“Magnesium supplementation improves sleep”

Does magnesium help you sleep? A little, for some people. The answer has grown more cautious as the trials have piled up, not less.

The broadest look so far was published in 2026. It searched five research databases, from their beginnings to June 2026, for every randomised controlled trial that tested oral magnesium on its own for sleep, and found twelve of them [1]. A randomised controlled trial is one where people are assigned by chance to the supplement or to a dummy treatment, so the two groups are otherwise alike.

Those twelve did not agree. They used different forms of magnesium, different amounts, different treatment lengths and different ways of measuring sleep, from questionnaires to brain-wave recordings to consumer sleep trackers, and the results were inconsistent across all of them [1]. The reviewers graded the certainty of the evidence low to very low and concluded that it does not support magnesium as a routine treatment for insomnia [1]. They left room for modest benefits on some self-reported measures in some adults.

Most of the older research sits in adults over 60 with long-standing insomnia, and that is where the numbers look best. A 2021 review pooled every randomised trial of magnesium for insomnia in that group, three trials and 151 people in total, and found the time taken to fall asleep was about 17 minutes shorter on magnesium than on a dummy pill [2].

Two details in that review are easy to miss. Total sleep time did not clearly improve, so what moved was dropping off rather than staying asleep [2]. And the reviewers judged all three trials to carry a moderate to high risk of bias, rating the overall evidence low to very low quality [2].

Younger adults were missing from the picture until recently. A 2025 trial randomly assigned 155 healthy adults aged 18 to 65 who reported sleeping badly to 250 milligrams of magnesium a day or a dummy pill for four weeks [3]. Both groups improved. Scores on the Insomnia Severity Index, a questionnaire running from 0 to 28 where a higher score means worse sleep, fell by 3.9 points on magnesium and 2.3 points on the dummy pill [3].

That gap of about 1.6 points is the whole effect, and the researchers describe it as small [3]. The dummy pill accounted for most of the improvement people felt.

One group did better than the rest: people who ate the least magnesium to begin with [3]. That fits the rest of the research. Magnesium looks most useful where there was a shortfall to correct, and least useful as a general sleep aid.

Is magnesium good before bed? No trial here was built to answer that. These studies gave a daily amount and measured sleep weeks later [1][2][3], and none compared taking it at night against taking it in the morning. A recommended time to take it is not something this evidence can supply.

The form question has the same hole in it. The trials used different magnesium compounds, and the 2026 review names that variation as one of the reasons their results cannot be pooled into a single number [1]. No trial has put one form head to head against another for sleep.

What the evidence does not cover:

• people with restless legs or night-time leg cramps, which the 2026 review analysed separately from ordinary insomnia [1]
• anyone taking magnesium for longer than the trials ran, which was weeks rather than months [1][3]
• children and teenagers, who sat outside the adult trials in the main review [1]
• how magnesium compares with sleep treatments that have stronger evidence behind them, because none of these trials ran that comparison [1]

Magnesium is a mineral most people already get from food, and the clearest benefit showed up in people whose intake was low [3]. Whether it adds anything on top of an adequate diet is the question these trials have not answered.