You can buy magnesium in a dozen forms, and almost every label promises better sleep. A 2024 randomized trial in Sleep Medicine: X tested one of them, magnesium L-threonate (MgT, sold under the brand name Magtein). After just 21 days, adults taking it had better deep and REM sleep scores on their Oura rings than those taking placebo. The result is interesting, but the details matter, and so do the caveats.
Why the magnesium form matters
Magnesium is a mineral, but supplements deliver it attached to a carrier molecule: citrate, glycinate, oxide, threonate and others. The carrier can influence absorption, digestive tolerance and, possibly, where the magnesium ends up.
MgT was designed around that last idea. It grew out of rodent research (Slutsky et al., Neuron, 2010) suggesting that threonate could help raise magnesium levels in the brain. The new trial's authors describe MgT as promising because of its brain bioavailability and its links to cognition, memory and mood. That is the rationale. It is not proof that the same effect translates into better sleep in people, which is what this study tried to examine.
The trial: who, what and how long
The study was randomized, double-blind, placebo-controlled and run in parallel groups.
- Participants: 80 adults aged 35 to 55 with self-reported sleep problems
- Intervention: 1 g/day of MgT (Magtein) or placebo
- Duration: 21 days
- Subjective measures: Insomnia Severity Index, Leeds Sleep Evaluation Questionnaire, Restorative Sleep Questionnaire, Profile of Mood States, and a daily diary
- Objective measure: an Oura ring
Pairing questionnaires with a wearable is a sensible design. People's sense of how they slept and what a device records don't always agree, so seeing both move in the same direction is more persuasive than either alone.
The Oura results
Compared with placebo, the MgT group showed statistically significant improvements (p < 0.05) in:
- Deep sleep score
- REM sleep score
- Light sleep time
- Activity score and daily movement score
- Readiness score, readiness activity balance and readiness sleep balance
The authors summarized this as improved sleep quality, especially in the deep and REM stages. Note that these are Oura's composite scores, not necessarily extra minutes in each stage, so avoid reading them as "X more minutes of deep sleep." Also, the abstract says the MgT group maintained good sleep quality and daytime functioning while the placebo group declined. Part of the headline gap therefore comes from the placebo group getting worse, not only from the MgT group improving.
Mood, energy and daytime function
The questionnaires pointed the same way. Versus placebo, MgT significantly improved behavior upon awakening, energy and daytime productivity, grouchiness, mood and mental alertness. Sleep supplements are often judged on how you feel the next day, and that is where this trial reported some of its clearer signals.
The authors say the findings fit what is known about MgT in neurons and animal models, and they suggest broader brain-health benefits. That is a hypothesis for future work, not something this study established.
Magnesium threonate vs glycinate vs citrate
This is the question most people searching "best magnesium for sleep" want answered, and this trial cannot answer it. It compared MgT to placebo, not to glycinate, citrate or any other form. Any ranking would be speculation.
What can be said, with hedging:
- Threonate: the form with a brain-delivery rationale and now a small positive sleep trial, though much of the supporting mechanistic work is in animals and the form is typically branded and more expensive.
- Glycinate: widely used for sleep and generally considered gentle on the stomach, with a plausible calming role for the glycine component. Head-to-head evidence against threonate is lacking.
- Citrate: well absorbed and common, but better known for its laxative effect at higher doses.
The authors themselves note that clinical evidence for magnesium and sleep has varied depending on the magnesium source. That is a reason to be cautious about assuming all forms behave alike, and equally cautious about assuming threonate is superior.
Practical takeaways
Dose. The trial used 1 g/day of MgT. That is the compound, not the elemental magnesium. At roughly 7 to 8 percent magnesium by weight, 1 g provides about 72 to 80 mg of elemental magnesium, which is modest next to the 300 to 400 mg many people associate with magnesium supplements. Check the label: some products list a different serving size, and the dose tested here may not match what's on the bottle.
Timing. The abstract doesn't specify when participants took it. Since the target is sleep, many people take MgT in the evening, but that is a practical convention, not a finding from this trial.
Who it might suit. Adults in roughly the 35 to 55 range who have self-described sleep complaints and are curious about a supplement trial. It isn't evidence for treating diagnosed insomnia, sleep apnea or other sleep disorders.
Side effects. The authors reported MgT was safe and well tolerated. Magnesium supplements in general can cause loose stools or stomach upset, and people with kidney disease or those taking medications that interact with magnesium (certain antibiotics and bisphosphonates, for example) should talk to a clinician or pharmacist first.
Honest limitations
- Short duration. Three weeks says nothing about long-term effects or whether benefits persist.
- Wearable sleep staging. Oura rings estimate sleep stages and are not polysomnography, the lab standard. Consumer devices can be useful for tracking trends but are less reliable at assigning specific stages.
- Self-reported sleep problems. Participants weren't necessarily clinically diagnosed, which limits how far the results generalize.
- Small sample. Eighty people is enough for a signal, not for firm conclusions.
- Branded ingredient. Magtein is a proprietary ingredient. Readers should check the paper's disclosures for industry ties or funding, and independent replication is still limited.
- Many outcomes. With numerous measures tested, some p < 0.05 results can arise by chance, which is another reason to want replication.
The bottom line
This trial is a promising, well-structured early signal that MgT may support deep and REM sleep scores, mood and daytime energy over three weeks. It doesn't show that threonate beats glycinate or citrate, and it doesn't settle the dose, timing or long-term question. Treat it as a reason for larger, independent, longer trials, and as a conversation starter with your clinician if you're considering it.
