Research ·
Magnesium glycinate for sleep: the trial and 125 labels
Dr. Eric Steiner · 6 min read
Image by Nick Sokoloff
When a patient asks whether magnesium glycinate will help them sleep, there is one placebo-controlled trial of that form to point to. It found a small improvement after four weeks at 250 mg of magnesium a night, and trials of other forms disagree with one another. We then read 125 magnesium glycinate labels in the National Institutes of Health (NIH) supplement label database. Fewer than half reach that dose in one serving, none say how much glycine they contain, and 11 also contain magnesium oxide.
What did the magnesium glycinate sleep trial find?
Researchers at Leibniz University Hannover gave 155 adults who reported poor sleep either placebo or two capsules of magnesium bisglycinate before bed for four weeks. The two capsules supplied 250 mg of magnesium and about 1.5 g of glycine. Insomnia Severity Index scores, on a 28-point scale, fell 3.9 points with magnesium and 2.3 with placebo, so magnesium added 1.6 points. That is a small effect, all of it self-reported, with no difference on other sleep, mood or fatigue questionnaires. The university paid for the study. 1
Does the form of magnesium matter for sleep?
No trial has compared magnesium forms head to head. These are three separate studies.
| Oxide, 2012 | L-threonate, 2024 | Bisglycinate, 2025 | |
|---|---|---|---|
| Who | 46 older adults with insomnia | 80 adults aged 35 to 55 with sleep problems | 155 adults aged 18 to 65 reporting poor sleep |
| Magnesium a day | 500 mg | About 75 mg | 250 mg |
| How long | 8 weeks | 3 weeks | 4 weeks |
| Result | Insomnia scores and time to fall asleep improved more than with placebo | Ring-tracked deep and rapid eye movement sleep scores improved more than with placebo | Insomnia scores improved slightly more than with placebo |
| Paid for by | No conflicts declared | AIDP, which sells magnesium L-threonate | The university running the trial |
Oxide is usually described as the poorly absorbed form, and it still has a sleep trial, at twice the glycinate dose. 2
The threonate trial used far less magnesium and was paid for by the company that sells the ingredient. All three trials are small and short, and none of them shows that the form matters. 3
Do magnesium glycinate products match the trial dose?
Only 48 products match the trial's dose and form, and none of those say how much glycine they contain.
- 42% reach the trial's 250 mg in one labeled serving. A typical capsule holds about 100 mg, and two-thirds of capsule and tablet products take three or more to reach 250 mg.
- None say how much glycine they contain. The trial gave about 1.5 g a night alongside the magnesium.
- 9% are partly magnesium oxide. Their panels list "magnesium glycinate and magnesium oxide" without saying how much comes from each, and their names don't mention oxide.
- Some names give the compound weight. Magnesium is about 14% of the compound, so a product called "Magnesium Glycinate 500 mg" can supply 70 mg of magnesium per capsule.
- 48% allow more than the upper limit at their highest dose. NIH sets 350 mg a day as the upper limit for magnesium from supplements, mainly because of diarrhea.
To compare products, find the Supplement Facts line that starts with "Magnesium," such as "Magnesium (as magnesium bisglycinate) 200 mg." That figure is the magnesium itself, and it is the number to hold against the trial's 250 mg. 4
What to tell a patient
Look past the glycinate trial and the evidence is mixed. A 2026 review of 12 randomized trials of magnesium alone found that they differed in form, dose and population, and that their results were inconsistent across questionnaires, sleep studies and wearables. It rated the certainty low to very low and concluded magnesium shouldn't be a routine treatment for insomnia. 5
Some trials found nothing. In 100 adults over 51 who slept poorly, 320 mg a day as citrate for seven weeks improved sleep scores no more than placebo did. 6
The clearest signal is in older adults with insomnia. Across three small trials, they fell asleep about 17 minutes sooner on magnesium than on placebo, on low-quality evidence. 7
So magnesium may help some people a little, the form hasn't been shown to matter, and placebo does much of the work. If a patient wants to try it, most trials used 250 to 500 mg of magnesium a day for four to eight weeks, which gives a dose and a time frame to judge it by. Above 350 mg a day, the upper limit for supplements, diarrhea becomes more likely.
Check kidney function and the medication list first. Impaired kidney function raises the risk of magnesium toxicity, and magnesium can reduce absorption of some antibiotics and oral bisphosphonates. 4
How we counted
On October 7, 2026 we searched the NIH Dietary Supplement Label Database for on-market products with "magnesium glycinate" or "magnesium bisglycinate" in the name, then kept the newest label for each product: 125 in all. The database doesn't hold every product sold. 8
The magnesium monograph compares the forms and covers the other uses studied, the doses and the interactions.
Sources
- 1Schuster et al., 2025PubMed CentralFour weeks, 155 adults with self-reported poor sleep. Insomnia scores improved 1.6 points more with magnesium than placebo. Academic funding.
- 2Abbasi et al., 2012PubMedEight weeks of 500 mg a day as oxide, in 46 older adults with insomnia.
- 3Hausenblas et al., 2024PubMedThree weeks of 1 g a day of magnesium L-threonate, about 75 mg of magnesium, in 80 adults. Funded by AIDP, the ingredient's supplier.
- 4NIH Office of Dietary SupplementsNIH ODSSupplement Facts list elemental magnesium. The upper limit for supplemental magnesium is 350 mg a day. Oxide is absorbed less completely than several other forms.
- 5Lopresti et al., 2026PubMed12 randomized trials of magnesium alone for sleep in adults, searched to June 2026. Lopresti also co-authored the 2025 glycinate trial.
- 6Nielsen et al., 2010PubMed100 adults aged 51 to 85 with poor sleep. Sleep scores improved as much on placebo as on magnesium citrate over seven weeks.
- 7Mah et al., 2021PubMedMeta-analysis of 3 trials in 151 older adults with insomnia. Sleep onset 17 minutes faster; total sleep time not significantly longer. Low to very low quality evidence.
- 8NIH Dietary Supplement Label DatabaseNIH label database162 labels entered between 2013 and 2025, reduced to 125 products. Capsule counts use the 97 sold as capsules, tablets or softgels.