Research Watch: What Recent Studies Actually Say About Supplements for Sleep

Welcome to the second Research Watch. Every Saturday I read the recent supplement research that crossed my desk and write up the studies worth knowing about. No product push, no marketing angle. Just the studies, what they found, and what I think it means for someone trying to make informed choices. This week: what the recent peer-reviewed evidence says about supplements for sleep.

Quick Answer

Melatonin has the strongest evidence for reducing sleep onset latency, with an optimal dose near 4 mg taken about 3 hours before bed, not the 5 to 10 mg most products contain. Magnesium bisglycinate shows a small but real benefit, mostly in people with low dietary magnesium intake. Ashwagandha and L-theanine show modest improvements in sleep quality and next-day function in people with anxiety-adjacent sleep issues. Glycine and magnesium L-threonate show benefit more on next-day function than on sleep architecture itself. Valerian and high-dose magnesium in people without baseline deficiency have weak recent evidence.

What This Week's Reading Covered

The sleep supplement market is enormous and uneven. I deliberately stayed in the highest-quality recent literature: systematic reviews and dose-response meta-analyses from 2024 and 2025, plus a small number of well-designed 2023 to 2026 randomized controlled trials. The goal is not to defend or attack any ingredient. It is to read what the studies actually report, in their own numbers, before anyone wraps a marketing story around them. Six studies anchor this post. Each one is linked to PubMed so you can read the abstract yourself.

Magnesium Bisglycinate in a Healthy-but-Poor-Sleeping Population

The best-powered recent magnesium sleep trial is a 2025 randomized double-blind study in Nature and Science of Sleep. Researchers enrolled 155 German adults aged 18 to 65 who self-reported poor sleep quality but had no diagnosed sleep disorder. Participants took 250 mg of elemental magnesium as bisglycinate (which also delivered about 1.54 g of glycine via the chelate) daily for four weeks.

The intervention group showed a statistically significant reduction in Insomnia Severity Index scores compared to placebo (mean change of minus 3.9 versus minus 2.3, p equals 0.049). The effect size was small (Cohen d 0.2). The most important detail buried in the paper is that responders were concentrated among participants with lower baseline dietary magnesium intake. If your diet already covers magnesium through leafy greens, nuts, seeds, and whole grains, the evidence that supplemental magnesium will meaningfully move your sleep is weaker than marketing suggests.

Source: Schuster J et al. Nature and Science of Sleep. 2025. PubMed 40918053

Ashwagandha for Anxiety-Adjacent Sleep Problems

A 2024 systematic review and meta-analysis in Human Psychopharmacology pooled five randomized controlled trials covering 254 adults with anxiety or insomnia. Ashwagandha root extract significantly improved sleep onset latency, total sleep time, sleep efficiency, and Pittsburgh Sleep Quality Index scores. It did not significantly change wake after sleep onset or total time in bed. The effect on anxiety (mean difference on the Hamilton Anxiety Scale of minus 5.96, p equals 0.008) was larger than the effect on sleep itself, fitting the working hypothesis that ashwagandha helps sleep mostly by reducing the cortisol and rumination that interferes with falling asleep.

The honest caveat is that pooled heterogeneity was very high at 98 percent, meaning the included trials varied significantly in extract type, dose, and population. The pooled estimate is suggestive, not definitive. Doses of 600 mg or higher of standardized root extract for at least 8 weeks showed the most consistent signal.

Source: Fatima K et al. Human Psychopharmacology: Clinical and Experimental. 2024. PubMed 39083548

L-Theanine: Smaller Effect on Sleep, Bigger Effect on Next-Day Function

A 2025 meta-analysis in Sleep Medicine Reviews pulled together 19 studies covering 897 participants. L-theanine produced a small but statistically significant improvement in subjective sleep onset latency (standardized mean difference 0.15, p equals 0.04) and overall subjective sleep quality (standardized mean difference 0.43, p equals 0.03). Its largest effect was on next-day daytime dysfunction (standardized mean difference 0.33, p less than 0.001), more than double the sleep onset effect.

The most accurate way to describe what the evidence shows is that L-theanine may not be a strong sleep onset aid, but it appears to soften the next-day consequences of imperfect sleep in people with stress-related disruption. The biggest limitation is that most included studies used L-theanine combined with other ingredients rather than in isolation, so attribution to L-theanine alone is partial.

Source: Bulman A et al. Sleep Medicine Reviews. 2025. PubMed 40056718

Melatonin: The Dose-Response Finding That Matters

This is the finding I want every reader to walk away with. A 2024 dose-response meta-analysis in the Journal of Pineal Research analyzed 26 randomized controlled trials covering 1,689 observations across people with insomnia and healthy volunteers. The authors modeled the relationship between dose, timing, and sleep outcomes.

Optimal effects on sleep onset latency and total sleep time occurred at approximately 4 mg taken about 3 hours before the intended sleep time. The dose-response curve plateaued. Higher doses did not produce proportionally better sleep, and may actually shift circadian timing in unintended ways. Walk into any pharmacy in Canada and you will mostly see melatonin sold at 5 mg or 10 mg, with some products going up to 20 mg. The dose-response data does not support those doses as more effective than 4 mg or below. Timing relative to your intended sleep onset matters more than how much you take.

Source: Cruz-Sanabria F et al. Journal of Pineal Research. 2024. PubMed 38888087

Glycine Before Bed: Modest, Specific, and Under-Tested

The cleanest recent test of bedtime glycine in humans is a 2023 crossover randomized controlled trial in European Journal of Nutrition. Thirteen physically active males with self-reported sleep complaints took 15 g of collagen peptides (delivering about 3.5 g of glycine) one hour before bed for seven nights, with polysomnography measuring outcomes. The dose was chosen to approximate the 3 g glycine dose used in earlier Japanese sleep research.

The primary hypothesized outcomes (sleep onset latency and sleep efficiency) did not significantly improve. However, the intervention significantly reduced the number of nighttime awakenings on polysomnography (21.3 versus 29.3, p equals 0.028) and improved next-morning Stroop test performance. The sample size is very small at 13, the population is all-male and athletic, and the active agent is a collagen peptide matrix rather than pure glycine. The honest read is that the recent direct human evidence for bedtime glycine remains thin.

Source: Thomas C et al. European Journal of Nutrition. 2023. PubMed 37874350

Magnesium L-Threonate: Cognition Yes, Objective Sleep No

A 2026 randomized controlled trial in Frontiers in Nutrition tested 2 g of Magtein (delivering 145 mg of elemental magnesium as L-threonate) versus placebo in 100 healthy adults aged 18 to 45 with dissatisfied sleep, for six weeks. The primary endpoint was cognition, not sleep. Magtein outperformed placebo on the NIH Toolbox Total Cognition Composite (p equals 0.043) and on reaction time (p equals 0.031). On sleep, self-reported daytime impairment improved (p equals 0.043) but the objective Oura Ring measurements of sleep disturbance, sleep efficiency, and restorative sleep showed no significant difference between groups.

The pattern is consistent with the L-theanine finding above: benefit is detectable on next-day function and on subjective daytime impact, but not on the underlying architecture of sleep itself. Worth noting: the lead investigator received funding through Clinical Research Australia, which conducts industry-supported trials. The paper is methodologically clean but the funding context warrants disclosure.

Source: Lopresti AL and Smith SJ. Frontiers in Nutrition. 2026. PubMed 41601871

What This Means for You

Reading across all six studies, the honest summary is this:

• If anxiety and rumination keep you awake: the evidence is strongest for ashwagandha (600 mg or higher of standardized root extract, daily for 8 weeks or longer) and L-theanine. Both work through reducing the cortisol and stress arousal that interfere with sleep onset, not by sedating you.
• If your dietary magnesium intake is low: magnesium bisglycinate at 200 to 250 mg of elemental magnesium daily has the cleanest small-effect signal in the recent literature. The benefit is most likely to land if you are currently under-consuming magnesium in your diet.
• If you are dealing with circadian misalignment: jet lag, shift work, or a late chronotype trying to wind back earlier. Melatonin at a low dose (1 to 4 mg) taken about 3 hours before your target sleep time has the strongest dose-response evidence. Higher doses are not better.
• If you are stacking everything at once: that is not what any of these studies tested. Combinations may help, but the evidence for the combination is much weaker than the evidence for individual ingredients matched to a specific cause.

What the Evidence Does Not Support

Three claims show up constantly in supplement marketing without solid recent backing. High-dose melatonin works better than low-dose: the 2024 dose-response analysis directly refutes this. Valerian reliably improves sleep: recent systematic reviews continue to find inconsistent randomized controlled trials, poor extract standardization, and significant risk of bias. Magnesium helps everyone sleep: the 2025 bisglycinate trial points to baseline dietary magnesium status as the key moderator. People meeting dietary requirements have minimal recent evidence of benefit.

What I Did Not Cover This Week

I deliberately stayed in the highest-quality systematic reviews and meta-analyses on the most-studied compounds. I did not cover apigenin, tart cherry, valerian individual extracts, CBD, or stacked branded sleep formulas. Some of those compounds may have mechanism-of-action research but lack the level of human trial evidence needed for the format of this post. We will get to them in future weeks as the literature develops.

This article is for informational purposes and is not medical advice. Sleep is complex and individual. If you have a diagnosed sleep disorder, take prescription medication, or are pregnant or breastfeeding, talk to a qualified healthcare practitioner before adding any supplement to your routine.

Sources

1. Schuster J et al. Magnesium Bisglycinate Supplementation in Healthy Adults Reporting Poor Sleep: A Randomized, Placebo-Controlled Trial. Nature and Science of Sleep. 2025.
2. Fatima K et al. Safety and efficacy of Withania somnifera for anxiety and insomnia: Systematic review and meta-analysis. Human Psychopharmacology: Clinical and Experimental. 2024.
3. Bulman A et al. The effects of L-theanine consumption on sleep outcomes: A systematic review and meta-analysis. Sleep Medicine Reviews. 2025.
4. Cruz-Sanabria F et al. Optimizing the Time and Dose of Melatonin as a Sleep-Promoting Drug: A Systematic Review of Randomized Controlled Trials and Dose-Response Meta-Analysis. Journal of Pineal Research. 2024.
5. Thomas C et al. Collagen peptide supplementation before bedtime reduces sleep fragmentation and improves cognitive function in physically active males with sleep complaints. European Journal of Nutrition. 2023.
6. Lopresti AL and Smith SJ. The effects of magnesium L-threonate (Magtein) on cognitive performance and sleep quality in adults: a randomised, double-blind, placebo-controlled trial. Frontiers in Nutrition. 2026.
7. Soh CY et al. The effect of glycine administration on physiological systems: a systematic review. GeroScience. 2023.

Related reading

• How to Sleep Better Without Melatonin

Shop Magnesium Bisglycinate