Magnesium

Magnesium has been studied in 695 papers since 1981: 216 trials and 234 observational studies in people, 25 in animals, 4 in cells, 208 reviews; 107 in the past year. The newest paper, Nov 15, 2026: “Age-related dysregulation of magnesium metabolism and food-based prevention”.

Intervention

695 papers study it336 more mention it538 findingsUpdated

PubChemChEBIMeSH

Research on it is growing 35% faster than aging research as a whole: 25 papers in the latest year counted

Part of Rhonda Patrick's supplement routineMore supplements

26 new papers in 30 daysSee all 870 papers on magnesium

At a glance

Mouse studies report lifespan extension with magnesium, replicated by independent labs. Independent labs also report lower systolic blood pressure in humans and associations with lower type 2 diabetes risk, while other human studies show no clear blood-pressure effect or diabetes link. Some papers have company-affiliated authors or declare company ties.

538 findings in 268 papers from 1988 to 2026, 22 of them since April 2026

Independent labs found the same change in mice and people. 146 independent labs in 34 countries; 58 trials, 1 about aging, 27 with no results posted.

“resulted in a significant reduction of SBP (ES = -1.25 mmHg; 95% CI: -1.98, -0.51, P = 0.001)”
Impact of Magnesium Supplementation on Blood Pressure: An Umbrella Meta-Analysis of Randomized Controlled TrialsCurrent therapeutic research, clinical and experimental2024doi:10.1016/j.curtheres.2024.100755

56 of its findings point the other way.

Mice and humans: replicated by independent labs. Cells and worms or flies: a single group.
  • 146 independent labs in 34 countries
  • 56 findings against
  • 58 trials, 1 about aging, 27 with no results posted
  • Funders: National Institutes of Health and others
  • 14 of 478 findings with company ties
  • Human trial93
  • Humans226
  • Animals19
  • Model organisms1
  • Cells4
  • In silico3

44 studies gave magnesium to people and measured what changed.

What changedDesignPeopleSource

down: Decreases C-reactive proteinSMD -0.327 (95% CI -0.602 to -0.053)

down: Decreases inflammatory markers

Review of trialsat least 2 months444 people pooledWomenPatient cohortsFrontiers in nutrition, Oct 31, 2025

down: Decreases blood pressuresystolic: 134.6 ± 6.8 to 126.3 ± 5.6 mmHg, diastolic: 84.1 ± 3.9 to 79.4 ± 1.6 mmHg, p < 0.01

down: Decreases IL-64.94 ± 3.30 to 3.01 ± 1.32 pg/mL, p < 0.01

up: Increases vitamin D17.93 ± 8.96 to 24.41 ± 10.20 ng/mL, p < 0.05

Randomized trial12 weeks27 peopleBoth60.2 ± 12.5 yearsOlder adults; metabolic syndromeInternational journal of clinical pharmacology and therapeutics, Dec 1, 2024

up: Increases serum magnesium0.76 ± 0.02 vs 0.69 ± 0.02 mmol/L

down: Decreases IFNG

Randomized trial6 weeks12 people7 malesmean ± SD age 67 ± 7 yearsAdults with insulin-treated type 2 diabetesThe Journal of clinical endocrinology and metabolism, Nov 18, 2024

down: Decreases systolic blood pressureES -1.25 mmHg (95% CI -1.98 to -0.51)

down: Decreases diastolic blood pressureES -1.40 mmHg (95% CI -2.04 to -0.75)

Review of trials8,610 people pooledCurrent therapeutic research, clinical and experimental, Jul 31, 2024

no change: Has no effect on insulin sensitivity

up: Increases serum magnesium0.75±0.02 vs 0.70±0.02 mmol/l

down: Decreases HDL cholesterol1.14±0.08 vs 1.20±0.09 mmol/l

Randomized trial6 weeks14 peopleBothmean 67±6 yearsAdults with type 2 diabetes treated with insulin for at least 1 year and a serum magnesium concentration ≤0.79 mmol/lDiabetologia, Nov 3, 2023

What did studies of magnesium find?

196

Studies of magnesium measured 196 outcomes; the most studied are systolic blood pressure (10 studies), C-reactive protein (8) and blood pressure (7).

Systolic blood pressure

downin humans9no changein humans1

10 studies
  1. Magnesium decreases systolic blood pressure in humans (ES -1.25 mmHg (95% CI -1.98 to -0.51)).

    Human trialn = 8,610

  2. Magnesium decreases systolic blood pressure in thiazide-treated hypertensive women (144 ± 17 vs. 134 ± 14 mmHg (SBP), 88 ± 9 vs. 81 ± 8 mmHg (DBP)).

    Human trialin thiazide-treated hypertensive womenmagnesium chelate600 mg/day6 monthsfemales

  3. Magnesium decreases systolic blood pressure in humans (versus placebo) (2.00 mm Hg reduction (95% CI 0.43-3.58)).

    Human trialversus placebomedian 368 mg/dmedian 3 monthsn = 2,028

  4. Magnesium decreases systolic blood pressure in humans (per 10 mmol/day increase in magnesium dose) (-4.3 mm Hg per 10 mmol/day (95% CI -6.3 to -2.2)).

    Human trialper 10 mmol/day increase in magnesium dose10 mmol/day increment

  5. Magnesium decreases systolic blood pressure in humans (WMD -2.50 mmHg, 95% CI -4.08 to -0.91).

    Review

  6. Magnesium decreases systolic blood pressure in humans (compared with placebo) (-2.81 mm Hg (95% CI, -4.32 to -1.29)).

    Reviewcompared with placeboelemental

  7. Magnesium decreases systolic blood pressure in diabetic and nondiabetic individuals (WMD = -3.056 mmHg, 95% CI = -5.509, -0.603, P = 0.015).

    Reviewin diabetic and nondiabetic individuals

  8. Magnesium decreases systolic blood pressure in humans (-18.7 mmHg, 95% CI -14.95 to -22.45, Cohen's d = 1.19).

    Reviewin hypertensive subjects on anti-hypertensive medication continuously for at least six months with starting SBP > 155 mmHgoralat least six monthsn = 135

  9. Magnesium decreases systolic blood pressure in humans (decrease in SBP of 3-4 mm Hg, Hedges G 0.32 (95% CI 0.23-0.41)).

    Review120-973 mg3 to 24 weeksn = 1,173

  10. Magnesium showed no clear effect on systolic blood pressure in humans (pooled random-effects meta-analysis of RCTs versus placebo, no treatment or usual care) (mean difference -1.3 mmHg, 95% CI -4.0 to 1.5).

    Reviewpooled random-effects meta-analysis of RCTs versus placebo, no treatment or usual careeight to 26 weeks follow-upn = 545

More on Systolic blood pressure

C-reactive protein

downin humans7no changein humans2OR 1.49

8 studies
  1. Magnesium decreases C-reactive protein in humans (versus placebo in patients with metabolic syndrome) (SMD -0.327 (95% CI -0.602 to -0.053)).

    Human trialversus placebo in patients with metabolic syndromeoralserumn = 444

  2. Magnesium decreases C-reactive protein in humans (compared with placebo at 6-month follow-up) (β = -3.2, 95% CI -6.0 to -0.4, p = 0.03).

    Human trialcompared with placebo at 6-month follow-up300 mg/day6 monthsn = 49

  3. Magnesium decreases C-reactive protein in humans (0.40).

    Human trial180 daysn = 49

  4. Magnesium decreases C-reactive protein in humans (baseline plasma CRP > 3.0 mg/L).

    Human trialbaseline plasma CRP > 3.0 mg/L7 weeks

  5. Magnesium is linked with lower C-reactive protein in humans (multivariate-adjusted; highest vs lowest quintile of dietary intake) (24% lower in highest vs lowest quintile (1.30 vs 1.70 mg/dL); P for trend = 0.03).

    Humansmultivariate-adjusted; highest vs lowest quintile of dietary intakedietaryfemalesn = 657

  6. Magnesium decreases C-reactive protein in humans.

    Revieworaln = 889

  7. Magnesium showed no clear effect on C-reactive protein in humans (oral supplementation versus control in adults) (WMD -0.49 mg/L (95% CI -1.72 to 0.75)).

    Revieworal supplementation versus control in adultsn = 927

  8. Magnesium decreases C-reactive protein in humans (baseline plasma CRP > 3 mg/L) (WMD: -1.12 mg/L, 95% CI: -2.05, -0.18, p=0.019).

    Reviewbaseline plasma CRP > 3 mg/L

  9. Magnesium showed no clear effect on C-reactive protein in humans (overall) (WMD: -0.11 mg/L, 95% CI: -0.75, 0.52, p=0.727).

    Reviewoverall

  10. Magnesium decreases C-reactive protein in humans (comparing the lowest to the highest group of Mg intake) (OR 1.49 (95% CI 1.18-1.89)).

    Reviewcomparing the lowest to the highest group of Mg intake

More on C-reactive protein

Blood pressure

no changein humans5downin humans2upin humans1

7 studies
  1. Magnesium showed no clear effect on blood pressure in humans.

    Human trial1 weekn = 21

  2. Magnesium showed no clear effect on blood pressure in humans.

    Human trialfemalesn = 199

  3. Magnesium showed no clear effect on blood pressure in humans (insulin-requiring Type 2 DM).

    Human trialinsulin-requiring Type 2 DMoral3 monthsn = 50

  4. Magnesium was not clearly linked with blood pressure in humans after adjustment for confounding factors.

    Humansafter adjustment for confounding factorsn = 2,504

  5. Magnesium improves blood pressure in humans (particularly in individuals with baseline hypomagnesemia).

    Reviewparticularly in individuals with baseline hypomagnesemia

  6. Magnesium decreases blood pressure in humans (added to antihypertensive drug therapy).

    Reviewadded to antihypertensive drug therapy

  7. Magnesium showed no clear effect on blood pressure in controlled hypertensives and normotensives.

    Reviewin controlled hypertensives and normotensives

  8. Magnesium decreases blood pressure in uncontrolled hypertensive patients taking antihypertensive medications.

    Reviewin uncontrolled hypertensive patients taking antihypertensive medications≥240 mg/day

Blood pressure

no changein humans6downin humans1

7 studies
  1. Magnesium decreases blood pressure in humans (metabolic syndrome) (systolic: 134.6 ± 6.8 to 126.3 ± 5.6 mmHg, diastolic: 84.1 ± 3.9 to 79.4 ± 1.6 mmHg, p < 0.01).

    Human trialmetabolic syndromeoral400 mg daily12 weeksn = 27

  2. Magnesium showed no clear effect on blood pressure in humans.

    Human trialover-the-counter magnesium supplementone weekn = 19

  3. Magnesium showed no clear effect on blood pressure in humans (office and 24-h ambulatory measures, magnesium vs placebo).

    Human trialoffice and 24-h ambulatory measures, magnesium vs placebocapsules350 mg/d24 wkmen and postmenopausal womensn = 52

  4. Magnesium showed no clear effect on blood pressure in humans (adults with high-normal diastolic blood pressure, placebo-controlled).

    Human trialadults with high-normal diastolic blood pressure, placebo-controlleddiglycine360 mg daily3 and 6 monthsn = 698

  5. Magnesium showed no clear effect on blood pressure in humans (untreated mild-to-moderate hypertension).

    Human trialuntreated mild-to-moderate hypertension15 mmol/daytwo monthsn = 39

  6. Magnesium showed no clear effect on blood pressure in humans (alongside cholesterol-poor, energy-restricted diet).

    Human trialalongside cholesterol-poor, energy-restricted diet500 mg dailyfour weeks

  7. Magnesium showed no clear effect on blood pressure in humans (Type 2 diabetes).

    ReviewType 2 diabetesoral15 mmol/day (360 mg/day)4-16 weeks

More on Blood pressure

Flow-mediated dilation

upin humans4no changein humans4

6 studies
  1. Magnesium showed no clear effect on flow-mediated dilation in humans.

    Human trialover-the-counter magnesium supplementone weekn = 19

  2. Magnesium showed no clear effect on flow-mediated dilation in humans (overweight and obese adults) (0.49 pp; 95% CI: -0.38 to 1.36 pp; P = 0.26).

    Human trialoverweight and obese adults350 mg daily24 weeksn = 52

  3. Magnesium improves flow-mediated dilation in thiazide-treated hypertensive women (+3.7 ± 2.1 vs. 2.4 ± 1.2%).

    Human trialin thiazide-treated hypertensive womenmagnesium chelate600 mg/day6 monthsfemales

  4. Magnesium improves flow-mediated dilation in humans (MD = 8.32; 95% CI: 3.45 to 13.19; p < 0.01).

    Revieworal> 400 mg/day

  5. Magnesium showed no clear effect on flow-mediated dilation in humans (MD = 2.86; 95% CI: - 1.31 to 7.03; p = 0.18).

    Revieworal

  6. Magnesium improves flow-mediated dilation in humans (unhealthy subjects, or BMI >= 25 kg/m2).

    Reviewunhealthy subjects, or BMI >= 25 kg/m2orallonger than 6 months

  7. Magnesium showed no clear effect on flow-mediated dilation in humans (MD 2.13, 95% CI -0.56 to 4.82, p = 0.12).

    Revieworal

  8. Magnesium increases flow-mediated dilation in humans (MD: 2.97; 95% CI: 0.23 to 5.70%, p = 0.033).

    Review

More on Flow-mediated dilation

Diastolic blood pressure

downin humans6

6 studies
  1. Magnesium decreases diastolic blood pressure in humans (ES -1.40 mmHg (95% CI -2.04 to -0.75)).

    Human trialn = 8,610

  2. Magnesium decreases diastolic blood pressure in humans.

    Human trialn = 7

  3. Magnesium decreases diastolic blood pressure in humans (p = 0.047).

    Human trial300 mg/d1 weekn = 13

  4. Magnesium decreases diastolic blood pressure in humans (compared with placebo) (-2.05 mm Hg (95% CI, -3.23 to -0.88)).

    Reviewcompared with placeboelemental

  5. Magnesium decreases diastolic blood pressure in humans (-10.9 mmHg, 95% CI -8.73 to -13.1, Cohen's d = 1.19).

    Reviewin hypertensive subjects on anti-hypertensive medication continuously for at least six months with starting SBP > 155 mmHgoralat least six monthsn = 135

  6. Magnesium decreases diastolic blood pressure in humans (decrease in DBP of 2-3 mm Hg, Hedges G 0.36 (95% CI 0.27-0.44)).

    Review120-973 mg3 to 24 weeksn = 1,173

  7. Magnesium decreases diastolic blood pressure in humans (pooled random-effects meta-analysis of RCTs versus placebo, no treatment or usual care) (mean difference -2.2 mmHg, 95% CI -3.4 to -0.9).

    Reviewpooled random-effects meta-analysis of RCTs versus placebo, no treatment or usual careeight to 26 weeks follow-upn = 545

More on Diastolic blood pressure

HbA1c

no changein humans4downin humans2

6 studies
  1. Magnesium decreases HbA1c in humans (-0.36%, p < 0.001).

    Human trialthree monthsn = 42

  2. Magnesium decreases HbA1c in T2D patients (8.32 to 7.96%, p < 0.001).

    Human trialin T2D patientselemental Mg250 mg/daythree monthsn = 42

  3. Magnesium showed no clear effect on HbA1c in humans (type 2 diabetes).

    Human trialtype 2 diabetes360 mg Mg/day8 weeksn = 11

  4. Magnesium showed no clear effect on HbA1c in humans (insulin-requiring Type 2 DM) (glucose: 10.7 +/- 3.8 vs 11.6 +/- 6.2 mmol l(-1), p = 0.8; HbA1c: 8.9 +/- 1.6 vs 9.1 +/- 1.2%, p = 0.8).

    Human trialinsulin-requiring Type 2 DMoral3 monthsn = 50

  5. Magnesium showed no clear effect on HbA1c in people with insulin-treated type 2 diabetes and low serum magnesium.

    Reviewin people with insulin-treated type 2 diabetes and low serum magnesiumoral supplement15 mmol/day6 weeksn = 14

  6. Magnesium decreases HbA1c in humans (at 500 mg/d) (-0.73 % (95 % CI: -1·25, -0·22), P = 0.004).

    Reviewat 500 mg/d500 mg/d

  7. Magnesium showed no clear effect on HbA1c in humans (Type 2 diabetes) (-0.31% (95% CI, -0.81 to 0.19)).

    ReviewType 2 diabetesoral15 mmol/day (360 mg/day)median duration of 12 weeks

More on HbA1c

HOMA-IR

downin humans5upin humans1OR 0.29

6 studies
  1. Magnesium decreases HOMA-IR in humans (HOMA-IR 3.04±0.29 before vs 2.43±0.19 after magnesium; p<0.05).

    Human trialuncontrolled before-after comparison in patients with rheumatoid arthritis, on routine treatmentoral300 mg/day6 monthsn = 35

  2. Magnesium decreases HOMA-IR in humans (-1.72, p < 0.001).

    Human trialthree monthsn = 42

  3. Magnesium decreases HOMA-IR in humans (metabolic syndrome) (OR: 0.29; 95% CI: 0.12, 0.72).

    Human trialmetabolic syndromedietary12 monthsn = 234

  4. Magnesium decreases HOMA-IR in humans (versus placebo in prediabetic adults, pooled from randomized trials) (MD -1.10).

    Reviewversus placebo in prediabetic adults, pooled from randomized trialsn = 384

  5. Magnesium decreases HOMA-IR in people at high risk of diabetes (SMD -0.57 (95% CI -1.17 to 0.03)).

    Reviewin people at high risk of diabetes

  6. Magnesium improves HOMA-IR in humans (WMD -0.67, 95% CI -1.20 to -0.14, p=0.013).

    Revieworal

More on HOMA-IR

Serum magnesium

upin humans4no changein humans1

4 studies
  1. Magnesium increases serum magnesium in humans (0.76 ± 0.02 vs 0.69 ± 0.02 mmol/L).

    Human trialmagnesium gluconate6 weeksn = 12

  2. Magnesium increases serum magnesium in adults with insulin-treated type 2 diabetes and low magnesium levels.

    Human trialin adults with insulin-treated type 2 diabetes and low magnesium levelsplasma-serum15 mmol/dayn = 12

  3. Magnesium increases serum magnesium in humans (0.75±0.02 vs 0.70±0.02 mmol/l, p=0.016).

    Human trialin people with insulin-treated type 2 diabetes and low serum magnesium, compared with placebooral supplement15 mmol/day6 weeksn = 14

  4. Magnesium increases serum magnesium in humans (0.75±0.02 vs 0.70±0.02 mmol/l).

    Human trial6 weeksn = 14

  5. Magnesium increases serum magnesium in humans (versus placebo) (0.05 mmol/L elevation (95% CI 0.03, 0.07)).

    Human trialversus placebomedian 368 mg/dmedian 3 monthsn = 2,028

  6. Magnesium increases serum magnesium in humans (baseline serum magnesium < 1.8 mg/dL).

    Human trialbaseline serum magnesium < 1.8 mg/dL7 weeksn = 37

  7. Magnesium showed no clear effect on serum magnesium in all participants.

    Human trialin all participants7 weeksn = 96

More on Serum magnesium

HDL cholesterol

upin humans2downin humans1no changein humans1

4 studies
  1. Magnesium decreases HDL cholesterol in people with insulin-treated type 2 diabetes and low serum magnesium (1.14±0.08 vs 1.20±0.09 mmol/l, p=0.026).

    Human trialin people with insulin-treated type 2 diabetes and low serum magnesiumoral supplement15 mmol/day6 weeksn = 14

  2. Magnesium decreases HDL cholesterol in humans (1.14±0.08 vs 1.20±0.09 mmol/l).

    Human trial6 weeksn = 14

  3. Magnesium was not clearly linked with HDL cholesterol in humans after adjustment for confounding factors.

    Humansafter adjustment for confounding factorsn = 2,504

  4. Magnesium increases HDL cholesterol in humans (WMD = 1.21 mg/dl, 95% CI: 0.58 to 1.85).

    Reviewserum

  5. Magnesium increases HDL cholesterol in humans (Type 2 diabetes) (0.08 mmol/l (95% CI, 0.03 to 0.14)).

    ReviewType 2 diabetesoral15 mmol/day (360 mg/day)4-16 weeks

More on HDL cholesterol

Inflammation

downin humans2no changein humans1downin reviews1

4 studies
  1. Magnesium showed no clear effect on inflammation in humans.

    Human trial24-weekn = 52

  2. Magnesium decreases inflammation in humans.

    Review

  3. Magnesium decreases inflammation.

    Review

  4. Magnesium decreases inflammation in humans.

    Revieworaln = 889

More on Inflammation

Fasting plasma glucose

downin humans4

4 studies
  1. Magnesium decreases fasting plasma glucose in people with diabetes, compared with placebo.

    Human trialin people with diabetes, compared with placebo

  2. Magnesium is linked with lower fasting plasma glucose in humans (β = -0.021, 95% CI: -0.033 to -0.009).

    Humansurinen = 2,348

  3. Magnesium decreases fasting plasma glucose in diabetic and nondiabetic individuals (WMD = -4.641 mg dL−1, 95% CI = -7.602, -1.680, P = 0.002).

    Reviewin diabetic and nondiabetic individuals

  4. Magnesium decreases fasting plasma glucose in people with diabetes (SMD -0.40 (95% CI -0.80 to -0.00)).

    Reviewin people with diabetes

More on Fasting plasma glucose

Systolic blood pressure

downin humans2no changein rats1downin rats1

3 studies
  1. Magnesium decreases systolic blood pressure in humans (essential hypertension receiving long-term thiazide diuretic treatment) (Systolic, diastolic, and mean blood pressures decreased by 7.5, 3.0, and 4.5 mm Hg, respectively).

    Human trialessential hypertension receiving long-term thiazide diuretic treatmentMgO600 mg Mg/day4 weeksn = 20

  2. Magnesium decreases systolic blood pressure in humans (type 2 diabetes mellitus) (WMD: - 5.78 mmHg, 95% CI: - 11.37 to - 0.19).

    Reviewtype 2 diabetes mellitus

  3. Magnesium showed no clear effect on systolic blood pressure in rats (old SHR).

    Animalsold SHRMg2+ rich diet0.8% Mg2+malesn = 5

  4. Magnesium decreases systolic blood pressure in rats (young SHR; rise in systolic blood pressure assessed at 7 weeks) (p < 0.05).

    Animalsyoung SHR; rise in systolic blood pressure assessed at 7 weeksMg2+ rich diet0.8% Mg2+7 weeksmalesn = 5

Carotid intima-media thickness

downin humans2no changein humans1

3 studies
  1. Magnesium prevents carotid intima-media thickness in thiazide-treated hypertensive women (0.79 ± 0.16 vs. 0.79 ± 0.19 mm in magnesium vs. 0.78 ± 0.13 to 0.89 ± 0.14 mm in placebo).

    Human trialin thiazide-treated hypertensive womenmagnesium chelate600 mg/day6 monthsfemales

  2. Magnesium decreases carotid intima-media thickness in humans (MD = - 0.08; 95% CI: - 0.15 to - 0.01; p = 0.03).

    Revieworal

  3. Magnesium showed no clear effect on carotid intima-media thickness in humans (MD: -0.13 mm; 95% CI: 0.27, 0.01; p = 0.077).

    Review

More on Carotid intima-media thickness

Oxidative stress and ROS signaling

downin humans1downin rat cells1downin reviews1

3 studies
  1. Magnesium decreases oxidative stress and ROS signaling in humans.

    Human trial340 mg per day2 weeks

  2. Magnesium decreases oxidative stress and ROS signaling.

    Review

  3. Magnesium decreases oxidative stress and ROS signaling in senescent BMSCs.

    Cellsin senescent BMSCs

More on Oxidative stress and ROS signaling

All 196

What is magnesium linked with?

134

Studies link magnesium with 134 measures, most often type 2 diabetes (15 studies), frailty (7) and colorectal cancer (7).

Type 2 diabetes

downin humans15OR 0.545–0.64

15 studies
  1. Magnesium is linked with a lower risk of type 2 diabetes in humans (first year after bariatric surgery) (OR 0.545).

    Humansfirst year after bariatric surgery

  2. Magnesium is linked with a lower risk of type 2 diabetes in humans (pooled across the three cohorts; quintile 5 vs quintile 1) (pooled multivariate HR 0.85 (95% CI 0.80-0.91), P < 0.0001).

    Humanspooled across the three cohorts; quintile 5 vs quintile 1quintile 5 vs quintile 1

  3. Magnesium is linked with a lower risk of type 2 diabetes in humans (RR 0.77, 95% CI 0.71-0.82).

    Humansn = 539,735

  4. Magnesium is linked with a lower risk of type 2 diabetes in those with baseline metabolic impairment (-32%, P trend = 0.05).

    Humansin those with baseline metabolic impairment7-yearn = 2,582

  5. Magnesium is linked with a lower risk of type 2 diabetes in humans (highest vs lowest quartile of magnesium intake) (OR 0.64 (95% CI 0.44 to 0.94) for highest vs lowest quartile).

    Humanshighest vs lowest quartile of magnesium intake5-year follow-upn = 17,592

  6. Magnesium is linked with a lower risk of type 2 diabetes in humans (adjusted for age and BMI) (P = 0.007 for trend (age- and BMI-adjusted); fully adjusted RR 0.89 highest vs lowest quintile, P = 0.05 for trend).

    Humansadjusted for age and BMIdietary intakefemales

  7. Magnesium is associated with type 2 diabetes in humans.

    Reviewdietary intake

  8. Magnesium is associated with type 2 diabetes in humans.

    Reviewdaily dietary intake

  9. Magnesium lowers the risk of type 2 diabetes in humans (comparing the highest magnesium intake to the lowest) (RR 0.78 (95% CI 0.75 to 0.81), AR reduction 0.120%).

    Reviewcomparing the highest magnesium intake to the lowest

  10. Magnesium lowers the risk of type 2 diabetes in humans (RR: 0.81; 95% CI, 0.77-0.86).

    Reviewdietary100 mg/day increment

All 15

More on Type 2 diabetes

Frailty

downin humans6upin humans1OR 0.43–2.85

7 studies
  1. Magnesium is associated with frailty in humans.

    Humans

  2. Magnesium is linked with a lower risk of frailty in patients with COPD, weighted logistic regression, adjusted (OR 0.85 (95% CI 0.76-0.96) per 100-unit increase).

    Humansin patients with COPD, weighted logistic regression, adjustedper 100-unit increase in Mg intaken = 1,696

  3. Magnesium is linked with a lower risk of frailty in humans (highest vs lowest sex-specific quartile, fully adjusted logistic regression) (OR 0.43 (0.21, 0.86); P-trend = .016).

    Humanshighest vs lowest sex-specific quartile, fully adjusted logistic regressionmean follow-up 3.5 yearsn = 1,900

  4. Magnesium is linked with a lower risk of frailty in humans (RR 0.88 (0.82, 0.94), Q5 vs Q1).

    Humansafter adjustment for lifestyle factors, medication use and dietary factors; highest vs lowest quintile (Q5 vs Q1)total energy-adjusted intake from diet and supplementsQ5 vs Q1median follow-up of 16 yearsfemales

  5. Magnesium is associated with frailty in humans (adjusted OR 2.85 (95% CI 1.23-6.97)).

    Humanslower-median serum magnesium group versus higher; adjusted for age, BMI, diabetes, cardiovascular disease, prevalent fractures, serum albumin and CRPn = 339

  6. Magnesium is linked with a lower risk of frailty in humans (0.5% reduced frailty risk per 1 mg/1000 kcal increase).

    Humansper 1 mg/1000 kcal increase in total intake, in individuals with or at risk of osteoarthritistotal intake from food sources and supplements1 mg/1000 kcal increase8-year follow-up

  7. Magnesium is linked with a lower risk of frailty in humans (intake at or above the RDA) (HR 0.51; 95% CI 0.26-0.93).

    Humansintake at or above the RDA8 yearsmalesn = 1,857

Colorectal cancer

downin humans7OR 0.65

7 studies
  1. Magnesium is linked with a lower risk of colorectal cancer in the total population (OR 0.65, 95% CI 0.52-0.81).

    Humansin the total population

  2. Magnesium is linked with a lower risk of colorectal cancer in humans (RR: 0.88; 95% CI: 0.81, 0.97).

    Humansevery 100-mg/d increase

  3. Magnesium is linked with a lower risk of colorectal cancer in humans (age-adjusted model only) (RR 0.67, highest vs lowest decile; P(trend)<0.001).

    Humansage-adjusted model only28 yearsfemalesn = 85,924

  4. Magnesium is associated with colorectal cancer in humans (Without previous diagnosis of cancer at baseline; adjusted for potential confounders) (RR 0.59, 95% CI 0.40–0.87; highest versus lowest quintile).

    HumansWithout previous diagnosis of cancer at baseline; adjusted for potential confoundersDietary intakeHighest versus lowest quintile of magnesium intakeMean follow-up of 14.8 yearsfemalesn = 61,433

  5. Magnesium lowers the risk of colorectal cancer in humans (RR 0.775, 95% CI 0.655-0.919).

    Review

  6. Magnesium lowers the risk of colorectal cancer in humans (RR 0.81 (95% CI: 0.70-0.92)).

    Reviewhighest versus lowest

  7. Magnesium lowers the risk of colorectal cancer in humans (highest vs lowest category) (RR 0.89, 95% CI 0.79-1.00).

    Reviewhighest vs lowest category

More on Colorectal cancer

Metabolic syndrome

downin humans6OR 0.61–0.73

6 studies
  1. Magnesium is associated with metabolic syndrome in humans across plasma magnesium quantiles (p-trend<0.05).

    Humansacross plasma magnesium quantiles

  2. Magnesium is linked with a lower risk of metabolic syndrome in humans (OR = 0.61, 95% CI = 0.41-0.91).

    Humans

  3. Magnesium is linked with a lower risk of metabolic syndrome in humans (total magnesium intake, highest vs lowest quintile) (RR 0.79 (0.63–0.98), highest vs lowest quintile).

    Humanstotal magnesium intake, highest vs lowest quintiletotal intake>437.9 vs <223.5 mg/day

  4. Magnesium is linked with a lower risk of metabolic syndrome in humans (OR 0.69, 95% CI 0.59, 0.81).

    Humansn = 24,473

  5. Magnesium is associated with metabolic syndrome in humans (adjusted for confounding lifestyle and dietary factors) (OR 0.73, 95% CI 0.60–0.88; P for trend = 0.0008).

    Humansadjusted for confounding lifestyle and dietary factorshighest versus lowest intake quintilefemalesn = 11,686

  6. Magnesium lowers the risk of metabolic syndrome in humans (RR 0.88 (95% CI, 0.84-0.93)).

    Review150 mg/day increment

More on Metabolic syndrome

Sarcopenia

downin humans2mixedin humans2upin humans1OR 0.63–0.998

5 studies
  1. Magnesium is associated with sarcopenia in humans (Q3: OR = 0.63, 95% CI: 0.42–0.95).

    Humansegg-derivedQ34 yearsmales

  2. Magnesium is linked with a lower risk of sarcopenia in humans (OR = 0.998 (95% CI = 0.996, 0.999, P < 0.001)).

    Humans

  3. Magnesium is associated with sarcopenia in humans (from included cross-sectional and observational studies).

    Reviewfrom included cross-sectional and observational studies

  4. Magnesium is associated with sarcopenia in humans (low Mg levels and low Mg intake).

    Reviewlow Mg levels and low Mg intake

  5. Magnesium is associated with sarcopenia in humans.

    Review

More on Sarcopenia

Hypertension

downin humans4upin humans1OR 0.4–2.19

5 studies
  1. Magnesium is linked with a higher risk of hypertension in humans (extreme-tertile comparison) (OR 2.19 (95% CI 1.36, 3.54)).

    Humansextreme-tertile comparisonwhole blood

  2. Magnesium is associated with hypertension in humans (OR 0.66 (95% CI: 0.51-0.87)).

    Humanshighest vs lowest quintile of dietary magnesium intake, adjusted for sociodemographic, anthropometric and lifestyle factorsn = 24,171

  3. Magnesium is linked with a lower risk of hypertension in humans (first year after bariatric surgery) (OR 0.584).

    Humansfirst year after bariatric surgery

  4. Magnesium is linked with a lower risk of hypertension in humans (obese women, highest vs lowest quartile of magnesium intake, adjusted) (adjusted OR 0.40 (95% CI 0.25–0.63), highest vs lowest quartile).

    Humansobese women, highest vs lowest quartile of magnesium intake, adjustedfemales

  5. Magnesium lowers the risk of hypertension in humans.

    Review

More on Hypertension

Stroke

downin humans5RR 0.89–0.93

5 studies
  1. Magnesium is linked with a lower risk of stroke in humans.

    Humans

  2. Magnesium is linked with a lower risk of stroke in humans (combined RR: 0.92; 95% CI: 0.88, 0.97).

    Humans100 mg Mg/dn = 241,378

  3. Magnesium lowers the risk of stroke in humans (comparing the highest magnesium intake to the lowest) (RR 0.89 (95% CI 0.83 to 0.94), AR reduction 0.281%).

    Reviewcomparing the highest magnesium intake to the lowest

  4. Magnesium lowers the risk of stroke in humans (7% reduction; RR: 0.93; 95% CI, 0.89-0.97).

    Reviewdietary100 mg/day increment

  5. Magnesium lowers the risk of stroke in humans (RR 0.89, 95% CI 0.82-0.97).

    Reviewn = 304,551

More on Stroke

Mortality

downin humans5−16%HR 0.6–0.74

5 studies
  1. Magnesium is linked with a lower risk of mortality in RA patients, fully adjusted weighted Cox model (HR 0.74, 95% CI 0.56-0.99; ARD -11.12%).

    Humansin RA patients, fully adjusted weighted Cox model

  2. Magnesium is associated with mortality in humans.

    Humans

  3. Magnesium is linked with a lower risk of mortality in humans (stroke survivors; highest vs lowest quartile of total intake, fully adjusted) (40% reduction in risk; HR 0.60 (95% CI 0.38-0.94)).

    Humansstroke survivors; highest vs lowest quartile of total intake, fully adjustedtotal intake from diet and dietary supplements≥18.5 vs ≤12.0 mg/(100 kcal*d)median follow-up 5.3 yearsn = 917

  4. Magnesium is linked with a lower risk of mortality in humans (high cardiovascular risk) (HR: 0.66; 95% CI: 0.45, 0.95; P < 0.01).

    Humanshigh cardiovascular riskmedian follow-up of 4.8 yn = 7,216

  5. Magnesium lowers the risk of mortality in humans (-16%).

    Reviewfemales

Ischemic stroke

downin humans4RR 0.88–0.91

4 studies
  1. Magnesium is linked with a lower risk of ischemic stroke in humans (RR: 0.91; 95% CI: 0.87, 0.96).

    Humansn = 241,378

  2. Magnesium lowers the risk of ischemic stroke in humans (comparing the highest magnesium intake to the lowest) (RR 0.88 (95% CI 0.81 to 0.95), AR reduction 0.246%).

    Reviewcomparing the highest magnesium intake to the lowest

  3. Magnesium lowers the risk of ischemic stroke in humans (highest vs lowest magnesium intake category, pooled prospective cohorts) (RR 0.88 (95% CI 0.81-0.95)).

    Reviewhighest vs lowest magnesium intake category, pooled prospective cohorts

  4. Magnesium lowers the risk of ischemic stroke in humans (RR 0.88, 95% CI 0.80-0.98).

    Review

More on Ischemic stroke

Mild cognitive impairment

downin humans2upin humans1OR 0.19–1.76

3 studies
  1. Magnesium is associated with mild cognitive impairment in humans (compared with the low-metal + low-FA subgroup) (OR 1.76, 95% CI 1.25–2.48).

    Humanscompared with the low-metal + low-FA subgrouphigh magnesium and high dietary folic acid

  2. Magnesium is associated with mild cognitive impairment in humans (OR 0.53 (95%CI 0.32-0.90), highest vs lowest quartile (median 48.4 vs 25.4 mg/L)).

    Humanswhole-blood magnesium

  3. Magnesium is linked with a lower risk of mild cognitive impairment in humans (OR = 0.190, p = 0.0351).

    Humansfemalesn = 612

More on Mild cognitive impairment

All 134

Has magnesium been tested in people?

1 of 58 about aging

Yes: 216 trials and 234 observational studies in people have been published, and 58 trials are registered.

Every condition in trials

40 of 59

Readouts due

Stopped

  • NCT01261702TerminatedPain

    not include patient anymore

  • NCT00250965TerminatedHeart valve disease

    Terminated by DSMB; futility analysis showing Mg treatment not efficacious

Trial sponsors

Funding

6 grants since 2016

National Institute on Aging$3.6MNational Heart, Lung, and Blood Institute$2.1MJapan Society for the Promotion of Science¥18.3M

How this record changed

  1. Summary rewritten as its findings changed.

    As it read before

    Magnesium extends lifespan in mice across independent labs; human trials report both lower systolic blood pressure and no clear effect on it. One registered trial concerns aging; across all uses, 27 trials ended over a year ago without posted results or citing papers. Some authors have company affiliations, and some papers declare company ties.

  2. Summary rewritten as its findings changed.

    As it read before

    Magnesium extends lifespan in LmnaG609G/+ mice in one study; human cohorts link it to lower biological age and longer healthspan. Independent labs reproduce associations with lower diabetes risk, while blood-pressure results disagree. No registered trials concern aging; 28 across uses ended over a year ago without posted results, and some papers have company-affiliated authors or declare company ties.

  3. Summary rewritten as its findings changed.

    As it read before

    Human studies link magnesium to less PhenoAge advancement and lower sarcopenia risk; a human study reports no effect on metabolic dysfunction-associated steatotic liver disease. Four independent labs contribute eight findings, five funded by Instituto de Salud Carlos III; one has a Nestlé-affiliated author. Of 57 registered trials, 28 ended over a year ago without posted results or citing papers.

  4. Summary written from the findings.

Latest

Dec 2025: 14Jan 2026: 26Feb 2026: 22Mar 2026: 20Apr 2026: 8May 2026: 11Jun 2026: 13Jul 2026: 22Aug 2026: 18Sep 2026: 18Oct 2026: 22Nov 2026: 9203 papers in 12 months

See all 870 papers on magnesium