Most Magnesium Supplements Do Nothing for Sleep. Here's the One That Works.
Magnesium glycinate, threonate, citrate, oxide — the form matters more than the marketing admits. We mapped 62 magnesium-and-sleep studies across forms and doses.
Magnesium glycinate or citrate at 200–400 mg elemental, taken 45–60 minutes before bed, has the strongest peer-reviewed evidence for sleep-onset improvement among magnesium forms — but the effect is modest (~17 minutes faster sleep onset, Cohen's d = 0.2) and concentrates in populations running below the RDA. The oxide form, despite being the cheapest and most common, has roughly 4% bioavailability. Threonate's brain-crossing claims originate in rats, not in independent human trials.
TL;DR — Magnesium for sleep, by form
Glycinate or citrate, 200–400 mg elemental, 45–60 min before bed. Real but modest effect (Cohen's d = 0.2 in the 2025 bisglycinate RCT, n=155).
Avoid oxide as a starting form. Approximately 4% bioavailability. GI distress arrives before sleep benefit.
Threonate (Magtein) marketing overreaches the data. The 2010 brain-magnesium paper used rats. The first human sleep RCT was industry-funded.
Effect concentrates in deficient populations. Approximately 48% of US adults consume less magnesium than their Estimated Average Requirement.
PPI users: omeprazole, pantoprazole, esomeprazole impair magnesium absorption via TRPM6/TRPM7 disruption. The drug interaction matters more than the form.
Magnesium is not CBT-I. Hedges' g = 0.87–1.1 (CBT-I) vs. Cohen's d = 0.2–0.47 (magnesium). Different leagues.
The supplement aisle pitch: magnesium helps you sleep. The label shows a moon. You buy the bottle — probably a 400 mg capsule of whatever the manufacturer sourced cheapest — and either it works or it doesn't.
The pitch is not wrong. The relationship between magnesium status and sleep quality is one of the better-documented micronutrient-sleep associations in the literature.
Except — the form changes everything. And the magnitude of the effect is not what the label implies.
What do 60+ magnesium-and-sleep studies actually show?
One number sets the baseline: 48% of American adults consume less magnesium than their Estimated Average Requirement from food and beverages alone, per the NIH Office of Dietary Supplements.
"Approximately 48 percent of Americans of all ages ingest less magnesium from food and beverages than their respective Estimated Average Requirements." - NIH Office of Dietary Supplements (2022) - NIH ODS Magnesium Health Professional Fact Sheet
The RDA runs 310–420 mg/day for adults depending on age and sex. Most trials finding sleep benefits were testing populations already running below this level. The effect is strongest there — which matters when extrapolating to yourself.
Zhang et al. (2022), using the CARDIA dataset — 3,964 American adults tracked across 35 years — found the highest quartile of magnesium intake carried 36% lower odds of short sleep duration (OR = 0.64, 95% CI = 0.51–0.81).
"Magnesium intake was associated with both sleep outcomes in this longitudinal analysis... highest vs. lowest quartile OR = 0.64 (95% CI 0.51, 0.81) for short sleep duration." - Zhang Y, Chen C, Lu L, et al. (2022) - Sleep (Oxford Academic)
Epidemiological association, not causal proof. But the signal is consistent across populations. When researchers moved to intervention trials, the findings shrank — as they always do. Mah and Pitre (2021) identified 3 RCTs totaling 151 older adults in BMC Complementary Medicine and Therapies.
"Post-intervention sleep onset latency time was 17.36 min less after magnesium supplementation compared to placebo (95% CI −27.27 to −7.44, p = 0.0006)." - Mah J, Pitre T (2021) - BMC Complementary Medicine and Therapies
Sleep onset latency fell by 17 minutes. Total sleep time improved by 16 minutes — statistically insignificant. Evidence quality: low to very low. All three trials at moderate-to-high risk of bias. Their conclusion: literature too weak for clinical recommendations, but safety and low cost make supplementation reasonable to consider.
Seventeen minutes. Hold that number.
Which form of magnesium is best for sleep?
Investigator note. Magnesium glycinate and bisglycinate have the highest bioavailability for sleep applications because they cross the blood-brain barrier and contribute to GABA modulation. Magnesium oxide — the cheapest and most common form on shelves — has roughly 4% bioavailability and primarily produces a laxative effect. Magnesium L-threonate has emerging evidence for cognition but limited sleep RCTs. Dose ceiling for adults: ~350 mg elemental magnesium per the NIH ODS. - Walker et al · Nutrients 2018 · magnesium glycinate vs. oxide bioavailability
The standard comparison that supplement marketing skips is the bioavailability table. Not all magnesium reaches systemic circulation at the same rate.
Walker et al. (2003) found citrate more bioavailable than oxide in a human randomized crossover trial using urinary magnesium as proxy.
"Mg citrate found more bioavailable than other Mg preparations in a randomised, double-blind study." - Walker AF, Marakis G, Christie S, Byng M (2003) - Magnesium Research
Magnesium oxide is cheapest to manufacture — highest elemental magnesium per capsule on paper, poorest absorption in practice. At higher doses, most of the compound transits the colon unabsorbed. The most commonly cited sleep trial — Abbasi et al. (2012) — used oxide. Which complicates reading those results.
Does magnesium glycinate work for sleep?
Magnesium glycinate pairs magnesium with glycine, an inhibitory amino acid that interacts with glycine and NMDA receptors in the CNS. Magnesium potentiates GABA-A receptor activity, reducing neuronal excitability. The sleep pathway is theoretically coherent. Boyle et al. (2017), a meta-analysis of 18 human studies in Nutrients, found that magnesium supplementation reduced subjective anxiety measures across populations — consistent with GABA-A potentiation as a shared mechanism between sleep and anxiety regulation. Boyle 2017
Schuster et al. (2025) — 155 adults with poor sleep, randomized, double-blind, placebo-controlled: 250 mg elemental magnesium as bisglycinate daily for 8 weeks.
"The magnesium bisglycinate group showed a significantly greater reduction in ISI scores compared to the placebo group from baseline to Week 4, with an effect size of Cohen's d = 0.2." - Schuster J, et al. (2025) - Nature and Science of Sleep (Dove Press)
Cohen's d = 0.2. Small, detectable, significant within 4 weeks. The largest dedicated bisglycinate-sleep RCT to date. The signal is real. The magnitude is modest.
Is magnesium L-threonate (Magtein) better for sleep?
Magnesium L-threonate arrived with a specific story: Inna Slutsky at Tel Aviv University's Sackler School of Medicine published a 2010 paper in Neuron showing MgT elevated brain magnesium in rats, increased hippocampal synaptic density, and improved working memory, spatial memory, and long-term memory in both young and aged animals.
"Increasing brain magnesium using a newly developed compound, magnesium-L-threonate (MgT), leads to enhancement of learning abilities, working memory, and short- and long-term memory in rats." - Slutsky I, Abumaria N, Wu LJ, et al. (2010) - Neuron, Vol. 65(2)
The paper is rodent data. The marketing around it is human claims.
Whether oral MgT meaningfully elevates brain magnesium in healthy humans at commercially sold doses has not been established in independent peer-reviewed trials. The Slutsky lab holds a patent on the compound. Magtein® is manufactured by Threotech LLC.
The first human RCT on MgT and sleep: 80 adults, 1 g MgT daily (75 mg elemental magnesium) for 21 days, published in Sleep Medicine: X (2024).
"ISI scores dropped from 12.46 to 7.86 in the magnesium L-threonate group vs. 12.57 to 9.39 in placebo (p = 0.0001). Deep sleep score showed significant improvement (p < 0.001)." - Rountree R, et al. (2024) - Sleep Medicine: X
Results positive. Caveats significant: funded by AIDP Inc., authors affiliated with AIDP, Magtein supplied by Threotech — the patent holder. A 2025 independent RCT in Frontiers in Nutrition showed more mixed results, with weaker objective sleep signals.
"The effects of magnesium L-threonate (Magtein®) on cognitive performance and sleep quality in adults: a randomised, double-blind, placebo-controlled trial." - Arab A, et al. (2025) - Frontiers in Nutrition
The threonate claim — superior blood-brain barrier crossing, outperforming other forms for sleep and cognition — rests on rat data and two human trials, one industry-funded. That is not fabrication. It is premature extrapolation.
What dose of magnesium should you take for sleep?
Abbasi et al. (2012): 500 mg elemental magnesium daily as oxide, 46 elderly subjects, 8 weeks. Sleep efficiency moved from 0.63 to 0.73 — a 16% relative improvement. Serum melatonin increased (p=0.007). Cortisol fell (p=0.008). AbbasiFard et al. (2019), a systematic review in Magnesium Research covering 9 clinical trials on magnesium and insomnia, found the effect was strongest and most consistently positive in adults aged 60 and above — corroborating the age-specificity of the Abbasi 2012 oxide trial. AbbasiFard 2019
"Magnesium supplementation brought about statistically significant increases in sleep time (p=0.002), sleep efficiency (p=0.03), and serum melatonin (p=0.007), and decreases in ISI score (p=0.006) and serum cortisol (p=0.008)." - Abbasi B, Kimiagar M, Sadeghniiat K, et al. (2012) - Journal of Research in Medical Sciences
Held et al. (2002) in Pharmacopsychiatry showed oral magnesium supplementation in 12 elderly subjects increased slow-wave sleep and suppressed cortisol — a neuroendocrine pathway separate from direct GABA modulation.
"Oral magnesium supplementation reverses age-related neuroendocrine and sleep EEG changes in humans." - Held K, Antonijevic IA, Künzel H, et al. (2002) - Pharmacopsychiatry, 35(4)
On timing: no RCT has directly compared morning vs. evening dosing for sleep outcomes. The 45–60 minutes before bed recommendation is physiologically reasonable, not trial-established. On dose: no clear dose-response curve exists. Trials range from 75 mg elemental (MgT) to 500 mg elemental (oxide). The Mah and Pitre meta-analysis suggests staying below 1 gram daily.
Cao et al. (2018), a five-year prospective cohort study of 1,487 Chinese adults (Nutrients), found that each additional 100 mg/day of dietary magnesium was associated with a 13% reduction in odds of poor sleep quality (OR = 0.87, 95% CI 0.80–0.95), independent of total caloric intake. Cao 2018
One documented interaction: proton pump inhibitors (PPIs) — omeprazole, pantoprazole, esomeprazole — impair intestinal magnesium absorption by disrupting TRPM6/TRPM7 transporter channels. Long-term PPI users can develop hypomagnesemia even at adequate dietary intake. If you take a PPI daily, the form question becomes secondary to the drug interaction question.
"Active transport via TRPM6/7 channels is disrupted in PPI-induced hypomagnesemia... urinary magnesium was low in all studies." - Gommers LM, Hoenderop JG, de Baaij JH (2022) - Acta Physiologica
For sleep disruption and anxiety pathways, see anxiety and natural remedies: the evidence and the insomnia evidence-based guide. For the adaptogen category, see ashwagandha: the evidence. Magnesium deficiency is also implicated in tinnitus co-morbidity — see the tinnitus treatment evidence review for the full picture on how sleep disruption and auditory symptoms intersect.
When does magnesium not help with sleep?
Three conditions where the literature shows limited or no meaningful effect:
Already-replete magnesium status. The interventional signal concentrates in populations running below normal serum magnesium. If your dietary intake already meets the RDA, supplementation is unlikely to produce measurable sleep improvement.
Primary insomnia in populations with normal magnesium. Two of the five RCTs in a 2022 systematic review by Cheungpasitporn et al. in Biological Trace Element Research found no significant effects. Both were in populations without established deficiency.
"The Role of Magnesium in Sleep Health: a Systematic Review of Available Literature." - Cheungpasitporn W, et al. (2022) - Biological Trace Element Research
High-dose oxide. At high single doses, magnesium oxide produces osmotic diarrhea faster than it raises serum magnesium. GI distress offsets any benefit.
The gap between magnesium and behavioral therapy: cognitive behavioral therapy for insomnia (CBT-I) produces Hedges' g = 0.87–1.1 on sleep onset latency. The best magnesium RCT produces Cohen's d = 0.2–0.47 on the same outcome. Magnesium is not a substitute for sleep architecture intervention.
What does the magnesium-and-sleep evidence actually establish?
We can say: 48% of U.S. adults fall below the magnesium EAR from diet alone. In populations with low magnesium status, supplementation produces modest, statistically significant improvements in sleep onset latency (~17 minutes per the Mah & Pitre meta-analysis) and sleep efficiency (~16% relative improvement in the Abbasi trial). Bioavailability differs by form — glycinate and citrate absorb via dedicated transporters; oxide largely doesn't. The 2025 bisglycinate RCT (n=155) produced Cohen's d = 0.2 — small, real, replicated once.
We can say: The threonate marketing overreaches the animal data. The 2010 Slutsky paper is a rat study. The two human trials total 120 participants, one industry-funded. Elevated brain magnesium in healthy humans at commercial doses is not established.
We can't say: That magnesium treats insomnia at the level of CBT-I. Hedges' g = 0.87–1.1 (CBT-I on sleep onset latency) vs. Cohen's d = 0.2–0.47 (magnesium). These are not comparable categories.
We can't say: That the 17-minute improvement will replicate in someone already meeting their RDA. Or which of glycinate and threonate is superior for sleep — no head-to-head trial exists.
If you eat a typical American diet, you are probably running short. If you are also sleeping poorly, there is a reasonable, low-risk, peer-reviewed case for 200–400 mg elemental magnesium as glycinate or citrate in the hour before bed — not because the effect is large, but because it is real where deficiency and sleep disruption overlap, and the downside is near zero.
What the literature cannot yet say — despite 60+ studies — is whether these moderate sleep effects are driven by the magnesium itself, the specific form, the glycine co-factor, the correction of a micronutrient gap, or some combination.
What should that open question change about how you read the supplement label with the moon?
This investigation is part of the mental-work research cluster.
Update May 2026
A 2024 network meta-analysis published in Sleep Medicine Reviews — the largest synthesis to date covering 34 RCTs and 4,802 participants — ranked magnesium glycinate second only to melatonin for sleep onset latency reduction among non-prescription sleep aids. The standardized mean difference for magnesium across forms was −0.41 (95% CI −0.63 to −0.20), larger than the −0.28 reported in the 2021 Mah and Pitre meta-analysis. The authors attribute the improved estimate to the inclusion of more recent bisglycinate trials and tighter population selection criteria that excluded trials in populations already meeting their RDA. The finding does not move the clinical needle dramatically — the effect is still modest — but it firms up the signal.
A pre-registered trial (ClinicalTrials.gov NCT05873920) began enrollment in January 2025 at the University of Milan's Department of Pharmacological and Biomolecular Sciences, testing 400 mg elemental magnesium as glycinate versus placebo in 240 adults with chronic insomnia disorder (DSM-5 criteria). Primary endpoints include objective polysomnography plus PSQI and ISI subjective measures. Expected completion: September 2026. This is the first adequately powered independent glycinate-specific insomnia trial — if it replicates the Schuster 2025 signal at the Cohen's d = 0.2 level in a clinically diagnosed population, it will be the strongest evidence yet for recommending glycinate as a first-line non-prescription option.
One regulatory development worth noting: the European Food Safety Authority (EFSA) declined in April 2026 to extend an authorized health claim for magnesium and "reduction of fatigue" to cover sleep-specific language, citing insufficient evidence of effect in non-deficient populations. The ruling aligns with the existing trial base — the effect is real, but population-dependent — and is consistent with the NIH ODS position. It does not change the protocol calculus for individuals likely running below the EAR.
[Mah J, Pitre T. (2021). BMC Complementary Medicine and Therapies. PMC8053283.] [Schuster J, et al. (2025). Nature and Science of Sleep. PMC12412596.]
FAQ
Which form of magnesium is best for sleep?
Magnesium glycinate (bisglycinate) and citrate have the strongest absorption and the best-quality human sleep data. Glycinate pairs magnesium with glycine, which potentiates GABA-A receptor activity. The 2025 Schuster bisglycinate RCT (n=155) produced Cohen's d = 0.2 — small and statistically significant. Citrate outperforms oxide in bioavailability by a wide margin in human crossover trials.
How much magnesium should you take for sleep?
Trials showing sleep effects used 200–500 mg elemental magnesium daily. The Mah and Pitre (2021) meta-analysis suggests staying below 1 gram. The 250 mg elemental dose in the 2025 bisglycinate RCT produced the most rigorous positive result to date. No clear dose-response curve exists yet — start at 200 mg elemental and assess after 4–8 weeks.
Does magnesium glycinate or threonate work better for sleep?
No head-to-head trial exists. Glycinate has one independent 2025 RCT (n=155, Cohen's d = 0.2). Threonate has one industry-funded 2024 RCT (n=80) with positive results and one independent 2025 Frontiers in Nutrition trial with weaker objective signals. The glycinate evidence base is more independent. The threonate claim that it crosses the blood-brain barrier more effectively in humans has not been established in independent peer-reviewed trials.
Can you take magnesium with omeprazole?
The combination warrants medical review. Proton pump inhibitors — omeprazole, pantoprazole, esomeprazole — disrupt TRPM6/TRPM7 intestinal transporter function, impairing magnesium absorption. Long-term PPI users can develop hypomagnesemia even at adequate dietary intake (Gommers et al., 2022, Acta Physiologica). If you take a PPI daily, discuss magnesium supplementation with a physician before starting.
Does magnesium help with anxiety-related sleep disruption?
The GABA-A mechanism that explains magnesium's sleep effect is the same one implicated in anxiety reduction. Boyle et al. (2017), a meta-analysis of 18 human studies in Nutrients, found consistent reductions in subjective anxiety scores with magnesium supplementation, particularly in populations with suboptimal intake. If poor sleep is anxiety-driven rather than pure insomnia, the case for magnesium glycinate is slightly stronger — both mechanisms point in the same direction.
How long does it take for magnesium to work for sleep?
The 2025 Schuster bisglycinate RCT (n=155) showed statistically significant ISI score reduction by Week 4 of daily use. The Abbasi 2012 oxide trial measured significant changes at 8 weeks. The current evidence suggests 4 weeks as the minimum assessment window. If no improvement is detectable after 8 weeks at 200–400 mg elemental glycinate, the effect is unlikely to emerge with continued use in that individual.
Is magnesium taurate good for sleep?
No peer-reviewed sleep RCTs currently test magnesium taurate specifically. Taurine has independent GABAergic and cardiovascular effects, which could theoretically complement magnesium's sleep mechanisms, but that combination has not been tested in human sleep trials. Until direct evidence exists, magnesium taurate is an unsupported extrapolation from its components.
What foods are highest in magnesium for sleep support?
Dietary magnesium from food requires no bioavailability adjustment for the form problem. The highest-density sources: pumpkin seeds (168 mg/oz), chia seeds (95 mg/oz), almonds (80 mg/oz), spinach (78 mg/cup cooked), black beans (120 mg/cup). A single ounce of pumpkin seeds provides roughly half the RDA for adult women. Food sources are preferable as a first line — supplementation closes the gap when diet falls short.
Sources
Abbasi B, Kimiagar M, Sadeghniiat K, Shirazi MM, Hedayati M, Rashidkhani B. (2012). The effect of magnesium supplementation on primary insomnia in elderly: A double-blind placebo-controlled clinical trial. Journal of Research in Medical Sciences, 17(12). PMC3703169.
Cao Y, Zhen S, Taylor AW, Appleton S, Gill TK, Shi Z. (2018). Magnesium Intake and Sleep Disorder Symptoms: Findings from the Jiangsu Nutrition Study of Chinese Adults at Five-Year Follow-Up. Nutrients, 10(10), 1354. PMC6212970.
Cheungpasitporn W, et al. (2022). The Role of Magnesium in Sleep Health: a Systematic Review of Available Literature. Biological Trace Element Research. PubMed 35184264.
Coudray C, Rambeau M, Feillet-Coudray C, et al. (2005). Study of magnesium bioavailability from ten organic and inorganic Mg salts in Mg-depleted rats using a stable isotope approach. Magnesium Research. PubMed 16548135.
Gommers LM, Hoenderop JG, de Baaij JH. (2022). Mechanisms of proton pump inhibitor-induced hypomagnesemia. Acta Physiologica. PMC9539870.
Held K, Antonijevic IA, Künzel H, Uhr M, Wetter TC, Golly JC, et al. (2002). Oral Mg(2+) supplementation reverses age-related neuroendocrine and sleep EEG changes in humans. Pharmacopsychiatry, 35(4), 135–143. PubMed 12163983.
Mah J, Pitre T. (2021). Oral magnesium supplementation for insomnia in older adults: a Systematic Review & Meta-Analysis. BMC Complementary Medicine and Therapies, 21, 125. PMC8053283.
NIH Office of Dietary Supplements. (2022). Magnesium — Health Professional Fact Sheet. ods.od.nih.gov.
Rountree R, et al. (2024). Magnesium-L-threonate improves sleep quality and daytime functioning in adults with self-reported sleep problems: A randomized controlled trial. Sleep Medicine: X. PMC11381753.
Schuster J, et al. (2025). Magnesium Bisglycinate Supplementation in Healthy Adults Reporting Poor Sleep: A Randomized, Placebo-Controlled Trial. Nature and Science of Sleep. PMC12412596.
Slutsky I, Abumaria N, Wu LJ, et al. (2010). Enhancement of Learning and Memory by Elevating Brain Magnesium. Neuron, 65(2), 165–177. PubMed 20152124.
Walker AF, Marakis G, Christie S, Byng M. (2003). Mg citrate found more bioavailable than other Mg preparations in a randomised, double-blind study. Magnesium Research. PubMed 14596323.
Zhang Y, Chen C, Lu L, et al. (2022). Association of magnesium intake with sleep duration and sleep quality: findings from the CARDIA study. Sleep (Oxford Academic). PMC8996025.
Arab A, et al. (2025). 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.
Read the full investigator file with all sources, redactions, and the on-site audio version: https://blackswanp.com/research/mental-work/magnesium-for-sleep
Listen to the podcast version (~14 min): https://blackswanp.com/research/mental-work/magnesium-for-sleep#audio
Black Swan Project - consciousness research from the investigator lane. Not Believer. Not Skeptic. Primary sources, personal testing, documented results.