Twelve randomised trials of magnesium given alone for sleep in adults, none described in athletes, gave inconsistent results across questionnaires, EEG, actigraphy and wearable estimates, with low to very low certainty: the authors do not support oral magnesium as a routine insomnia treatment, while allowing a possible modest subjective benefit in some adults. From the abstract

Lopresti, Smith and Drummond reviewed randomised trials in adults in which oral magnesium was the only active treatment, or had an arm of its own, and at least one sleep outcome was reported. They searched PubMed/MEDLINE, Scopus, Cochrane CENTRAL, ClinicalTrials.gov and the WHO trials registry from inception to 1 June 2026.

Twelve trials on sleep, insomnia or poor sleep formed the main synthesis; three trials of sleep disturbed by cramps or limb movements were analysed on their own, one of them a restless-legs trial that enrolled people aged 15-50 because adult-only data were not available.

The trials differed in population, formulation, elemental dose, duration, the role sleep played among the outcomes, and the measures used, so they were described rather than pooled. Results did not agree across self-report, EEG, research-grade actigraphy, consumer wearables and physiological markers, and the certainty was low to very low.

The authors conclude that the evidence does not back oral magnesium as a routine insomnia treatment, that some adults may see a modest benefit on some subjective outcomes, and that well-designed, adequately powered trials are needed before routine recommendations.

Adults in 12 trials on sleep, insomnia or poor sleep (main synthesis), plus three trials of sleep disturbed by night cramps, restless legs or periodic limb movements (one enrolled people aged 15-50). The abstract gives no numbers of participants, ages or sexes.