Magnesium: Everything Science Proves (and Debunks) in 2026

Magnesium is an essential mineral that acts as a cofactor in hundreds of enzymatic reactions: energy production in the form of ATP, protein synthesis, nerve transmission, muscle contraction and relaxation, bone structure. This omnipresence explains both its genuine importance and the ease with which it is attributed to just about everything.

Officially recognized functions

It is important to distinguish what magnesium does physiologically from what supplementation changes clinically. These are two different questions, and confusion between the two feeds most of the sector's marketing.

On the first count, European regulations recognize nine contributions under authorized health claims: magnesium contributes to reducing fatigue, to normal functioning of the nervous system, to normal psychological functions, to normal muscle function, to normal protein synthesis, toelectrolyte balance, to normal energy metabolism, as well as to the maintenance of normal bone structure and dentition and to cell division.

These formulations are precise and deserve to be read carefully. "Contributes to the maintenance of normal bone structure" is not equivalent to "prevents fractures." "Contributes to normal psychological functions" is not equivalent to "reduces stress." The entire rest of this article hinges on this distinction.

Actual needs and intakes

Reference intakes for an adult are around 300 to 400 mg per day of elemental magnesium. According to PubMed, a 2025 review devoted to athletes notes that these needs are likely higher in very active individuals, due to losses through sweat and increased metabolic demand.

The main dietary sources are leafy green vegetables, nuts and seeds, legumes, whole grains, dark chocolate and certain mineral waters. The refining of cereals removes a large portion of magnesium, which explains why an industrialized diet inherently provides less.

This is the most important and least known technical point. The serum magnesium, which is what a standard blood test measures, represents only an infinitesimal fraction of the body's magnesium. The bulk is found inside cells and in bones. A normal result therefore does not exclude insufficiency.

A critical reading note is necessary: this is a narrative review, with an advocacy tone, and not a meta-analysis. Its authors defend a strong position on the extent of the problem. The physiological fact they recall, however—the distribution of magnesium in the body—is not disputed.

What this means in practice

The body maintains stable serum magnesium by drawing on bone and cellular reserves. Blood is therefore the last compartment to deteriorate. When blood magnesium levels become frankly abnormal, the situation is already advanced.

This leads to an uncomfortable consequence for everyone: in routine clinical practice, there is no simple and reliable way to know if you have a magnesium deficiency. Therefore, we reason based on clusters of indicators : poor dietary intake, highly processed foods, intense physical activity, age, certain chronic conditions, and certain medications such as diuretics or proton pump inhibitors.

This uncertainty also explains why so many clinical trials on magnesium produce heterogeneous results: they often include people who were not deficient, in whom no benefit is expected.

Paradoxically, the best-demonstrated effect of magnesium is neither sleep, nor stress, nor cramps: it is blood pressure reduction. And above all, it is the only area where a dose threshold clearly emerges from the data.

A decrease of 2 mmHg may seem trivial. At the individual level, it is. At the population level, a reduction of this magnitude shifts the cardiovascular risk curve in a non-negligible way.

The decisive role of dosage

But the most useful information comes from an umbrella meta-analysis, that is, an analysis that compiles ten previous systematic reviews, published in 2024 and covering a total of 8,610 participants.

The difference is spectacular: 1.25 mmHg across all doses, compared to 6.38 mmHg above 400 mg per day. In other words, most studies concluding a modest effect were simply testing insufficient doses. This is the common thread throughout this entire report, and we dedicate the following section to it.

A significant portion of disappointing results published on magnesium are explained by doses that are too low or durations that are too short. The thresholds below are all expressed in elemental magnesium, and not in magnesium salt weight, a distinction that many labels keep deliberately vague.

Magnesium element or salt weight: the confusion to avoid

A technical point that trips up almost everyone. The figure displayed on a label often corresponds to the weight of the salt, and not to that of the magnesium element it actually provides. Yet the gap is considerable depending on the form: magnesium oxide contains approximately 60% magnesium element, whereas bisglycinate contains approximately 14%. At equal powder weight, one therefore provides four times more than the other.

This is the paradox of this market: the most concentrated forms are often the least well absorbed, and the best absorbed are the least concentrated. Only the line indicating magnesium elementmatters, generally accompanied by the percentage of reference intake values.

Magnesium reduces sleep onset latency by approximately 17 minutes in elderly people with insomnia. But this result is based on three trials and 151 participants, with a moderate to high risk of bias and quality of evidence that the authors describe as low to very low. Total sleep time, for its part, is not significantly increased.

The authors conclude that the literature is insufficient for a physician to formulate a strong recommendation, while noting that since magnesium is inexpensive and widely available, existing evidence may justify a trial in affected individuals.

This is an honest position, and it is the one we adopt: there is a signal, it points in the right direction, but presenting magnesium as a proven natural sleep aid would be an exaggeration. Let us also note that this research concerns elderly people with insomnia, not adults who sleep properly.

« Magnesium anti-stress » is the dominant commercial positioning of this mineral. Yet a systematic review devoted precisely to this question reaches an embarrassing conclusion: none of the 18 studies examined administered a validated measure of subjective stress as an evaluation criterion. Regarding anxiety, a benefit appears, but only in populations already vulnerable and on poor quality evidence.

What can and cannot be concluded from this

It cannot be concluded that magnesium plays no role in nervous system balance. Its contribution to the normal functioning of the nervous system and to normal psychological functions is recognized at the European level, and its physiology justifies this.

Rather, it must be concluded that the anti-stress effect as it is marketed has never been properly tested. When a benefit appears on anxiety, it is in people who are already anxious, suffering from premenstrual syndrome or hypertension, and in only about half of the trials.

On depressive mood, a meta-analysis of 52 studies and 4,049 participants found a statistically significant effect but of a very small size, with an SMD of 0.16, below the threshold usually retained for speaking of a small effect. In the same analysis, zinc reached 0.59 and coenzyme Q10 0.97 (DOI: 10.1016/j.jad.2022.11.072).

The honest formula would therefore be: magnesium participates in the normal functioning of the nervous system, and correcting a deficiency can improve how one feels, but it is not an anxiolytic and has never demonstrated a measurable effect on stress.

The association between magnesium and cramps is probably the most deeply rooted in the general public. The reference Cochrane review nevertheless concludes that it is unlikely that magnesium provides clinically significant benefit in older adults suffering from idiopathic cramps. In pregnant women, results are contradictory.

The case is not entirely closed

In the interest of fairness, a subsequent trial should be mentioned. A randomized double-blind placebo-controlled trial, published in Nutrition Journal, tested a particular form of magnesium with enhanced cellular absorption in people suffering from nocturnal leg cramps, and observed a reduction in their frequency and duration (DOI: 10.1186/s12937-021-00747-9).

How to reconcile the two? The most reasonable reading is that the form used and the baseline status of participants probably matter more than magnesium itself. A single trial does not overturn a Cochrane review, but it suggests that the question deserves to be revisited with better-absorbed forms.

In the meantime, presenting magnesium as the remedy for cramps is not supported by the best available evidence.

This is probably the area where magnesium is most effective, and paradoxically the least promoted by marketing. A dose-response meta-analysis published in 2024, covering 22 randomized trials and comparing seven different supplements, places magnesium at the top for reducing the number of attacks.

Why this result deserves to be highlighted

Three elements make it robust. First, it's a direct comparison : magnesium is not evaluated alone but pitted against Coenzyme Q10, riboflavin, alpha-lipoic acid, probiotics, vitamin D, and omega-3s. Next, it achieves the strongest reduction in the number of attacks of the entire panel, with 2.51 fewer attacks per month. Finally, it acts on all three measured dimensions: frequency, intensity, and number of migraine days per month.

For comparison in the same analysis, omega-3s achieve no significant results on any criterion, which shows that the analysis truly discriminates between supplements.

Moreover, this result is not isolated: magnesium has long been part of migraine preventive approaches, and its presumed mechanism, modulation of neuronal excitability, is consistent with what we know about its physiology.

Magnesium contributes to the maintenance of normal bone structure, a claim recognized at the European level. But no data show that supplementation reduces fracture risk. On blood sugar, a network meta-analysis of 178 studies does not include magnesium among effective supplements for type 2 diabetes.

Bones: marginal correlation, no demonstrated protection

The distinction is essential and rarely made. A authorized health claim describes an established physiological contribution, not a clinical benefit demonstrated by supplementation. Magnesium participates in bone structure, that is a fact. That taking magnesium reduces the risk of breaking your hip is another matter, and that one is not demonstrated.

Blood sugar: absent from the ranking of effective supplements

On type 2 diabetes, the most rigorous comparison available is a network meta-analysis of 178 studies, which puts eight supplements in competition on the same criteria.

Magnesium was indeed among the supplements evaluated. It does not emerge in any of the efficacy categories. This is a clear negative result, and it deserves to be reported as plainly as positive results.

This is the commercial battlefield of the sector. The data says two things at once: the organic forms (citrate, malate, glycinate, bisglycinate) are better absorbed than the inorganic forms (oxide, sulfate, chloride); but in a healthy person without prior deficiency, all forms maintain normal physiological levels.

This double conclusion is more useful than it appears. It means that the choice of form becomes truly determinant for three populations: the elderly, the sick, and those whose status is already low. For a healthy adult who simply wishes to ensure adequate intake, the consistency and the dose reached weigh more than the name of the salt listed on the label.

What each form actually provides

The case of marine magnesium deserves further discussion, as its commercial argument relies entirely on its natural origin. Two things must be distinguished that the word conflates: "marine" designates an origin, seawater, while "oxide," "citrate" or "bisglycinate" designate a chemical form. Now it is the form, not the origin, that determines absorption.

Concretely, what is extracted from seawater by evaporation and then purification is a mixture of inorganic salts, mainly oxide and hydroxide, with chloride and sulfate, in proportions that vary depending on the source and the manufacturer's process. These forms are those that the literature places on the side of poorest absorption. In return, their elemental magnesium content is high, which explains the flattering figures displayed on the labels.

In a person whose kidneys function normally, excess magnesium is eliminated through urine, which makes the risk of accumulation low. The main adverse effect is digestive : laxative effect and diarrhea, especially with inorganic forms and at high doses. In case ofkidney failure, on the other hand, magnesium can accumulate and supplementation requires medical advice.

Interactions to know about

Magnesium is known to interact with several classes of medications, mainly by decreasing their absorption when taken simultaneously: this is notably described for certain antibiotics and for bisphosphonates used in osteoporosis. The usual approach is to space out doses by several hours. These interactions are listed in the instructions of the medications involved: your pharmacist is the best person to verify your specific situation.

Conversely, several medications increase losses of magnesium or reduce its absorption, notably diuretics and proton pump inhibitors used long-term for reflux. These are situations where deficiency is more likely.

Decision table

What the data can tell us, situation by situation.

FAQ, all your questions about magnesium