skip to Main Content

Does this sound like
you, or someone you know?

An 86-year-old female patient came in with her daughter. She
had Chronic Obstructive Pulmonary Disease (COPD), depression, anxiety, and was
brought in by her daughter because she has had trouble sleeping, mostly with
falling asleep, and less so with staying asleep. She had tried mirtazapine and
trazodone (other sleep medications) without success. Her daughter asked, “Can mom try melatonin?” and I
wondered: Is melatonin more effective than placebo as a treatment of insomnia?

I searched the terms “insomnia melatonin” on PubMed Clinical
Queries looking specifically for systematic reviews, which usually provide an
exhaustive summary of current literature and offer higher quality evidence than
individual journal articles. I found a review article summarizing results from
19 randomized-controlled trials with a total sample of 1683 participants. The
review looked at 3 outcomes: sleep latency (the amount of time it takes to fall
asleep), total sleep time, and sleep quality. For each of the outcome, it explored 3 of its components – objective
(actual amount of time), subjective (patients’ feeling of how long it takes),
and overall (both subjective and objective combined). Individual patient data
was used for analysis, and there was no publication bias on Egger’s test on any
of the outcome measures.

The results were consistent across the 19 studies included
in the review, and shown below in a table format. Of note, for sleep quality,
the review reported the findings as standardized mean difference, because each
study used different scales to measure sleep quality. This makes it difficult
to interpret what the findings mean to patients.

In summary, it seems that melatonin helps patients feel that
they fall asleep faster, sleep longer, and get better quality sleep, although
in reality the actual effects are not as great as what patients feel. Also,
although statistically significant, these improvements may not be clinically
significant for every patient. For example, falling asleep faster by 7.06
minutes might not mean much for many patients, especially when we compare these
effects to those of other sleep medications, such as non-benzodiazepine sleep
aides (reduce sleep latency by 12.8 to 17 minutes), or benzodiazepine sleep
aides (reduce sleep latency by 10.0 to 19.6 minutes).

Back to the patient: Although
I was not impressed with melatonin’s ability to reduce sleep latency by 7
minutes, melatonin is safe, cheap, with no side effects of addiction, so we
decided to give it a try. At a follow-up visit, patient reported no benefits
from melatonin, and the search for a better sleep aide continued.

Take-home points:

1. Melatonin is safe, cheap, with no side effects of

2. Melatonin makes patients believe they have better sleep,
although when measured in reality, the improvements are small.

– By June Howell, a geriatrics fellow who blogs at
and tweets @junehowellmd

This post is part of the #GeriCases
series, in which we discuss a clinical case in geriatrics and the
attempt to provide patient-centered care with the use of best available

Ravishankar Ramaswamy, MD


Section Editor, #GeriCases

Reference: Ferracioli-Oda E1, Qawasmi A, Bloch MH
Meta-analysis: melatonin for the treatment of primary sleep disorders. PLoS
One. 2013 May 17;8(5):e63773. doi: 10.1371/journal.pone.0063773.

Image source: Moyan Brenn, no modification

This Post Has 5 Comments

  1. I depend on Melatonin and follow with interest work like the Tulane study on breast cancer association when we don't produce our own. Several studies are now spotlighting Melatonin.

    But, when we consider side effects, PLEASE read the label: after two months, we are advised to take a week off. I do take a week off.

    During that week, I need more gabapentin and Tylenol at night.

  2. Hi Margaret,

    Thank you for your comment.

    You brought up an interesting point about side effects which I did not include in the review for brevity, but what the systematic review found is that melatonin does not produce tolerance, meaning the more often/higher dose you take, the more sleep you will get, which is why they concluded that melatonin has no side effects of addiction, because one of the main features of addiction is that you will need higher and higher dose to experience the same effect you had before.

  3. Melatonin production is something maybe that is impaired in many of us, with our artificial lights, nightlights, etc. Google melatonin and parkinson's, or malatonin and dementia, or even melatonin and macular degeneration.

    In dogs we use it a lot for anxiety, problems with fireworks, sundowners/doggie dementia.

    I read a study long ago, so I don't know if the advice still holds or was validated, that melatonin in the evening was a cancer fighter, but given regularly during the day was a cancer promoter. I've tried to find the study again, but never have – it was long ago.

    Melatonin is a hormone, I don't think the "stop for a week" suggestion applies here, as it would for some herbals?

    I do think dosage is tricky – I often see high dosages frequently recommended, from 1-6 mg. I take a 500 mcg lozenge and that's all I need to knock me out. I get a better quality sleep with it, but can only take it when I know I have time for a full night's sleep.


  4. Hi Lisa T,

    Thank you for your comment.

    You're absolutely right, dosing is tricky. Rigorous study on appropriate dosage is hard to find, and this systematic review did not comment on an appropriate starting dose. There are studies that showed efficacy with low dose (0.1 to 0.3mg, which is physiologic), so in the elderly population, a good way to go is to start low, and go slow. This way, in case you experience any side effect or drug interaction, it won't be as severe as at a high dose.

    In terms of relation to cancer, it has been suggested that melatonin has anti-oxidant effects, but there have been no rigorous clinical trials, and it has NOT been approved for use as anti-oxidants.

    Hope this is helpful

  5. There are some specialized situations where you might, as a geriatrician (in a few short months), consider giving melatonin another look.

    One is for REM sleep disturbance in Lewy Body Dementia. It may be a reasonable alternative to clonazepam (which would be first line in a younger patient but poses extra risks in elderly patient or one with sleep apnea).

    There are also some studies on sleep disturbance in dialysis patients which suggest a possible role for melatonin. The hypothesis is disruption of the day-night cycle.

    It's too early for me to feel comfortable acting on, but there is some reason to think it may be useful in preventing some ICU delirium.

    Unfortunately, in the US we don't have access to standardized melatonin. It falls under the Dietary Supplement Health and Education Act. Therefore it is essentially unregulated. (Multiple investigations of different supplements from different companies have shown that what is asserted on the bottle label is often unrelated to what is found in the bottle.)

Leave a Reply

Your email address will not be published. Required fields are marked *

Back To Top