A new magnesium trial found real cognitive gains and a sleep signal too soft to call a win

A new randomized, placebo-controlled trial of magnesium L-threonate — sold under the brand name Magtein and marketed heavily as a sleep and brain-health supplement — found real, statistically significant improvements in cognitive performance after six weeks, and sleep results the study's own authors describe as too soft to call a clear win [s1].

What the trial actually tested

Researchers Adrian Lopresti and Stephen Smith enrolled 100 adults aged 18 to 45 with self-reported unsatisfying sleep, randomizing them to 2 grams of Magtein daily (providing 145 mg of elemental magnesium) or a placebo for six weeks [s1]. Ninety-six participants completed the trial, and 92% took more than 80% of their assigned capsules — a high compliance rate that strengthens confidence in the results, since poor adherence is a common way supplement trials produce weak or noisy findings regardless of whether the product actually works [s1]. The trial was funded by Threotech Inc., which makes Magtein, and the funder was involved in the study's design and supplied the product — a standard industry-trial structure, and one worth keeping in view when reading results this favorable to the sponsor's own product.

The cognitive results, which were genuinely positive

On the trial's primary outcome — the NIH Toolbox Total Cognition Composite Score — the Magtein group improved 8.40 points against 5.60 points for placebo, a statistically significant difference (p=0.043) that researchers translated into an estimated 7.5-year reduction in cognitive age [s1]. Two secondary cognitive measures also reached significance: working memory (p=0.033) and a reaction-time and hand-eye coordination test (p=0.031) [s1]. A measure of fluid intelligence showed no difference at all between groups (p=0.953) [s1] — a useful reminder that "cognitive improvement" in this trial was real but not uniform across every mental function tested.

The sleep results, which were much weaker

This is where the trial's own data complicates the marketing case for a "sleep supplement." Self-reported sleep-related impairment improved more with Magtein across the full sample (p=0.043) [s1]. But self-reported sleep disturbance, restorative sleep, and general wellbeing all showed no significant difference between groups [s1]. Most tellingly, every objective sleep measure recorded by an Oura Ring — sleep score, duration, efficiency, latency, deep sleep, and REM sleep — showed no significant difference [s1]. That's a meaningful gap between what participants reported feeling and what a wearable device that continuously tracked their actual sleep detected. The authors' own framing is direct: "no changes in sleep parameters as measured by the Oura Ring were demonstrated" [s1].

A limitation the authors flagged about their own sample

The study's authors point to a specific reason the sleep results may have come back muted: the enrolled population averaged 6 hours and 53 minutes of sleep with 86% sleep efficiency — figures the authors themselves describe as "within healthy levels," suggesting the objective sleep disturbance in this group was minimal to begin with [s1]. A supplement can't move a sleep metric that wasn't meaningfully impaired in the first place; the authors explicitly recommend future trials enroll people with magnesium deficiency or genuinely poor objective sleep, rather than people who mostly just feel their sleep is unsatisfying.

The one physiological finding that stands apart

Separate from both the cognitive and sleep measures, the trial found the Magtein group's average heart rate fell 1.32 beats per minute more than placebo (p=0.030), and heart rate variability rose 1.45 milliseconds against a 1.31 millisecond decline in the placebo group (p=0.036) [s1]. The authors interpret this as a signal of increased parasympathetic (rest-and-digest) nervous system activity [s1] — a real, statistically significant physiological effect, though one step removed from the subjective "better sleep" claim most buyers are actually looking for.

What to watch next

Whether a longer trial, or one enrolling people with clinically poor sleep rather than merely dissatisfying sleep, produces a clearer signal on the objective sleep measures where this trial came back flat — the authors' own limitations section treats that as the natural next study, not a settled negative result.