How to choose a magnesium supplement: forms, labels and price

Start with elemental magnesium, not the big number

A magnesium label can carry two different numbers: the weight of the whole compound and the weight of the magnesium inside it. Only the second is your dose. A product that lists “500 mg magnesium glycinate” and nothing else has not told you how much magnesium you are getting. Our primer on elemental magnesium covers the basics.

When a label gives both numbers, you can run a quick ratio check. Divide the elemental amount by the compound weight. Pure magnesium glycinate cannot exceed about 14% magnesium by weight, because that ceiling comes from the molecule itself. A label claiming 200 mg of elemental magnesium from 1,000 mg of “glycinate” works out to 20%, which pure glycinate cannot produce.

A result like that usually means the product has been blended with something richer in magnesium, most often oxide, which is about 60% elemental by weight. The practice is commonly called “buffered.” It is not illegal, but it is not what the front of the bottle implies.

Forms compared: what the chemistry and the trials say

There is no single best form. A 2017 review of magnesium absorption research concluded that the type of salt matters less than is often claimed, and that evidence for one form being better absorbed than another is inconsistent. Dose and a person’s existing magnesium status appear to matter more. [PMID 29123461]

What does differ reliably is how much magnesium each gram of compound contains. That drives capsule size, serving count and cost.

The trials that exist do not tell a tidy story. In a 60-day trial of 46 healthy adults taking 300 mg of elemental magnesium daily, citrate raised blood magnesium most, an amino-acid chelate outperformed oxide, and oxide did not differ from placebo. [PMID 14596323] The only direct glycinate-versus-oxide trial used a single 100 mg dose in 12 people who had part of their small intestine removed, and found essentially a tie, 23.5% versus 22.8%. [PMID 7815675]

Other results cut the opposite way. In a 10-day trial in 61 young women, oxide raised ionized blood magnesium while carbonate lowered it. [PMID 35041150] A 2001 study found lactate absorbed better than oxide and about as well as chloride and aspartate. [PMID 11794633] Taken together, the sensible reading is that short trials with different endpoints disagree, not that any one salt has won.

Branded forms deserve extra caution. The case for magnesium L-threonate rests on a rat study showing higher brain magnesium. [PMID 20152124] The human trials are industry-funded, such as one in adults with sleep problems that used 1 g a day for three weeks. [PMID 39252819] A Sucrosomial® comparison was run by the technology’s own maker. [PMID 29630135] Neither is disqualifying, but neither has been independently replicated in the material we reviewed. For a side-by-side of more forms, see this comparison.

Label red flags worth noticing

• Only a compound weight, or only one unlabeled number. “Magnesium 400 mg” might be elemental or might be the whole compound. If the label does not say, you cannot know the dose.
• An elemental-to-compound ratio that exceeds the form’s ceiling. This points to blending, as described above.
• Just “magnesium,” with no form named. This is usually oxide or an undisclosed mix of forms.
• “Chelated” used as a performance claim. Chelation describes a chemical bond. It does not tell you the amount, the purity or the absorption. In the one direct comparison, the chelate and oxide absorbed about equally. [PMID 7815675]
• Proprietary blends that hide the amount of each magnesium source.
• Health claims about anxiety, asthma, sleep or other conditions. A systematic review rated the evidence on magnesium and anxiety as poor. [PMID 28445426] The largest oral magnesium trial for asthma, 450 mg a day for 16 weeks in 300 adults, found no benefit. [PMID 14519140]
• Laxative-scale doses sold as daily supplements. Magnesium oxide at 1.5 g a day improved constipation in one trial, which is a laxative effect, not a reason to take it daily for general use. [PMID 32969946]

Read the “other ingredients” line too. It must list everything besides the active ingredient. Common entries such as a hypromellose or gelatin shell, magnesium stearate (a lubricant) and silicon dioxide (a flow agent) are routine processing aids, not scandals. A “no fillers” claim is worth checking against that line, and a capsule with an empty one deserves a second look.

Third-party testing: what to verify

“Third-party tested” can mean an accredited lab with no stake in the brand checked the product. It can also mean nothing beyond the words. The difference is whether something specific stands behind it.

• USP Verified Mark. A voluntary program that checks identity, strength, purity and manufacturing consistency. Verified products are listed by name on USP’s own site, so search for the exact product.
• NSF Certified for Sport. A similar program that also screens for substances banned by major sports organizations, with its own searchable list.
• Certificate of analysis (COA). A lab report on a specific batch. It is a document, not a certification program, so any brand can obtain one. A brand that tests should be able to send you the COA for your lot number on request.

A badge with no verification number, or a mark you cannot find in the program’s database, has not been verified for that product. Treat a claim you cannot trace the same as no claim.

Price per effective dose

Bottle price and price per capsule both mislead, because servings and elemental content vary. The fair unit is the cost of 100 mg of elemental magnesium.

• Find the elemental magnesium per serving on the panel, and note how many capsules, scoops or gummies a serving is.
• Multiply by servings per container to get total elemental milligrams.
• Divide the price by that total, then multiply by 100.

A hypothetical example: a $24 bottle with 30 servings of 200 mg elemental magnesium holds 6,000 mg, or sixty 100 mg units. That is $0.40 per 100 mg. A $15 bottle with 30 servings of 100 mg holds thirty units and costs $0.50, so the cheaper bottle is the pricier magnesium.

This is where form shows up in your wallet. Glycinate needs far more raw material than oxide to reach the same elemental dose, so it tends to cost more per milligram. Whether that premium buys anything is unresolved, since absorption evidence between forms is mixed.

Powders, capsules and gummies

Format changes convenience and dosing precision more than it changes magnesium. We found no controlled trial comparing the same magnesium salt across capsule, powder and liquid. The absorption studies compared salts, not formats, so any format claim borrows evidence that measured something else.

• Capsules: fixed dose, no taste, and the most elemental magnesium per serving. Check how many capsules make up a serving.
• Powders: the dose depends on how accurately you scoop, and they more often contain flavoring or sweeteners. Check the other-ingredients line.
• Gummies: limited by size. A gummy is mostly sugar or sweetener, gelling agent and flavoring, so there is little room for magnesium.

The gummy ceiling is easy to illustrate. A hypothetical 3 g gummy that is 15% glycinate by weight holds about 450 mg of compound. At the 14% ceiling, that is roughly 60 to 65 mg of magnesium. Reaching a typical dose could mean several gummies, which adds sugar or sugar alcohols. For a longer look, see the gummies versus capsules breakdown.

Two other formats to treat skeptically: “magnesium tea” is usually an herbal blend with no stated elemental amount, and topical sprays, oils and bath products are usually magnesium chloride. A 2017 review found no scientific support for meaningful absorption through skin. [PMID 28788060]

When the product matters less than the person

Choosing well is one half of the decision. The other is whether a supplement fits your situation. Reduced kidney function is the main concern. Severe magnesium excess is rare and happens almost only when kidney function is substantially reduced and intake from supplements, laxatives or antacids is high. [PMID 30220246] A review of chronic kidney disease says both too little and too much magnesium carry risk, and that supplementing in advanced disease needs careful monitoring. [PMID 39256329]

Medications matter too. Magnesium can bind to tetracycline and fluoroquinolone antibiotics in the gut, and doctors advise separating the doses by several hours. [PMID 21942976] Taking it at the same time as levothyroxine reduced absorption of the thyroid medicine in a small trial. [PMID 41221788] Some diuretics increase urinary magnesium loss, so ask a pharmacist before combining them with a supplement. [PMID 10653441]

Finally, do not assume a blood test settles whether you are low. Low blood magnesium is often symptom-free, and a serum test does not necessarily reflect total body stores. [PMID 24402002] [PMID 26069819] More background lives on our magnesium topic page.