Pick an organic form such as bisglycinate or citrate, stay under 250 mg a day from supplements and take it at the same time every day. In a double-blind trial, magnesium oxide was absorbed no better than a placebo. And the effect on sleep is small. That it works mainly in people who get too little is plausible. It has not been proven.
Which magnesium should you take for sleep? An organic form, so bisglycinate or citrate. How much? Stay under 250 mg a day from supplements, the safe upper limit in Europe. When? At the same time every day, because no study has compared morning with evening.
And here is the answer you hear less often: the effect is small. In older adults with insomnia, falling asleep took about seventeen minutes less. If your intake is already fine, you will probably barely notice it.
Below you will find, question by question, what the research shows, including the form that was absorbed no better than a placebo in a double-blind trial.
Here the research is surprisingly clear. In a randomised double-blind trial, 46 healthy adults took 300 mg of magnesium for sixty days as an amino acid chelate, as citrate or as oxide. Both organic forms were absorbed better than oxide. Citrate produced the highest blood level.
Magnesium oxide was absorbed no better than the placebo on any measure. And that happens to be the cheapest form, the one you find in almost every supermarket.
Between bisglycinate and citrate the evidence is thin. Anyone claiming one clearly wins is going beyond what was measured. The practical difference: citrate pulls water into your gut and can loosen your stools at a higher dose, bisglycinate does that far less.
Source: Walker et al.3 2003, Magnesium Research
Want your training and nutrition to adjust to your own data every week? That is what I am building an app for. Join the waiting list
The studies that found something used 250 to 500 mg a day. The most recent one, with 155 participants and 250 mg as bisglycinate, saw a small measurable effect on insomnia after four weeks.
Treat 250 mg a day from supplements as your ceiling. That is the safe limit in Europe. Above that dose you mainly risk loose stools.
Do not count your food towards that limit. Green vegetables, nuts and wholegrains deliver magnesium without that risk. Anyone eating enough of those probably does not need the supplement.
Source: Abbasi et al.1 2012, Journal of Research in Medical Sciences; Nielsen et al.2 2010, Magnesium Research; Schuster et al.6 2025, Nature and Science of Sleep; EFSA, Wetenschappelijk Comité voor de menselijke voeding7 2006, Tolerable Upper Intake Levels for Vitamins and Minerals
No study has compared taking it in the morning with taking it before bed. Anyone telling you there is one right moment is filling in that gap themselves.
What does count is taking it every day. The trials in which magnesium did something ran four to eight weeks without a break. So pick a moment you can keep up and attach it to something you already do daily.
Source: Abbasi et al.1 2012, Journal of Research in Medical Sciences; Schuster et al.6 2025, Nature and Science of Sleep
In 46 older adults with insomnia, 500 mg of magnesium a day shortened the time to fall asleep after eight weeks. Cortisol dropped and melatonin rose. Total sleep duration did not change. In another study with 100 participants, sleep improved just as much in the magnesium group as in the placebo group.
An earlier summary, a 2021 systematic review, found only three usable studies with 151 participants between them. On average people fell asleep 17 minutes faster. The researchers themselves rate that evidence as low to very low. In the 2025 trial the effect was strongest in those with a low intake beforehand.
The newest review, from 2026, brought together twelve randomised trials and reaches the same verdict. The results contradict each other and the certainty is low to very low. The researchers do not recommend magnesium as a routine treatment for insomnia.
That is the heart of it: the effect is small. That magnesium mainly works in people who get too little of it is a plausible explanation. It has not been proven. In the 2025 trial it came from an exploratory analysis. The positive study in older adults only included people with a low intake.
Source: Abbasi et al.1 2012, Journal of Research in Medical Sciences; Nielsen et al.2 2010, Magnesium Research; Mah & Pitre4 2021, BMC Complementary Medicine and Therapies; Schuster et al.6 2025, Nature and Science of Sleep; Lopresti et al.5 2026, Journal of Dietary Supplements
Bisglycinate is often recommended because the glycine in it would deepen your sleep. Glycine does work. In 11 people, 3 grams before bed shortened both falling asleep and the path into deep sleep, measured in the sleep lab. Two other small studies saw less tiredness the next morning.
The dose is the problem. One bisglycinate tablet with 100 mg of magnesium holds about 0.6 grams of glycine, a fifth of what worked in the studies. For 3 grams you would need five tablets, which means 500 mg of magnesium: double the safe upper limit.
On top of that, all three glycine studies were run with authors from Ajinomoto, the largest glycine producer in the world. In Bannai the participants were even employees of that company. Around 33 participants in total, without independent replication by another group.
So choose bisglycinate for the gentle absorption. The glycine that comes with it is a bonus too small to count on.
Source: Yamadera et al.9 2007, Sleep and Biological Rhythms; Inagawa et al.8 2006, Sleep and Biological Rhythms; Bannai et al.10 2012, Frontiers in Neurology; EFSA, Wetenschappelijk Comité voor de menselijke voeding7 2006, Tolerable Upper Intake Levels for Vitamins and Minerals
I recently saw a post from another coach: chamomile tea, sea salt and magnesium as a natural sleeping pill. It sounded logical, so I repeated it in my own story. A fine example of what Socrates called false knowledge: believing you know something because it sounds plausible.
I am a historian by training and a data geek, so I want to see the studies first, even when it concerns my own story. So I read the research behind those three remedies. On chamomile tea and on salt and night-time urination I wrote a separate piece. That last claim I take back.
Supplements are the last five per cent. The first 95 sit in your structure.
To bed around the same time every day. Your phone outside the bedroom. No caffeine in the six hours before bed. A moment outside in the daylight first thing in the morning.
You need no tracker for that or fancy supplements. Stick to the basics.
Source: Windred et al.13 2024, Sleep; Drake et al.14 2013, Journal of Clinical Sleep Medicine
| Form | Absorption in research | Stools | What it means |
|---|---|---|---|
| Bisglycinate | better than oxide | gentle | solid default choice for daily use |
| Citrate | highest blood level of the three | can loosen at a higher dose | fine choice, watch the dose |
| Oxide | no better than placebo | can loosen | the cheapest form and the weakest |
An organic form, so bisglycinate or citrate. In a double-blind trial with 46 adults both were absorbed better than magnesium oxide, which did no better than a placebo on any measure. Between bisglycinate and citrate the evidence is too thin to name a winner.
The studies that found something used 250 to 500 mg a day. Treat 250 mg a day from supplements as your ceiling, the safe limit in Europe. Magnesium from food does not count towards it.
No study has compared morning with evening, so there is no proven best moment. What counts is taking it daily: the trials ran four to eight weeks without a break.
A little. In older adults with insomnia, falling asleep took an average of seventeen minutes less, according to a review of three studies with 151 participants. The researchers themselves rate that evidence as low to very low.
It is the form that is absorbed worst. In a double-blind trial, magnesium levels rose no more after sixty days than with a placebo. Worth noting: the study in older adults with insomnia that did see faster sleep onset used magnesium oxide, at 500 mg a day. An organic form remains the better choice for absorption.
About 0.6 grams in a tablet with 100 mg of magnesium. In the studies where glycine improved sleep it was 3 grams, so five times as much. Reaching that would mean 500 mg of magnesium, double the safe upper limit.
There is no separate research on that, so I cannot make a claim. What is established: the effect on sleep is small on average. That there is more to gain for people who get too little magnesium is plausible. It has not been proven.
Timon offers High Performance 1:1 Health Coaching, wherever you live. Coaching clients can also train with him in person in Ghent, Belgium.
Leave a comment below. Say what you think, or tell me which claim to look into. Your comment appears right away and everyone can read it.
What I researched that week, summed up in three sentences, sources included. Unsubscribe any time.
Every number in this article was checked against the original publication.
This article is information, not medical advice. If you have symptoms, are pregnant or take medication, talk to your doctor before changing anything.
Chamomile tea works well as a calming evening ritual. As a sleep aid the evidence is thin: in the only trial in 34 people with chronic insomnia, chamomile extract did no better than placebo. The positive studies used extract in capsules. Allergic to the daisy family or on blood thinners? Check first.
Read the article
A pinch of salt before bed does not help against night-time urination. In 223 people who ate less salt, night-time urination fell from 2.3 to 1.4 times a night. What also helps: looking at what you drink across the day and not staying in bed longer than you need. Regularly up twice or more? Get it checked.
Read the articleYou log, the app reads your trend and moves your targets every week. Training, nutrition, lifestyle and sleep in one plan. On the waiting list you hear first when it opens.
