin in children and adolescents over 6 with difficulty initiating sleep or with delayed sleep phase syndrome. The article, which is favorable to the use of melatonin, discusses dosage and time of administration, but where it is stated that duration has not been safely established for children and adolescents, a period of up to four weeks is discussed.

In 2015 another consensus statement from international experts was published on the role of melatonin in pediatric neurology. The authors describe the relevance of the determination of the DLMO in order to adjust the moment of melatonin administration, which would be between the hours of 19:00 and 23:00 for children aged 6 to 12 and between the hours of 20:00 and 00:00 for adolescents. According to the authors of these consensuses, there is no evidence to show that delayed-release melatonin has advantages over the immediate release melatonin that is usually present in food supplements.

The consumption of hypnotics and sedatives

As shown in Annex 6, from 2008 to 2016, the distribution of hypnotic and sedative prescriptions (ATC group N05C) has increased markedly: lormetazepam (derived from benzodiazepine) and zolpidem (Z drugs, related to benzodiazepines) account for the biggest consumption rates. The consumption of these two drugs has increased, particularly for the 18 to 64 age group and for older age groups (65 to 74 and over 75), for the latter remaining at a high level, while for the working-age population a recent (small) drop has taken place.

When it comes to melatonin as a drug, it must not be forgotten that although a prescription is required, it is not publicly funded and, therefore, no accurate records are available. The number of people recorded to have received it has gone from 68 in 2015 to 148 in 2016 for over 18s and from 21 to 26 for under 18s.


It seems paradoxical that a change of only 0.02 mg of melatonin makes the difference between being a dietary supplement and being prescription medication, even though it is not publicly funded. Depending on melatonin’s status as either a dietary supplement or a drug, the requirements for its marketing authorization from regulatory agencies (food safety and medication) in Europe are different. The variability in product quality, formulation, doses, and combinations of supplements is recognized and this partly explains the heterogeneity between studies observed in some of the meta-analyses.

As a drug, it is approved for adults aged 55 or over with primary insomnia, for a maximum of 13 weeks. The EMA describes a small effect size on a small fraction of the adult and elderly population and has called for reliable studies on the pediatric population. In cases of circadian sleep disorders, such as phase delay syndrome, the effect of melatonin has been shown to be greater, which could be more related to the adjustment of the biological clock than to the hypnotic effect.

When sleep hygiene and psychological interventions have failed, melatonin, which has shown a relatively favorable risk-benefit ratio compared to other hypnotics and sedatives, could be useful for children with disorders like ADS or ADHD. However, it seems advisable to not use a longer duration than that used for adults with the minimum effective dose and evaluating the treatment at 3 weeks. The evidence is not entirely conclusive and the ongoing studies on this pediatric population must be seen.

Randomized controlled trials of a pragmatic nature with a larger sample size and a longer follow-up period, as well as studies on the comparative effectiveness of hypnotics, are required.

Update:  28.02.2018