Does magnesium glycinate help sleep? The trials, in plain terms

A little, and mostly at the start of the night. In 155 adults who slept badly, 250 mg of magnesium as bisglycinate cut insomnia scores by 3.9 points in four weeks, against 2.3 points on a dummy capsule.¹ Older adults with insomnia fell asleep about a quarter of an hour sooner.²

The one trial that names its form

Bisglycinate is the compound sold as glycinate: magnesium bound to glycine, an amino acid. It is the only form any trial in this set names.

That trial enrolled 155 adults aged 18 to 65 who said they slept poorly.¹ Half took 250 mg of elemental magnesium a day, half took a dummy capsule, and everyone was scored on the Insomnia Severity Index over four weeks.¹ That score comes from a questionnaire, where a higher number means worse sleep.

Note the entry ticket. Nobody was screened for a magnesium shortage. Sleeping badly was the whole requirement, which makes this a test of magnesium in ordinary poor sleepers rather than in depleted ones.

The word elemental earns a line as well. A bisglycinate capsule weighs far more than the magnesium inside it, so 250 mg of elemental magnesium comes in a bigger capsule than the number suggests.¹ A label quoting the weight of the whole salt is quoting a larger number for the same mineral.

What it moved, and by how much

Insomnia scores fell 3.9 points in the magnesium group and 2.3 in the dummy group.¹ The gap of 1.6 points scraped past the usual significance line, at p = 0.049.¹ Nothing else the trial measured shifted at all.¹

The effect size came out at 0.2.¹ On that scale 0.2 counts as small, 0.5 as moderate and 0.8 as large, so the authors' own word for it, modest, is the right one.¹

Two features of that result matter more than the headline. The first is the dummy group, which improved by more than half as much as the magnesium group.¹ Most of what a person taking bisglycinate would feel across four weeks is not the magnesium.

The second is that the benefit was not spread evenly. Participants who reported eating less magnesium at the start improved noticeably more, which the authors read as a possible group of high responders.¹ That is an exploratory result inside one trial, so treat it as a lead rather than a finding.

The bigger number comes from the weakest trials

Across three trials in 151 older adults with insomnia, magnesium cut the time taken to fall asleep by 17.36 minutes against a dummy, at p = 0.0006.² Total sleep time rose by 16.06 minutes, which did not reach significance.² So people got to sleep sooner without clearly getting more sleep.

A quarter of an hour is a real change for someone lying awake. The trouble is what stands behind it. All three trials were at moderate-to-high risk of bias, and the reviewers graded the findings low to very low quality.²

Their own summary deserves quoting closely. The literature, they wrote, is substandard for making well-informed recommendations.² But magnesium is cheap and easy to get, so the evidence may support it for insomnia symptoms in amounts under 1 g, given up to three times a day.² That is a shrug with a dose attached.

Who improved, and how anyone would tell

The low-intake thread runs through the rest of this evidence. A review of supplement trials found that five of the eight sleep studies reported improvement on at least one sleep measure, two reported none, and one was mixed.³ Its authors expected the benefit to land mainly on people not getting much magnesium from food in the first place.³ They also listed what held them back: few participants per study, and dosages, formulations and durations that changed from trial to trial.³

A trial in people with diabetes comes at the same idea from a group where shortfall is plausible. Of 320 patients with insomnia, 290 finished two months on a dummy, magnesium, potassium, or magnesium and potassium together.⁶ Insomnia severity differed across the groups, at p = 0.0001, and cortisol and melatonin moved as well.⁶ Two cautions apply. The trial was single blind, so people knew what they were taking, and potassium alone shifted the same measures.⁶

The widest survey draws the line most clearly. Across nine studies covering 7,582 people, the observational data tied magnesium status to sleep quality, including daytime sleepiness, snoring and sleep duration, while the randomized trials came back contradictory.⁴ Having more magnesium in the body goes with sleeping better. Taking more has been much harder to show.

Where the form question actually stands

Glycinate's reputation rests on absorption and on being easy on the stomach, not on sleep trials that compared one salt against another. The bisglycinate trial names its form and had nothing to compare it against.¹ The older-adult review does not report which forms its trials used.² A review of 31 sleep-supplement trials combined results for amino acids, melatonin and vitamin D, and found magnesium had too little data to combine at all.⁵

That leaves a plain statement. There is a small effect for bisglycinate in poor sleepers, and a larger, shakier one for magnesium of unstated form in older adults with insomnia.^1,2 Whether glycinate beats citrate or oxide for sleep is a question nothing here asked.

What remains unknown

The bisglycinate trial measured sleep by questionnaire, and its authors want objective measurement, longer courses, and a proper description of those possible high responders.¹ The older-adult result rests on three biased trials, and its own reviewers say the literature cannot guide use.² The wider survey asks for larger trials followed beyond 12 weeks.⁴ The most useful forms and doses remain unestablished.³

And no abstract here says when in the day the magnesium was swallowed, beyond up to three times a day.² The evening-dose habit, which is how almost everyone takes it, has no trial behind it.