Magnesium
mineral · also known as Mg
This is general information, not medical advice. Supplements can interact with medicines and health conditions. Always check with your doctor or pharmacist before starting or changing supplements.
What it is
Magnesium is an essential mineral involved in several hundred enzymatic reactions in the body, including energy production, muscle and nerve function, blood glucose regulation, and blood pressure control. Many adults in high-income countries consume less magnesium than recommended, and mild deficiency is more common in people with poor diets, gastrointestinal disease, or those taking certain medications such as proton pump inhibitors or diuretics. People take magnesium supplements to correct low intake, ease muscle cramps, support sleep, or as an adjunct for migraine prevention or blood pressure control. Different salts (glycinate, citrate, oxide) vary widely in how much elemental magnesium they provide and how well that magnesium is absorbed. Most healthy adults tolerate supplemental magnesium well; the main limiting side effect at higher doses is diarrhoea, which is both dose- and form-dependent.
Which form to choose
Grades here rate how well the evidence supports choosing a form, which is not the same as how much that form has been studied: a well-studied form can be graded low precisely because the studies argue against it. This is a different scale from the claim grades below — see our editorial policy.
Elemental content: 16.1% by weight. A higher figure means more of the mineral per gram, not necessarily more absorbed.
Magnesium bound to citric acid; well absorbed relative to oxide, with a mild-to-moderate laxative effect that becomes more noticeable at higher doses.
Graded on comparative absorption, and supported by trial use. Three randomised comparisons in healthy adults all place citrate above oxide: a 1990 solubility-and-absorption study, a 2003 double-blind 60-day comparison in which citrate also produced the highest serum magnesium of the three forms tested, and a 2017 single-dose crossover finding higher 24-hour urinary excretion and serum levels than oxide. Citrate is also one of the forms used in the migraine prevention trials this monograph cites. Held at B rather than A because every comparison rests on absorption surrogates in small samples of healthy volunteers, not on clinical outcomes, and because no trial establishes that better absorption produces better results.
Elemental content: 14.1% by weight. A higher figure means more of the mineral per gram, not necessarily more absorbed.
Magnesium bound to the amino acid glycine; well absorbed and generally better tolerated (less GI upset) than magnesium oxide, which is why it is often chosen for sleep- or relaxation-focused use.
Graded on comparative absorption, not on trial use. The one randomised head-to-head of the three common forms tested an amino-acid chelate alongside citrate and oxide: both organic forms were absorbed better than oxide at 60 days, but citrate, not the chelate, produced the highest serum magnesium. So the evidence supports glycinate over oxide and does not support it over citrate. The tolerability advantage it is usually chosen for is widely reported but, as far as we can find, not established by a randomised comparison, which is the main reason this is a C rather than a B.
Elemental content: 60.3% by weight. A higher figure means more of the mineral per gram, not necessarily more absorbed.
Highest elemental magnesium content by weight but poorly absorbed (studies have found single-digit to low-double-digit percent absorption); most useful as a low-cost osmotic laxative or antacid rather than for correcting magnesium deficiency.
Graded on comparative absorption, and the grade reflects what that evidence SHOWS rather than how much of it there is. Oxide is the best studied of the three forms here and comes off worst in every randomised head-to-head we can find: less absorbed than citrate in 1990, less than both citrate and an amino-acid chelate over 60 days in 2003, and lower urinary excretion and serum levels than citrate in 2017. Its one real advantage is that it carries far more elemental magnesium per gram, so a smaller tablet delivers a given label dose. It is graded D as a choice, not as a subject of study.
Evidence-based benefits
- Grade B· Limited evidence[1]
Oral magnesium supplementation may modestly reduce migraine attack frequency when used preventively, though the supporting trial evidence is inconsistent across formulations and doses.
An evidence-based review of oral magnesium trials for migraine prevention (mainly citrate or oxide, roughly 400-600 mg/day) found some trials reported fewer migraine attacks versus placebo, but rated the overall evidence as limited due to small sample sizes and heterogeneous dosing and formulations.
- Grade A· Strong evidence[1]
Magnesium supplementation modestly lowers blood pressure on average, with the largest reductions in people who are hypertensive or magnesium-deficient; little to no effect is seen in normotensive people.
A 2025 systematic review and meta-analysis of 38 randomised controlled trials (2,709 participants, median dose 365 mg elemental magnesium/day for 12 weeks) found average reductions of 2.8 mmHg systolic and 2.1 mmHg diastolic blood pressure versus placebo. Reductions were larger (roughly 6-8 mmHg systolic) in people with hypertension or hypomagnesemia, but not statistically significant in normotensive groups.
- Grade B· Limited evidence[1]
Magnesium glycinate supplementation may modestly improve subjective sleep quality in adults who report poor sleep, though the effect size is small.
A 2025 randomised, placebo-controlled trial in 155 adults with self-reported poor sleep found that 250 mg/day elemental magnesium (as bisglycinate) produced a small but statistically significant improvement in Insomnia Severity Index scores versus placebo after 4 weeks (Cohen's d approximately 0.2), with larger gains in those with lower baseline dietary magnesium intake.
- Grade B· Limited evidence[1]
Pooled randomised-trial evidence does not show a significant benefit of magnesium supplementation on glycemic control (fasting blood glucose or HbA1c) in people with type 2 diabetes; a trend toward improvement exists but does not reach statistical significance, even in those with hypomagnesemia.
A meta-analysis of randomised controlled trials of magnesium supplementation in type 2 diabetes found a non-significant trend toward improved glycemic control in the magnesium-treated group (mean difference -0.19, 95% CI -0.58 to 0.21), with a stronger but still non-significant trend in patients with hypomagnesemia (mean difference -1.16, 95% CI -2.92 to 0.6). The authors concluded that routine magnesium supplementation for glycemic control in type 2 diabetes cannot be recommended based on this evidence.
Risks & contraindications
- Grade B· Limited evidence[1]
Magnesium oxide, and other magnesium salts at higher doses, commonly cause diarrhoea and gastrointestinal upset from their osmotic laxative effect in the gut.
A 2023 review of intervention studies reporting diarrhoea as an adverse event found variable rates depending on the magnesium salt and dose; seven studies spanning 128-1200 mg/day found no significant difference in diarrhoea versus control, while other trials (particularly with magnesium oxide) reported diarrhoea in close to half of participants, reflecting oxide's very poor absorption.
- Grade C· Preliminary evidence[1]
In people with reduced kidney function, magnesium supplementation carries a risk of hypermagnesemia, which can cause neuromuscular, cardiac, and respiratory complications.
A 2023 narrative review found hypermagnesemia is uncommon in people with normal kidney function but occurs mainly in those with impaired renal clearance or high combined intake from supplements, laxatives, or antacids; reported effects range from muscle weakness to respiratory failure and coma in severe cases.
Dosage & how to take it
- Magnesium glycinate: 200–400 mg/day, in the evening, with or without food. Dose is elemental magnesium. Glycinate's gentler GI profile makes it a common choice for evening/sleep-support use. On sourcing: the sleep trial cited on this monograph (PMID 40918053) used 250 mg/day of elemental magnesium as bisglycinate, and the blood-pressure meta-analysis (PMID 41000008) had a median dose of 365 mg/day -- both inside this range. The 200 mg/day lower bound is below every dose any trial cited here used; it is a starting and tolerability step, not a dose shown to do anything, so do not read the bottom of the range as an evidenced dose.
- Magnesium citrate: 200–400 mg/day, in the evening, with or without food. Dose is elemental magnesium. May loosen stool at the higher end of the range; splitting the dose can reduce GI effects. On sourcing: the migraine review cited on this monograph (PMID 25533715) covers trials of mainly citrate or oxide at roughly 400-600 mg/day, so its lower bound is this range's upper bound and most of its range sits above ours; the blood-pressure meta-analysis (PMID 41000008) had a median dose of 365 mg/day. The 200 mg/day lower bound is below every dose any trial cited here used -- a starting and tolerability step, not an evidenced dose.
- Magnesium oxide: 200–400 mg/day, in the evening, with or without food. Dose is elemental magnesium. Poor absorption means a larger share acts as an osmotic laxative in the gut; not the preferred form for correcting deficiency. On sourcing: the migraine review cited here (PMID 25533715) covers oxide among the forms trialled at roughly 400-600 mg/day, above this range, and the 200 mg/day lower bound is below every dose any trial cited here used -- a tolerability step rather than an evidenced dose. Diarrhoea is common with oxide specifically (PMID 37487817 reports close to half of participants in some oxide trials), which is the practical limit on dosing this form upward regardless of what the range allows.
Interactions
Long-term PPI use can lower the body's magnesium, most likely by reducing how much magnesium the intestine absorbs. A 2019 meta-analysis of 16 observational studies covering more than 131,000 patients found PPI users had significantly higher odds of low blood magnesium (adjusted odds ratio 1.71), with high-dose PPI use roughly doubling the odds compared with low-dose use. The severe end of this is well documented: a published case series — one patient was on omeprazole itself — describes dangerously low magnesium after one to thirteen years of PPI use, together with low calcium, low potassium, and heart rhythm abnormalities. The evidence is observational, so it cannot prove every PPI user loses magnesium, and most users never develop a problem. Taking a magnesium supplement alongside a PPI is a supportive pairing rather than a risk, but do not use it to self-treat symptoms: if you are on a PPI long-term, ask your prescriber to check your magnesium level, and report muscle cramps, tremor, or palpitations promptly, because severe cases usually need the PPI itself reviewed rather than just more magnesium.
Magnesium is a polyvalent cation, and levothyroxine can bind to polyvalent cations in the gut, forming a complex that your body absorbs poorly, which can lower the medication's effectiveness. A 2025 randomised crossover trial in healthy adults tested this directly: taking magnesium aspartate together with a levothyroxine dose significantly reduced how much levothyroxine reached the bloodstream (AUC reduced by 12%, with peak level also reduced and delayed); magnesium citrate showed a smaller, seven percent reduction that did not reach statistical significance in this study. Because the effect was clearest with one magnesium form and only a non-significant trend with another, and because the trial was small (15 healthy subjects, not people with hypothyroidism), it's reasonable to treat any magnesium supplement as a candidate for the same interaction rather than assume citrate is exempt. If you take levothyroxine, take your magnesium supplement at least 4 hours before or after your dose, and mention any new magnesium supplement to your doctor if your thyroid levels have been stable.
Take at least 4h apart from Thyroid hormones.
Our medication coverage is not exhaustive. Absence of a warning here is not evidence of safety. Always confirm with your pharmacist.
Where to buy
Checking current prices…