For sleep complaints, a trial of the studied bisglycinate may be considered; the expected difference is small.
Study findings. Over 4 weeks, the Insomnia Severity Index (ISI) improved by 1.6 points more than with placebo (95% confidence interval 0.0–3.3); sleep quality and daytime sleepiness measures did not differ noticeably.
Studied regimen250 mg elemental Mg/dayBisglycinate with accompanying glycine
Duration and population4 weeksAdults aged 18–65 with poor sleep
Practical guidance. Studied reference: 250 mg elemental Mg as bisglycinate for 4 weeks; not optimal for everyone.
May slightly improve subjective sleep quality.
Migraine frequency
Confidence2/5
For frequent migraine, magnesium may be discussed as an addition to prevention; the studied high dose requires monitoring.
Study findings. In an RCT, attack frequency during weeks 9–12 fell from baseline by 41.6% with magnesium and 15.8% with placebo; this is a change in frequency, not the proportion of people who improved.
Studied regimen600 mg elemental Mg (24 mmol) dailyTrimagnesium dicitrate
Duration and population12 weeks81 adults aged 18–65 with migraine
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
May reduce attack frequency as an addition to usual care.
Blood pressure
Confidence2/5
With normal blood pressure, do not add magnesium solely to lower it; with high blood pressure a small additional reduction is possible.
Study findings. In the Argeros review, systolic −2.81 mmHg (95% CI −4.32 to −1.29), diastolic −2.05 (−3.23 to −0.88); I² 78% and 88%.
Studied regimen—Different oral forms
Duration and populationDifferent durations in 38 RCTsMixed adult groups with normal and elevated blood pressure
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
May slightly lower blood pressure in studied groups. An effect at normal blood pressure has not been established.
Night leg cramps
Confidence2/5
For ordinary night cramps, do not add magnesium solely to prevent them.
Study findings. In one RCT, the change was −3.41 versus −3.03 cramps per week, p=0.67; no between-group advantage.
Studied regimen520 mg elemental Mg/day in the key RCT; above ULOxide in the key RCT
Duration and population4 weeksNon-pregnant adults with ordinary night cramps
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
No convincing consistent reduction in ordinary night cramps has been shown.
Constipation
Confidence2/5
For constipation, the result concerns magnesium oxide at a laxative dose; do not equate an ordinary supplement with the medicinal regimen for this effect.
Study findings. In two small RCTs, response was 32/47 versus 9/47, RR 3.32 (95% CI 1.59–6.92) over 4 weeks.
Studied regimen1.5 g magnesium oxide/day; compound weight, not an elemental Mg dose specified hereMagnesium oxide
Duration and population4 weeksPatients with chronic constipation
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
Magnesium oxide at studied laxative doses may relieve constipation. The result does not transfer to ordinary supplement doses or other forms.
Glucose and insulin
Confidence1/5
A healthy person should not add magnesium solely for blood sugar; with diabetes or prediabetes, the conclusion depends on baseline status and treatment.
Study findings. In a mixed review, fasting glucose −3.60 mg/dL (95% CI −6.13 to −1.06), HbA1c −0.15 percentage points (−0.26 to −0.03); this is not a prediction for healthy people.
Studied regimen—Different forms
Duration and populationDuration for an everyday regimen not establishedMainly diabetes, prediabetes and other at-risk groups
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
Predictable improvement in blood sugar in a healthy person has not been established. Data in at-risk groups and people with diabetes are considered separately.
Depressive symptoms
Confidence1/5
For depressive symptoms, magnesium may be discussed only as an optional addition; there is no established regimen with a predictable effect.
Study findings. A reliable between-group symptom effect size cannot be extracted from available materials; in one placebo RCT, final HDRS was 10.7 versus 10.4, p=0.908.
Studied regimenIn the key RCT, 40 mg elemental Mg three times dailyDifferent forms; a specific form for the overall conclusion was not identified
Duration and population8 weeks in the key RCTPeople with depressive symptoms; taking fluoxetine in the key RCT
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
May reduce symptom severity as an addition to usual care.
For everyday stress, do not expect a consistent noticeable effect; no trial regimen has been established for this goal.
Study findings. In a sleep RCT, the PSS difference was 0.4 points (95% CI −1.7 to 2.5), PHQ-4 anxiety 0.0 (−0.4 to 0.4); in a fibromyalgia RCT, DASS-stress was 21.8 versus 21.6, p=0.930.
Studied regimen250 and 100 mg elemental Mg/day in different RCTsBisglycinate and chloride in different RCTs
Duration and population4 weeksAdults with poor sleep or fibromyalgia; not selected for ordinary stress
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
May relieve subjective tension and stress.
Memory and attention
Confidence1/5
Do not buy magnesium for proven improvement in everyday memory and attention.
Study findings. In one RCT, NIH Total Cognition Composite changed by +8.40 versus +5.60; the arithmetic difference in changes was 2.80 points, p=0.043, CI not found.
Studied regimen2 g compound (145 mg elemental Mg)/dayL-threonate
Duration and population6 weeksAdults aged 18–45 dissatisfied with sleep
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
Noticeable improvement in everyday memory and attention has not been established.
Recovery after exercise
Confidence1/5
A small effect on post-exercise soreness is possible, but a reliable regimen and effect size for the ordinary exerciser cannot be derived.
Study findings. Soreness decreased in several small studies; no directly comparable between-group difference in points is available in the reviewed studies.
Studied regimenIn individual experiments, 350 mg/day and 500 mg product/day; elemental Mg basis is not always clearDifferent or unspecified forms
Duration and population7–10 days in individual experimentsSmall groups of young exercisers, including runners with low baseline Mg intake
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
May reduce muscle soreness after exercise. Consistent gains in strength and performance from supplementation have not been established.
Exercise and recovery
Confidence2/5
When Mg intake is adequate, do not add it solely for strength, power or endurance.
Study findings. No consistent gain was shown; in a small 2025 crossover RCT, VO₂max was 41.3 versus 44.4 mL/kg/min and mean power 415 versus 439 W with Mg versus placebo.
Studied regimenIn the newer RCT, 300 mg MgCl₂ twice daily; elemental dose unknownChloride in the newer RCT; different forms in the review
Duration and population9 days in the newer RCTHeterogeneous healthy and exercising groups
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
May reduce muscle soreness after exercise. Consistent gains in strength and performance from supplementation have not been established.
Adults should not add magnesium specifically to prevent fractures.
Study findings. No direct estimate of fracture reduction is available; in an RCT in 50 girls, one BMC measure had p=0.0534, total hip p=0.5769.
Studied regimen300 mg elemental Mg/day in a pediatric RCTOxide in the key RCT
Duration and population1 yearKey RCT: girls aged 8–14 with relatively low intake; not equivalent to adults
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
Fracture prevention and noticeable bone strengthening in adults from a separate supplement have not been established.
Quality of life
Confidence1/5
In PCOS, a signal of improved quality of life is possible as an addition, but it should not be generalized to all symptoms and women.
Study findings. In one RCT, SF-36 improved in an adjusted comparison; a verified between-group effect size and CI were not established. Acne, alopecia and bleeding did not differ convincingly.
Studied regimen250 mg magnesium oxide/day; elemental dose unknownOxide according to the tablet description
Duration and population10 weeks64 women with PCOS from an infertility clinic who wanted pregnancy
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
There is a signal of improved quality of life in a specific studied group of women with PCOS. Other symptoms and measures are assessed separately.
Restless legs symptoms
Confidence1/5
Do not choose magnesium alone for restless legs; the signal concerns addition to pramipexole.
Study findings. At 2 months, IRLS was 16.08 with Mg versus 24.24 with placebo; p=0.001 refers to comparison of three groups, and the adjusted pairwise contrast and CI were not established.
Studied regimen250 mg MgO/day; elemental dose unknownMagnesium oxide
Duration and population2 months75 people aged 15–50 with restless legs; all received pramipexole
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
May relieve symptoms when added to pramipexole in the studied group. Benefit from use alone has not been established.
Premenstrual symptoms
Confidence2/5
A trial may be considered for individual premenstrual complaints, but improvement in every symptom cannot be predicted.
Study findings. In Walker 1998, the composite fluid retention measure improved only in the second cycle (p=0.009); there was no difference in the first. A comparable between-group effect size was not extracted.
Studied regimen200 mg Mg/day in Walker 1998Oxide in Walker 1998, with amino acids in the product
Duration and population2 cyclesWomen with premenstrual symptoms
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
May relieve some premenstrual symptoms.
Chronic pain
Confidence2/5
For chronic pain overall, do not add magnesium for consistent pain relief; specific narrow groups require separate conclusions.
Study findings. Cochrane Ferraro 2025: short-term pain −0.55 on a 0–100 scale (95% CI −8.32 to +7.21) in two oral RCTs.
Studied regimen—Different oral forms
Duration and populationShort-term outcome in the reviewAdults with different chronic pain conditions
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
Consistent noticeable relief of chronic pain from oral use has not been established.
Recurrent kidney stones
Confidence2/5
For recurrent stones, do not use magnesium alone as proven prevention; do not attribute potassium–magnesium citrate mixture results to it.
Study findings. One RCT of Mg hydroxide versus placebo: RR 0.59 (95% CI 0.29–1.20) and 0.74 (0.36–1.51) for two doses; both intervals include no effect.
Studied regimenTwo doses in one trial; exact elemental dose not established in the current recordMagnesium hydroxide in a separate trial
Duration and populationDuration for an everyday regimen not establishedPeople with recurrent stone formation
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
No convincing reduction in recurrent stones has been shown for magnesium alone. Potassium–magnesium citrate results apply to the mixture.
Altitude sickness prevention
Confidence1/5
Do not choose magnesium to prevent altitude sickness.
Study findings. Cochrane 2018: RR 1.09 (95% CI 0.55–2.13) for altitude sickness; no convincing reduction.
Studied regimenAbout 1200 mg elemental Mg/day; a high research doseOral form in Dumont 2004
Duration and populationDuration for an everyday regimen not established70 randomized participants in an altitude trial, 61 in the efficacy analysis
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
No convincing preventive benefit was shown.
Menstrual pain
Confidence1/5
An additional symptomatic effect is possible for menstrual pain, but a form and reliable everyday dose cannot be selected.
Study findings. Cochrane 2001: two of three small placebo comparisons were positive, one found no difference; a comparable pain reduction effect size was not extracted.
Studied regimenIn the later study, 150 or 300 mg of a substance not confirmed as elemental MgSalt identity conflicts in the similar later publications: stearate or sulfate
Duration and population2 cycles in the later studyWomen with primary menstrual pain
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
May reduce menstrual pain, including as an addition to pain medication under the studied conditions.
A healthy person should not buy magnesium for weight loss.
Study findings. 2026 review: weight −0.70 kg (95% CI −1.30 to −0.09) in mixed groups; body fat percentage −0.02 percentage points (−0.46 to +0.43).
Studied regimen—Different forms
Duration and populationDuration for an everyday regimen not establishedMixed adult groups, not equivalent to a healthy person
Practical guidance. A practical effective dose for independent use has not been established; the studied dose is described separately.
Noticeable weight loss or reduction in body fat percentage in a healthy person from a supplement has not been established.
