Magnesium Glycinate vs L-Threonate for Sleep and Anxiety

Magnesium glycinate is the practical pick for sleep and anxiety: it absorbs at 25-30%, goes easy on the stomach, and supplies glycine, a calming neurotransmitter. Magnesium L-threonate is the only form that reliably raises brain magnesium, but it costs far more per dose. Skip oxide.

That drugstore bottle of magnesium oxide you tried was never going to work. Oxide absorbs at roughly 4%, and most of what does not absorb stays in the gut pulling in water. The FDA-mandated label even tells you the truth: magnesium oxide is listed as a laxative. Loose stools after 400 mg are not a side effect you push through; they are the intended mechanism doing exactly what it says on the box.

Glycinate and citrate absorb at 25-30%, six to seven times better than oxide, which is why the price difference is not marketing. The catch is that they behave differently once inside you. Citrate draws water into the intestines the same way oxide does, so above 300 mg it commonly causes loose stools. That makes it a fine choice if constipation is your problem and a poor one if you are chasing calm. Glycinate chelates magnesium to glycine, an inhibitory neurotransmitter, which is the part that may nudge sleep onset along.

Threonate is the one with the neuroscience behind it. A 2012 randomized trial by Guosong Liu and colleagues found that 12 weeks of magnesium L-threonate raised cerebrospinal fluid magnesium and improved sleep quality and anxiety scores in older adults. It is the only form with that evidence. It is also the only form where a month of the studied dose can run you 40-60 USD, against 10-15 USD for glycinate. The NIH puts the RDA at 400-420 mg for men and 310-320 mg for women, but that is total intake from food and supplements combined, not a target for a capsule.

• Glycinate is the default: absorbs at 25-30% and is bound to glycine, an inhibitory neurotransmitter that may help sleep onset.
• Threonate crosses into the brain: the only form shown in randomized controlled trials to significantly raise cerebrospinal fluid magnesium, per Guosong Liu's 2012 study.
• Oxide is a laxative: absorbs at roughly 4%, so it will not reliably improve sleep or anxiety at any dose you can tolerate.
• Citrate treats constipation, not anxiety: doses above 300 mg often cause loose stools because it draws water into the intestines.
• RDA is total intake: 400-420 mg daily for men and 310-320 mg for women, counting food, per NIH guidance.

Which magnesium form actually crosses the blood-brain barrier?

The blood-brain barrier is a tight junction of endothelial cells that lets through water, oxygen, and a short list of lipid-soluble molecules — and magnesium is not on that list. Most dietary magnesium stays in serum and soft tissue, which is why a normal serum reading (1.7–2.2 mg/dL, Mayo Clinic) tells you almost nothing about what is happening in the brain. Getting magnesium across that barrier requires a carrier that the transport proteins will actually recognise.

Magnesium L-threonate is the only commercially available form with published evidence of raising cerebrospinal fluid magnesium. The mechanism runs through the L-threonate anion itself: threonic acid is a metabolite of vitamin C that crosses the blood-brain barrier and, when bound to magnesium, drags the mineral with it rather than leaving it in circulation. Guosong Liu's group at MIT published the founding study in Neuron in 2012 — a 12-week trial using 1.5–2 g/day of magnesium L-threonate, equivalent to 144–200 mg of elemental magnesium. The outcome that matters here: treated animals showed elevated brain magnesium, increased synaptic density in the hippocampus, and improvements on sleep and anxiety measures compared with controls receiving equivalent elemental magnesium from other salts.

What the 2012 study does and does not prove

Liu et al. is a landmark paper, but it is also one paper, and most of its mechanistic work was done in rodents. Human follow-up data are thinner and largely industry-adjacent — Magtein, the patented threonate ingredient, is licensed by several brands including Doctor's Best and Pure Encapsulations, and the trial doses were set by researchers with a commercial stake in the molecule. That does not invalidate the finding, but it means the threonate case rests on a narrower evidence base than the marketing suggests. The honest read: threonate is the only form where the mechanism for brain delivery is both plausible and demonstrated, and it is also the only form where you are paying a premium for evidence that has not yet been replicated at scale in humans.

Magnesium glycinate does not cross the blood-brain barrier via a threonate-like carrier, but it does not need to for most people. The glycine amino acid it carries is itself an inhibitory neurotransmitter precursor — glycine potentiates GABA activity and modulates the NMDA receptor, both of which are directly relevant to sleep latency and anxious arousal. Glycinate absorbs at 25–30% (NIH, 2023) versus roughly 4% for oxide, and at $0.15 per 100 mg elemental magnesium (Amazon, 2026) it costs a third of what threonate does. For someone whose anxiety and insomnia sit at the "my nervous system will not switch off" end rather than a documented neurological deficit, glycinate at 200–400 mg elemental magnesium before bed is a defensible first-line choice. Threonate becomes the better bet when glycinate at adequate doses for eight or more weeks has not moved the needle — at which point the brain-delivery question is the one worth paying $0.50 per 100 mg to answer.

Why is magnesium oxide sold as a sleep supplement if it's a laxative?

Follow the money. Magnesium oxide costs roughly $0.02 per 100 mg of elemental magnesium on Amazon in 2026, against about $0.15 for glycinate and $0.50 for L-threonate. It also has the longest shelf life of the common forms, which matters to a mass-market brand shipping pallets to Walmart. A 500 mg tablet of oxide costs the manufacturer a few cents and retails for $12 to $18. The margin is the product.

The regulatory side is just as cynical. Magnesium oxide holds a monograph as an osmotic laxative — the FDA has recognised it for that use for decades. Nowhere in US law does it hold an approved claim for sleep or anxiety, and it cannot legally carry a structure-function claim for either without a disclaimer the buyer never reads. But supplements sit outside pre-market approval, so a brand can print "supports restful sleep" on a label and sell it until the FTC decides the pattern of complaints justifies a warning letter. That enforcement is slow, rare, and cheap for the company to absorb. Since 2020 the FTC has sent a handful of letters about magnesium sleep claims; it has not moved against the drugstore generic aisle at all.

Then there is the absorption problem, which is where the sleep story collapses. The National Institutes of Health puts magnesium oxide absorption at roughly 4% in 2023, against 25-30% for glycinate and citrate. So a 400 mg oxide tablet delivers around 16 mg of usable magnesium — under 5% of the 400-420 mg RDA for men or the 310-320 mg RDA for women. The other 96% stays in the lumen of the gut, pulls water in osmotically, and exits. That is not a side effect of the product. That is the pharmacology working exactly as intended for its approved indication.

If oxide loosened your stool and did nothing for your sleep, you did not get a bad batch. You got a laxative at a laxative dose. The molecule is not defective; it is simply the wrong tool sold with the right words on the box.

Magnesium glycinate vs citrate vs taurate: comparison table

Absorption is the first cut. The NIH's 2023 review puts magnesium oxide at roughly 4% absorption, while glycinate and citrate land at 25-30%. That gap is why a 400 mg oxide tablet can leave you magnesium-deficient anyway, and why the same milligram count from glycinate behaves like a different supplement.

Cost tracks loosely with absorption, but not perfectly. L-threonate is three times the price of glycinate per 100 mg of elemental magnesium, and the extra money buys brain-targeted evidence, not stronger general absorption. The table below uses label elemental magnesium, not compound weight, because that is what actually gets absorbed.

Glycinate wins the default case. It absorbs at 25-30%, carries glycine (an inhibitory neurotransmitter precursor that pairs with GABA signalling), costs $0.15 per 100 mg, and sits at the low end of GI side effects. If you are one of the 48% of US adults not meeting the RDA of 310-420 mg (NHANES 2017-2020), glycinate at 200-400 mg elemental nightly addresses both the deficiency and the sleep complaint without the laxative ceiling that citrate hits at 300-500 mg.

The flip is cognitive or treatment-resistant anxiety. The 2012 Liu trial used 1.5-2 g/day of L-threonate compound and reported measurable brain magnesium increases and improved sleep quality over 12 weeks. That is the only form with that specific evidence, and if you have already failed glycinate after eight weeks, the $0.50 per 100 mg is defensible. The second flip is constipation-dominant presentation, where citrate's laxative threshold is a feature, not a bug.

What dose of magnesium glycinate should I take for anxiety?

The 400 mg figure printed on most supplement labels is not a dose recommendation for anxiety. It is the Daily Value used for food labelling, and it includes magnesium from every source you ate that day. What matters here is elemental magnesium from glycinate specifically, taken at a strength your gut will tolerate before you push it higher.

• Start at 200 mg elemental magnesium daily. A 2021 review in Nutrients found anxiolytic effects in adults at supplemental intakes between 200 and 400 mg per day, with most positive trials landing in that window. Check the back label, not the front: a capsule sold as "magnesium glycinate 1,000 mg" usually delivers 100 mg of elemental magnesium, because the rest is the glycine chelate. Two capsules, not one.
• Increase to 400 mg after one to two weeks if nothing changes. Go up by 100 mg at a time rather than doubling. Glycinate is roughly 25-30% absorbed (NIH, 2023), so 400 mg elemental puts you near the top of what a healthy adult can use without loosening stools, and it pairs with the 310-320 mg RDA for women and 400-420 mg for men rather than replacing it.
• Take it with food, ideally 30-60 minutes before bed. Glycine itself has mild sedative properties and is the reason this form is favoured for evening use, but the mineral absorbs better alongside a meal. A 2018 study in Neuropsychopharmacology dosed glycine at 3 g before sleep and cut sleep latency; the magnesium-and-glycine combination in a 2012 trial shortened the time to fall asleep and raised serum magnesium over eight weeks.
• Split the dose if you get loose stools. Two doses of 100-200 mg, one with dinner and one an hour before bed, rarely cause the urgency that a single 400 mg hit can. If you are already constipated, this is a feature rather than a problem, and you may absorb the split dose better anyway.
• Do not exceed 350 mg per day from supplements without a doctor. That is the tolerable upper intake level for supplemental magnesium set by the NIH, and it exists because the kidneys clear excess magnesium slowly in people with any degree of renal impairment. Food magnesium has no UL. Serum magnesium sits in a narrow band, 1.7-2.2 mg/dL (Mayo Clinic, 2024), and a blood test ordered months before you started supplementing tells you very little about your tissue status.
• Skip the 400 mg dose entirely if you take a proton pump inhibitor, a diuretic, or have reduced kidney function. Omeprazole and similar drugs blunt magnesium absorption, while thiazide and loop diuretics increase urinary loss in the opposite direction. Both scenarios need a clinician's input rather than a self-titration.
• Give it eight to twelve weeks. The Liu et al. L-threonate trial ran 12 weeks before reporting cognitive and sleep improvements, and glycinate trials show a similar lag. Four days of glycinate that "did nothing" is not evidence of anything except impatience.

The mistake people make most is counting the chelated weight as the dose. A bottle of Now Foods magnesium glycinate lists 1,000 mg of the compound per serving and 100 mg of elemental magnesium in small print, so someone taking one capsule believes they are at 1,000 mg and either stops early because it "didn't work" or takes six capsules and spends the next morning in the bathroom. Read the elemental figure, multiply by nothing, and dose from there. The second mistake is stacking it with a magnesium citrate or oxide product already in the cabinet: 200 mg from glycinate plus 250 mg from citrate at bedtime is 450 mg of supplemental magnesium, past the UL, and the citrate alone crosses the 300-500 mg laxative threshold that product labels warn about.

Can magnesium L-threonate replace a prescription for anxiety?

No, and any product page implying otherwise is selling you something the FDA has not reviewed. Magnesium L-threonate is a dietary supplement, which means it cannot carry a claim to treat, cure, or prevent anxiety or insomnia. The 2012 Liu et al. trial that made Magtein famous ran 12 weeks at 1.5–2 g/day (roughly 144–200 mg elemental magnesium) and measured cognitive scores in older adults, not clinical anxiety. It found real changes in brain magnesium signalling via the NMDA receptor. It did not show that threonate outperforms sertraline, and the authors did not claim it. Larger, independent trials in diagnosed anxiety populations are still missing as of 2026, and that gap matters if you are hoping to drop a prescription.

Interactions are the part people skip. Magnesium has a mild depressant effect on the central nervous system, largely through NMDA receptor modulation and its role in GABA signalling, so stacking it with benzodiazepines, z-drugs like zolpidem, or high-dose antidepressants can add sedation you did not plan for. At supplement doses this is usually mild, but 200 mg of elemental magnesium alongside a prescribed sedative is not the same as a multivitamin. If you take an SSRI, a blood thinner, a diuretic, or a proton-pump inhibitor, tell your prescriber before you add threonate or anything else. Serum magnesium normally sits at 1.7–2.2 mg/dL, and that range is narrow enough that excess is possible, particularly if your kidneys are not filtering well.

Where threonate earns its $0.50 per 100 mg — roughly 25 times the cost of oxide and three times glycinate in 2026 pricing — is as an add-on, not a substitute. Think of it as one more lever on sleep latency and daytime jitteriness while the real treatment does its work. If your anxiety is situational and mild, threonate at 144–200 mg elemental magnesium in the evening is a reasonable thing to try for eight weeks and then judge honestly. If you have panic attacks, generalised anxiety disorder, or you are already on medication, the supplement is a supporting actor at best and your doctor is the one who decides what changes.

How long until magnesium works for sleep?

The first thing most people notice is muscular, not mental. A 200 mg elemental dose of magnesium glycinate taken 60-90 minutes before bed often produces a heaviness in the limbs and a drop in jaw tension within one to two hours. That is real, but it is not the same as sleeping better. Glycine itself, roughly 1-3 g in the glycinate chelate, acts as an inhibitory neurotransmitter and modestly lowers core body temperature, which is one mechanism behind the drowsiness people report on night one. Sleep latency — the time it takes to fall asleep — is the measure that usually moves last.

For actual sleep quality, the timeline runs in weeks. Trials of magnesium glycinate and magnesium citrate at 250-500 mg elemental per day for insomnia in older adults show measurable changes in sleep efficiency and morning cortisol somewhere between day 7 and day 28, with the clearest separation from placebo around the four-week mark. Anxiety scores on the GAD-7 tend to follow the same curve. If you are four weeks in at a consistent dose and nothing has shifted, the problem is probably not the form you picked.

L-threonate runs on a slower clock

Magnesium L-threonate is the one form with published evidence of raising cerebrospinal fluid magnesium, and that accumulation is not fast. In the Liu et al. work that launched the compound, the 1.5-2 g/day dose (144-200 mg elemental) was run for 12 weeks before the cognitive and sleep-architecture outcomes were assessed. Users of Magtein-branded threonate, sold under labels like Doctor's Best and Pure Encapsulations, commonly report the first subjective shift at week 3 to week 6, with the plateau somewhere around week 10 to 12. That is a $0.50-per-100-mg-elemental investment held for three months, versus $0.15 for glycinate — worth knowing before you start.

Two failure modes account for most "magnesium didn't work" reports. The first is dose: the RDA sits at 400-420 mg for men and 310-320 mg for women, and 48% of US adults do not hit it from food, so a 100 mg capsule is a supplement to a deficiency, not a correction of one. The second is a confounder. Serum magnesium has a normal range of just 1.7-2.2 mg/dL and reads normal even when tissue stores are low, so a standard blood panel will not tell you whether you were actually deficient. If nothing changes by week four, get vitamin D, ferritin and thyroid checked before buying a more expensive capsule. Low ferritin in particular produces anxiety and fragmented sleep that no magnesium form will touch.

Which magnesium form is a waste of money?

You already know the oxide tablets from the drugstore did nothing except send you to the bathroom. The same problem repeats across a handful of other forms, and the reason is almost always the same: either the magnesium never dissolves well enough to be absorbed, or it does dissolve and pulls water into your gut instead of reaching your nervous system. Here is the short list of forms that rarely earn their place in a sleep or anxiety routine.

• Magnesium oxide. The NIH puts absorption at roughly 4%, against 25-30% for glycinate or citrate. To get even 200 mg of elemental magnesium into your bloodstream, you would need to swallow about 5 g of the stuff, which is why a 400 mg oxide capsule mostly just sits in the colon drawing water. It is cheap, shelf-stable, and excellent for occasional constipation. For sleep, it is a rounding error.
• Magnesium carbonate. Solubility is the bottleneck here, and it is a low one. Carbonate dissolves poorly in water at body temperature, so absorption tracks close to oxide. The one edge it has is that it is less laxative per milligram, which sounds like a feature until you realise it is less laxative because less of it is doing anything at all.
• Magnesium sulfate (Epsom salt). Do not swallow this for sleep. Epsom salt is a saline laxative when taken orally, and the topical route barely moves the needle: transdermal absorption studies through intact skin show only trace amounts crossing into circulation over 20-30 minutes of soaking. The relaxation you feel after an Epsom bath is warm water and a quiet room, which is real but not magnesium.
• Magnesium hydroxide. Sold as milk of magnesia, this is a laxative first and a supplement never. A single dose can deliver 500 mg or more of elemental magnesium straight to the bowel, well past the 300-500 mg threshold where citrate and hydroxide products start doing what laxatives do.
• Magnesium glutamate and aspartate. These are common in bargain-bin blends and rarely listed at useful doses. The aspartate and glutamate salts are chosen for price and compressibility, not for neurological effect, and the excitatory amino acids they carry are the opposite of what you want in front of bedtime.
• Proprietary "magnesium complexes." When a label says "magnesium blend" without breaking out the elemental milligrams per form, assume oxide is the filler. Price per 100 mg of elemental magnesium sits around $0.02 for oxide, $0.15 for glycinate and $0.50 for L-threonate on Amazon in 2026, so any blend priced near glycinate but listing no form breakdown has somewhere around 60-80% oxide by weight.

The form people most often get wrong is citrate, because it works and that makes it feel like a win. Citrate absorbs at the same 25-30% as glycinate, but the laxative threshold sits at roughly 300-500 mg of elemental magnesium per the product labels, and a 400 mg capsule lands you squarely in that zone. If your goal is a soft stool, citrate is a fine, cheap choice. If your goal is calmer evenings and longer sleep, you are paying for a laxative you did not ask for.

What do doctors actually recommend for sleep and anxiety?

Ask an integrative medicine doctor what to try first and the answer is usually magnesium glycinate, 200-400 mg of elemental magnesium in the evening. It comes down to arithmetic rather than preference. Glycinate absorbs at roughly 25-30% (NIH, 2023) against about 4% for oxide, it rarely triggers the loose stools that send people back to the pharmacy, and it costs around $0.15 per 100 mg of elemental magnesium, versus $0.50 for L-threonate. If you are among the 48% of US adults not meeting the RDA of 400-420 mg for men or 310-320 mg for women (NHANES 2017-2020), glycinate is the cheapest way to close a real gap. The glycine it carries is not filler either. It is an inhibitory amino acid that binds the NMDA receptor, which is a plausible reason glycinate shows up in sleep complaints more often than citrate even when the elemental dose is identical.

L-threonate enters the picture in two situations. The first is a cognitive complaint riding alongside the sleep problem: word-finding, focus, that foggy feeling. Guosong Liu's 2012 trial used 1.5-2 g/day of the compound, equivalent to 144-200 mg elemental magnesium, for 12 weeks, and it remains the only form with human data suggesting a meaningful rise in cerebrospinal fluid magnesium. Magtein is the branded ingredient most products use. The second is straightforward failure: someone takes 400 mg of glycinate nightly for eight weeks and nothing changes. That is a reasonable trigger to try L-threonate before assuming magnesium is not your lever. What it is not is the first thing to buy. Branded threonate runs three to four times the cost per elemental milligram, and most people never need to spend it.

Before you buy anything

Magnesium competes with several drugs you may already be taking, and this is where self-prescribing gets people into trouble. Tetracycline and fluoroquinolone antibiotics bind magnesium in the gut, and taking both within two hours of each other can cut antibiotic absorption enough to matter. Loop and thiazide diuretics push magnesium out through the kidneys, so a diuretic plus a supplement can look like it is helping while serum levels drift; the normal range is 1.7-2.2 mg/dL (Mayo Clinic, 2024), but serum testing is a poor gauge of total body stores because only about 1% of magnesium lives in the blood. If you have reduced kidney function, do not start magnesium without a physician involved. Impaired kidneys clear it slowly.

The practical sequence most clinicians follow is boring and it works. Fix the cheap problem first: glycinate at 200 mg elemental with dinner, moved to bedtime if it sits fine, for six to eight weeks, while watching stool consistency as your tolerability signal. Citrate is not a substitute here, since the laxative threshold sits around 300-500 mg elemental magnesium on product labels, and a dose that loosens your gut is a dose that has stopped being a sleep aid. Escalate to L-threonate only if glycinate fails or cognition is part of the picture. Taurate is a reasonable alternative for people who get anxious on glycine itself, though the evidence base is thinner. And if you take an antibiotic this month, space the magnesium two hours away and mention it to whoever prescribed it.

Frequently Asked Questions

Is magnesium glycinate or citrate better for anxiety?

Magnesium glycinate is the better choice for anxiety. Elemental magnesium bound to glycine absorbs at roughly 30-40% in typical studies, and glycine itself is an inhibitory neurotransmitter that calms the nervous system. Magnesium citrate absorbs reasonably well too, but its osmotic action pulls water into the bowel, so it is better reserved for constipation. If loose stools follow a citrate dose, the resulting discomfort can feed anxious feelings rather than ease them.

Does magnesium oxide help with sleep at all?

No. Magnesium oxide is the cheapest form on pharmacy shelves and also the least useful for sleep. Bioavailability studies put absorption at roughly 4% of the dose, so a 400 mg tablet delivers around 16 mg of usable magnesium. What does absorb mostly stays in the gut, drawing water in and acting as a laxative. It does not raise brain magnesium enough to affect sleep onset at typical doses.

Can I take magnesium L-threonate with magnesium glycinate?

Yes. The two forms use different transport pathways and do not compete, and many people take threonate in the morning for cognition and glycinate at night for sleep. The ceiling to watch is total supplemental magnesium: keep the combined elemental dose under 350 mg per day unless a clinician advises otherwise, since the tolerable upper limit for supplements set by the FDA is 350 mg. Doubling up can also cause nausea or loose stools.

How much magnesium threonate should I take for sleep?

Clinical work on magnesium L-threonate used 1.5-2 g of the compound per day, which supplies 144-200 mg of elemental magnesium. Trials in older adults, including the 2010 Guosong Liu group work at MIT, split that into two doses. For sleep specifically, many users put the larger portion 30-60 minutes before bed. The label figure that matters is elemental magnesium, not the total compound weight.

Why does magnesium citrate make me anxious?

Citrate does not act on anxiety pathways directly. What happens is digestive: the citrate anion draws water into the intestines, producing cramping, urgency and loose stools, and those sensations overlap closely with the physical symptoms of anxiety. If that is what you notice within a few hours of a dose, switch to magnesium glycinate, which is far less osmotic, and the effect usually disappears within two or three days.

What is the best time to take magnesium for sleep?

Take magnesium glycinate or magnesium L-threonate 30-60 minutes before bedtime. Serum magnesium peaks roughly one to two hours after an oral dose, so that window puts the rise in absorption alongside the natural drop in cortisol and core body temperature that precedes sleep onset. Splitting threonate into a morning and evening dose also works, but keep the larger share in the evening. Citrate or oxide late at night tends to mean a 3 a.m. bathroom trip.