Sleep
Magnesium for sleep: does it work? What the newest evidence says
Magnesium for sleep: a 2026 review of 12 trials found small, mixed effects. Whether to keep taking it, which form has evidence and what to check first.
1h Protocols · October 4, 2026 · 10 min read
Magnesium may help some people sleep a little, but the best evidence so far can't promise it will help you. A systematic review published in August 2026 gathered the randomized trials of magnesium for sleep, taken on its own, and found inconsistent results, with low to very low certainty (Lopresti et al., 2026).
If you've bought a tub of magnesium glycinate because a podcast swore by it, you've met the confusion. Neuroscientist Matthew Walker argued on The Diary of a CEO in 2025 that most magnesium never reaches the brain (The Diary of a CEO), Oxford circadian neuroscientist Russell Foster called the data "still not absolutely solid" (Feel Better, Live More), and sleep specialist Michael Breus ran one of the trials himself (The Diary of a CEO).
Does magnesium help you sleep? What 12 trials found
The review kept only randomized trials of magnesium on its own against a placebo, so products that mix it with melatonin, zinc or herbs are left out. One of the more encouraging results came from a 2025 trial of magnesium bisglycinate in German adults with poor sleep. After four weeks, scores on a standard sleep-problem questionnaire improved by 3.9 points with magnesium and by 2.3 points with the placebo. In other words, the placebo group got more than half as much improvement with no magnesium at all, and the reviewers call the extra effect small. Other trials found no benefit beyond placebo, and measured sleep barely moved: a research-grade wrist monitor found no clear effect, ring estimates were mixed, and the one brain-wave study was tiny.
The reviewers rated the certainty of the evidence low or very low for every sleep outcome and concluded that current evidence doesn't support taking magnesium routinely for long-term sleep problems, though modest improvements in some self-rated outcomes "remain possible" (Lopresti et al., 2026).
Should you keep taking magnesium for sleep?
Whether you're about to buy a tub or already have one, run through these checks before you think about which form to take.
Fix the basics first
Asked about supplements, Walker goes straight to the fundamentals: a regular schedule, less caffeine and alcohol, dimmer lights in the evening and fewer screens. On Huberman Lab he suggests dimming at least half your lights in the last hour before bed (Huberman Lab). Start with how to sleep better and the caffeine cut-off. The 1h Protocol for Better Sleep is built on the same basics and ends with a personal protocol for your own nights.
Ask whether your intake might be low
The case for magnesium is strongest where people lack it. Walker traces the sleep story to older studies in which some people with disrupted sleep turned out to be deficient and slept better once that was corrected. If your levels are already normal, he argues, extra magnesium can't push you past the ceiling, and "all you're doing is creating probably expensive urine" (The Diary of a CEO). Breus starts from the same place: "Fix your deficiencies first."
The review is more cautious. It says magnesium may be worth considering when low intake or deficiency is suspected for other reasons, but the trials didn't consistently measure people's magnesium intake or levels, so they can't show that people with low levels benefit more. Whether you need a blood test is a question for your doctor.
Check safety with a doctor or pharmacist
In the trials, magnesium was generally well tolerated, and gut symptoms were the most likely side effect. The reviewers add that side effects were recorded inconsistently and many trials were short, and that any decision should take into account your diet, kidney function, other health conditions, the medicines you take and the elemental magnesium in the product. The NHS says: "Check with your doctor before taking anything for your sleep problems" (NHS).
Expect a small change, if any
The review's own phrase for any expected benefit is "uncertain and likely modest." The trials ran for two to twelve weeks. If you try magnesium, a simple sleep diary (when you go to bed, wake in the night and get up, plus naps, caffeine and alcohol) gives you something better than memory to judge by, and something to show your doctor (CDC).
Which magnesium is best for sleep?
No form has been shown to beat another: no trial has compared forms head to head, and the trials also differed in dose, length, population and quality. The 12 trials in the main analysis tested seven kinds of magnesium:
| Form | Trials | What they found |
|---|---|---|
| Bisglycinate (glycinate) | 1 | Small improvement on a sleep-problem questionnaire after four weeks; other sleep and daytime measures not clearly better |
| L-threonate | 3 | Some gains in daytime functioning and selected ring-measured outcomes; sleep-problem scores not consistently better |
| Oxide | 3 | Mixed: one positive trial in older adults, one with no benefit, and a small brain-wave study |
| Chloride preparations | 2 | One small two-week trial with a borderline gain in sleep quality and no change in sleep-problem scores; one with no clear benefit |
| Citrate | 1 | Both groups improved; no clear benefit from magnesium |
| Gluconate | 1 | Improvements reported, but the reviewers judge the trial at high risk of bias |
| Food-based preparation | 1 | Better diary-rated sleep quality; no clear effect on wrist-monitored sleep |
If you searched for magnesium glycinate for sleep, the German bisglycinate trial above is the only glycinate trial in the review. Its effect was small, and a single trial can't show a pattern.
Threonate is Walker's pick. Most forms, such as oxide and citrate, don't cross the blood-brain barrier, he says, and "sleep is produced by your brain"; magnesium L-threonate is the one form he thinks has some evidence (The Diary of a CEO). The review notes that laboratory studies suggest threonate may raise magnesium in the brain at low doses, but says this "has not translated consistently into better sleep outcomes." In its three trials, daytime functioning improved more often than sleep itself.
Breus said on The Diary of a CEO that the head of a supplement company handed him a $75,000 check to test its magnesium, and that "His magnesium really made people sleep better" (The Diary of a CEO). In the review, that small trial (two weeks on magnesium and two on placebo) showed no change in sleep-problem scores and a borderline improvement in sleep quality, with some ring estimates favoring magnesium. The reviewers class it as a small pilot with partial benefit.
Melatonin vs magnesium: does melatonin work?
Melatonin is a weaker bet still. Walker describes it as a timing signal: "Melatonin is like the starting official at the 100-m race." By his reading of the meta-analyzes, it helps people fall asleep only about 3.4 minutes faster, "not much more than placebo," and high doses, far more than the body makes, can leave you in a fog the next morning (The Diary of a CEO). Foster agrees that melatonin "is most emphatically not a sleep hormone" (Feel Better, Live More). Both point to jet lag as a good use, because melatonin taken at the right time helps the body clock catch up with a new time zone.
The European Insomnia Guideline (2023) fits that picture. It doesn't recommend ordinary fast-release melatonin for trouble sleeping, because it doesn't work well enough, except where body-clock timing plays a part. It also warns that melatonin quality varies widely and advises buying it only from licensed pharmacies or on prescription (Riemann et al., 2023).
Start with what costs nothing
The habits Walker lists first (a regular wake time, less caffeine and alcohol, dimmer evenings) cost nothing, and in his view they do far more for sleep than any supplement. Run them for a few weeks with a sleep diary, and you'll have your own evidence before the next order.
If you're suffering from sleep problems, and you're looking to supplements, you're stepping over dollars to pick up pennies. (Matthew Walker, The Diary of a CEO)
If you've been sleeping badly for three months or more, book a visit with your GP or doctor. Long-term sleep problems have treatments that work.
Before you buy another bottle, try the free part of the 1h Protocol for Better Sleep: it finds your sleep pattern and gives you one first step built on those basics. Start free.
Frequently asked questions
Does magnesium actually help you sleep?
Possibly a little, for some people. A 2026 systematic review of 12 randomized trials found that some reported small improvements in self-rated sleep, while several found no benefit beyond placebo. The reviewers rated the certainty of the evidence low to very low.
Which magnesium is best for sleep?
No form has been shown to be best, because no trial has compared forms directly. One trial of magnesium bisglycinate (glycinate) found a small benefit on a sleep-problem questionnaire, and three trials of L-threonate improved some daytime measures more than sleep itself.
How much magnesium should I take for sleep?
The trials don't identify a best dose. Several reported only the weight of the whole compound, not the elemental magnesium it contains; where elemental doses were reported, they ranged from about 75 to 729 mg a day, and positive and null results appeared across that range. Ask a doctor or pharmacist before starting, particularly if you take medicines or have kidney problems.
Does magnesium help with sleep and anxiety?
The idea is plausible: the 2026 review lists magnesium's role at brain receptors involved in arousal, and possible effects on GABA, the brain's main calming messenger, and on stress hormones such as cortisol, as ways it could help sleep. The trials haven't confirmed it. The one trial in people with both sleep and anxiety complaints, adults aged 50 to 70 with memory problems, found no clear sleep benefit over placebo.
Is melatonin better than magnesium?
Not for everyday trouble sleeping. Matthew Walker reads the meta-analyzes as showing that melatonin helps people fall asleep only about 3.4 minutes faster, and the 2023 European Insomnia Guideline doesn't recommend fast-release melatonin for trouble sleeping unless body-clock timing is involved, as with jet lag.
Sources
This article draws on the 2026 systematic review by Lopresti, Smith and Drummond, podcast interviews with Matthew Walker, Russell Foster and Michael Breus, and guidance from European, UK and US health bodies. One disclosure from the review: two of its three authors ran two of the included trials, and the third, who wasn't involved in either, checked those assessments independently.
- Magnesium supplementation for sleep in adults: a systematic review of randomized controlled trials (systematic review, Lopresti, Smith and Drummond, Journal of Dietary Supplements, 2026)
- World No.1 Sleep Expert: Magnesium Isn't Helping You Sleep! (podcast episode, The Diary of a CEO with Matthew Walker, November 2025)
- Separate beds, the truth about sleep supplements and naps, with Professor Russell Foster (podcast episode, Feel Better, Live More #424, 2024)
- Human sleep expert: Dr Michael Breus (podcast episode, The Diary of a CEO, 2026)
- The European Insomnia Guideline: an update on the diagnosis and treatment of insomnia 2023 (clinical practice guideline, European Sleep Research Society, 2023)
- Protocols to improve your sleep, with Matthew Walker (podcast episode, Huberman Lab Guest Series, Episode 2)
- Insomnia (public health guidance, NHS, 2024)
- About sleep (public health guidance, CDC, 2024)
