How much magnesium should I take?

The trials used under 1 g in divided doses. They were too few and too varied to establish an optimal dose, so start low and speak to a clinician.

Which form is best for sleep?

The evidence cannot say. Glycinate and citrate are better absorbed and better tolerated than oxide, which is mostly a laxative.

Does it work if I am not deficient?

The trials did not select for deficiency, so this is unresolved. Effects of supplements are generally larger in people who start with less.

Generally, at ordinary doses. Excess causes diarrhoea, and anyone with reduced kidney function should be cautious, since the kidneys clear it.

ResearchRestoration capacity

Does magnesium help you sleep?

Last reviewed 17 August 2026

The verdict

Possibly, by a small margin, on weak evidence. Pooling three randomised trials of 151 older adults, magnesium reduced time to fall asleep by 17.4 minutes against placebo. Total sleep time rose by 16 minutes and did not reach significance. Every trial carried moderate to high risk of bias and the authors graded the evidence low to very low.

Strength of evidence

traditional
emerging
informed
established

Early findings that look promising and need better studies before anyone can be sure. How we grade

What is the claim?

Magnesium is involved in hundreds of processes in the body, including the regulation of neurotransmitters and the control of muscle and nerve activity. That biological role is the basis for the claim that supplementing it helps you sleep, and it is a mechanism rather than a result.

The distinction matters here more than usual, because magnesium's involvement in so many systems makes almost any claim sound plausible.

What did the trials find?

A modest improvement in falling asleep. A 2021 systematic review and meta-analysis pooled three randomised trials covering 151 older adults across three countries, finding sleep onset latency 17.36 minutes shorter with magnesium than placebo.

Mah and Pitre published this in BMC Complementary Medicine and Therapies. Seventeen minutes is a real difference if you spend that time lying awake, so the direction is genuinely useful.

Total sleep time rose by 16.06 minutes and did not reach statistical significance, so the evidence supports falling asleep faster more than sleeping longer.

How good is the evidence?

Weak, in the authors' own assessment. All three trials carried moderate to high risk of bias, and the outcomes were supported by low to very low quality evidence. Their conclusion is that the literature is substandard for physicians to make well-informed recommendations.

Three trials and 151 people is a small base for a supplement taken by millions. Every participant was an older adult, so nothing here speaks to younger people, and the trials could not establish which form or dose works best.

The authors still note something practical. Magnesium is cheap and widely available, so the balance of a small possible benefit against low cost and low risk is reasonable, even on thin evidence.

What about the different forms?

The trials cannot settle this. Glycinate, citrate, threonate and oxide are all sold with distinct claims, and the evidence base is not large enough to compare them for sleep.

One practical difference is well established. Magnesium oxide is poorly absorbed and has a strong laxative effect, which is why it is the cheapest and why it appears in laxatives. Glycinate and citrate are better tolerated.

Magnesium L-threonate is marketed for brain effects specifically, and its sleep evidence sits at the same insufficient level as the rest.

Is it worth trying?

It is a reasonable low-cost experiment for trouble falling asleep, with modest expectations. Seventeen minutes is the size of the effect in the evidence, and anything promising more is going beyond it.

Keep doses sensible and speak to a clinician if you take medication or have kidney problems, since magnesium clearance depends on kidney function. Excess causes diarrhoea before anything else.

If falling asleep is a persistent problem, the components of cognitive behavioural therapy for insomnia have far stronger evidence than any supplement, and the largest analysis of that literature is on this site.

Questions people ask

How much magnesium should I take?: The trials used under 1 g in divided doses. They were too few and too varied to establish an optimal dose, so start low and speak to a clinician.
Which form is best for sleep?: The evidence cannot say. Glycinate and citrate are better absorbed and better tolerated than oxide, which is mostly a laxative.
Does it work if I am not deficient?: The trials did not select for deficiency, so this is unresolved. Effects of supplements are generally larger in people who start with less.
Is it safe?: Generally, at ordinary doses. Excess causes diarrhoea, and anyone with reduced kidney function should be cautious, since the kidneys clear it.

Sources

Every linked source below was opened and checked on the review date. Where a record could not be opened, it is named in the text with no link.

Mah J, Pitre T (2021). Oral magnesium supplementation for insomnia in older adults: a Systematic Review & Meta-Analysis. BMC Complementary Medicine and Therapies.
Three randomised trials, 151 older adults. Sleep onset 17.36 minutes faster, total sleep time 16.06 minutes longer and not significant. All trials at moderate to high risk of bias, evidence graded low to very low.