Evidence Files

Magnesium: One of the Few Supplements That Actually Has Solid Evidence — With One Important Caveat

Magnesium is the rare supplement with genuinely good trials behind it — which is exactly why the numbers deserve reading rather than summarising. Blood pressure falls, by 2.0 mmHg. Migraine prevention rates as *possibly effective*. And for the leg cramps that sell more magnesium than anything else, Cochrane concludes it is unlikely to provide clinically meaningful relief.

Published: 20 June 2026

Verdict

TREATMENT (partially confirmed)

We searched PubMed and Cochrane Library for magnesium evidence. Unlike most supplements, the evidence base is genuinely solid — which makes the specifics worth stating precisely rather than in the usual "magnesium is important" register. Correcting a confirmed deficiency is well justified. Beyond that: a meta-analysis of 34 trials found systolic blood pressure fell by 2.00 mmHg; a systematic review rates magnesium as Grade C, possibly effective, for migraine prevention; and Cochrane's 2020 review concludes that for skeletal muscle cramps in older adults, magnesium is unlikely to help at all.

Why magnesium actually matters

Magnesium is a cofactor for over 300 enzymatic reactions including ATP synthesis, DNA replication, and neuromuscular signalling. When levels drop — through inadequate intake, medication effects (PPIs, diuretics, antibiotics), alcohol use, or diabetes — the consequences are measurable: muscle cramps, heart arrhythmias, elevated blood pressure, impaired glucose metabolism.

Deficiency is a genuine clinical entity rather than a supplement-industry invention, and it has identifiable causes: long-term diuretics or proton pump inhibitors, alcohol misuse, malabsorption, poorly controlled diabetes. In those situations, correcting it is treatment.

The good news is that unlike most supplements discussed in evidence-based medicine circles, magnesium has been studied seriously. The complication is that the results are more specific — and more modest — than the reputation: 2.00 mmHg off blood pressure, migraine prevention rated *possibly effective*, and no meaningful benefit for the leg cramps it is most often bought for.

What the evidence shows

Taken claim by claim, with the actual numbers rather than the usual summary.

Leg cramps: the most common reason people buy it, and the weakest evidence. Cochrane's 2020 review concludes it is unlikely that magnesium provides clinically meaningful cramp prophylaxis to older adults. For pregnancy-associated cramps the literature is conflicting, and for exercise-associated cramps the reviewers found no randomised trials at all
Blood pressure: real, and small. A meta-analysis of 34 trials in 2,028 participants found a median dose of 368 mg/day over three months reduced systolic pressure by 2.00 mmHg and diastolic by 1.78. Genuine and statistically significant — and roughly an order of magnitude less than antihypertensive medication
That trial included everyone, not only the deficient. Participants were both normotensive and hypertensive; there was no "only if deficient" condition. We previously stated otherwise, which was inaccurate
Migraine: possibly effective, which is an honest grade. A systematic review in *Headache* assigns Grade C — possibly effective — for migraine prevention, and describes 600 mg of magnesium dicitrate as a safe and cost-efficient option. Worth trying, not established
Type 2 diabetes: modest, and not yet guideline material. A dose-response meta-analysis of 18 trials estimated HbA1c fell by 0.73% at 500 mg/day. The authors state plainly that the trials so far are not sufficient to make clinical practice guidelines
Deficiency itself is a real, treatable condition. It turns up with long-term use of diuretics or proton pump inhibitors, with alcohol misuse, and with malabsorption. This is the case where magnesium is treatment rather than habit

Why "magnesium works" and "this supplement will fix it" are different claims

What the trials measured

Specific outcomes, in specific numbers: 2.00 mmHg off systolic blood pressure, 0.73% off HbA1c at 500 mg/day, migraine prevention rated *possibly effective*. Each is a real finding from controlled trials — and each is modest enough to be worth picturing concretely before buying a year's supply.

What the label implies

Energy, calm, better sleep, fewer cramps, general wellbeing. The cramp claim is the one Cochrane examined directly, and it came back unlikely to help. The rest are mostly extrapolations from magnesium's role in enzymatic reactions — biology, not outcomes.

Frequently asked questions

Does magnesium help with leg cramps?

Most likely not — which matters, because this is the single most common reason people buy it. Cochrane's 2020 review concludes that magnesium is unlikely to provide clinically meaningful cramp prophylaxis in older adults. For cramps in pregnancy the evidence is conflicting, and for exercise-associated cramps there are no randomised trials at all.

Should I get tested before supplementing?

If the goal is to correct a deficiency, yes — otherwise you don't know whether there is anything to correct. One caveat worth knowing: serum magnesium is an imperfect marker, since most of the body's magnesium sits inside cells and in bone. That makes the result something to interpret with a clinician alongside your symptoms and your medication list, rather than a simple pass/fail.

How much does magnesium lower blood pressure?

About 2 mmHg systolic. The 34-trial meta-analysis found 2.00 mmHg systolic and 1.78 diastolic at a median 368 mg/day. That is a real effect worth neither dismissing nor overselling: it is not a substitute for antihypertensive treatment.

Is magnesium worth trying for migraine?

Possibly, and the honest grading says so. A systematic review rates the evidence Grade C — possibly effective — and notes that 600 mg of magnesium dicitrate appears safe and inexpensive. That combination, plausible benefit at low cost and low risk, makes it a reasonable thing to discuss with a clinician who manages your migraines.

Which form should I buy?

Formulation differences in absorption are real, but they are a second-order question — and one where supplement marketing runs well ahead of the trial evidence. The first-order questions are whether you have a reason to supplement at all and what outcome you expect. Note that the migraine review specifically studied magnesium dicitrate at 600 mg, so if migraine is the target, the form and dose studied are worth matching.

Can I just eat more magnesium instead?

For most people, yes, and it is the better default. Dark leafy greens, nuts, seeds, legumes and whole grains supply magnesium alongside fibre, potassium and protein. Supplementation earns its place when there is a specific reason — a confirmed deficiency, a medication that depletes magnesium, or a target like migraine prevention agreed with a clinician.

Our Conclusion

Magnesium earns its reputation in a way most supplements don't: the trials exist, they are decent, and the mechanism is understood. That is precisely why the numbers deserve to be read rather than summarised — and why we corrected two things we had previously written. Blood pressure falls by 2.00 mmHg, not 3–4, and the effect was not limited to people who were deficient. Migraine prevention rates as *possibly effective*. And for leg cramps, the single most common reason anyone buys magnesium, Cochrane concludes it is unlikely to provide clinically meaningful benefit. None of that makes magnesium a scam; it makes it a mineral with specific, modest, well-characterised effects. The useful question is therefore not *which magnesium should I buy* but do I have a reason to take it — a confirmed deficiency, a medication that depletes it, or a target like migraine prevention worth trying with a clinician. Absent one of those, nuts and dark leafy greens are the better-evidenced route.

Sources

Every claim in this article rests on a primary publication. Links go to the publisher's site or to PubMed, the US National Library of Medicine database.

This article is for informational purposes only and does not constitute medical advice. Trick or Treatment analyses the presence of clinical studies in open scientific databases — PubMed and Cochrane Library. The absence of studies in these databases does not automatically mean a drug is ineffective, but it does mean its effectiveness has not been confirmed by evidence-based medicine standards. Any treatment decisions should be made together with your doctor.