Does magnesium help protect against osteoporosis?

There is no trial showing that it does. We found no meta-analysis of randomized trials of magnesium supplements for bone density or fractures in the general or postmenopausal population, and a systematic review of supplement trials for osteoporosis in women found no magnesium trial that met its criteria. Observational data point to a small link with bone density and no consistent link with fractures. In 73,684 postmenopausal women, those taking in more than 422.5 mg of magnesium a day from food and supplements had 3% higher hip bone density than those under 206.5 mg, and no fewer hip or total fractures.

Unsettled. Whether magnesium prevents osteoporosis is untested, and that is why the answer is open. The randomized trials that exist are small, short and mostly measure blood markers of bone turnover, and most found no change. The larger studies are observational, and they disagree about fractures. The NIH Office of Dietary Supplements says more research is needed.

What the trials found

Randomized trials of magnesium and bone

In a one-year randomized trial of healthy girls aged 8 to 14 whose diets held less than 220 mg of magnesium a day, 300 mg of magnesium oxide a day increased the gain in hip bone mineral content against placebo, at the edge of statistical significance (P = 0.05). The spine did not differ. The abstract does not say how many girls were randomized. The trial dose was chosen by researchers. The upper limit for magnesium supplements in children depends on age, as set out below.

In a crossover trial of 26 healthy women aged 20 to 28, doubling magnesium intake with a supplement for 28 days raised magnesium in urine by 36% and had no effect on markers of bone formation or breakdown. In a 12-week randomized trial of 78 adults with overweight or obesity, adding 360 mg of magnesium glycinate to vitamin D did not change bone turnover markers. Bone turnover was not the main aim of that trial.

One study often cited for magnesium in osteoporosis gave 20 postmenopausal women with osteoporosis magnesium citrate for 30 days. Markers of bone breakdown fell compared with matched women who took nothing. The abstract describes untreated controls, without placebo or randomization, which is why we rate it as observational. It measured markers only, without bone density.

People on dialysis

A meta-analysis of 8 randomized trials in 309 people on hemodialysis found that magnesium lowered parathyroid hormone, which drives kidney-related bone disease, by a weighted mean of 236.56 (95% CI 123.41 to 349.71). The abstract does not give the units. Bone density and fractures were not measured, and this applies only to dialysis.

Observational studies of intake

A 2016 meta-analysis of 12 observational studies found no significant link between high magnesium intake and hip fracture (summary effect 1.92, 95% CI 0.81 to 4.55) or fractures overall (1.01, 0.94 to 1.07). It found a borderline correlation with bone density at the femoral neck, 0.14 (0.001 to 0.28), and none at the lumbar spine. A 2022 meta-analysis in adults aged 60 and over found a small positive link with hip bone density, a pooled beta of 0.03 (0.01 to 0.06), and no link with fractures.

One cohort stands apart. In 3,765 US adults at risk of knee osteoarthritis followed for 8 years, women in the top fifth of magnesium intake had a 62% lower risk of reported fracture than those in the bottom fifth (hazard ratio 0.38, 95% CI 0.17 to 0.82). For men the ratio was 0.47 (0.21 to 1.00). Fractures were self-reported, and the abstract does not give absolute rates.

Blood magnesium

A 2023 meta-analysis of 4 observational studies with 119,755 people found that lower blood magnesium was linked to a 58% higher risk of fractures (relative risk 1.579, 95% CI 1.216 to 2.051). The abstract does not give absolute fracture rates. In 2,245 Finnish men followed for 25.6 years, low blood magnesium was also linked to more fractures, while magnesium from the diet showed no association at all. That split is the main uncertainty in this topic. Blood magnesium can reflect illness, and it does not show that eating or swallowing more would change anything.

Who should be careful

Not for everyone. The upper limit for magnesium from supplements is 350 mg a day for adults, and 65 to 350 mg for children depending on age, according to the NIH Office of Dietary Supplements. The limit does not apply to magnesium in food. If your kidneys work poorly, the risk of magnesium toxicity rises because the excess cannot be cleared. That is a question for your doctor before any supplement.

If you already take an oral bisphosphonate such as alendronate for osteoporosis, magnesium-rich supplements or medicines can reduce its absorption. The NIH fact sheet advises taking them at least 2 hours apart.

Magnesium oxide, used as a laxative and antacid, was linked to more hip fractures in a Taiwanese cohort of 52,138 people over 65: 1.23 against 0.74 hip fractures per 100 person-years, an adjusted hazard ratio of 1.66 (95% CI 1.54 to 1.80). This is observational, and people prescribed a laxative may already be frailer or fall more. In the 73,684 postmenopausal women above, the top intake group had more lower-arm and wrist fractures, which the authors connect with more activity and falls.

Pregnancy is the one place where magnesium has a recognized bone warning, and it concerns a hospital drug. In 2013 the FDA warned against intravenous magnesium sulfate for more than 5 to 7 days to stop preterm labor, because of bone problems seen in infants. In a blinded radiograph study, 6 of 11 newborns whose mothers had it for more than 7 days had changes in the upper arm bone, against none of 22 matched unexposed newborns. A later Taiwanese cohort of 4,092 births found no clear rise in infant fractures after more than 5 days of the injections (hazard ratio 1.48, 95% CI 0.59 to 3.71). A Cochrane review of 3 trials with 360 women found no clear difference in newborn osteopenia or fracture between high-dose and low-dose regimens. None of this is about magnesium from food or tablets.

What expert bodies say

The NIH Office of Dietary Supplements says more research is needed to know whether magnesium supplements can help prevent or manage osteoporosis. We found no authorized health claim or guideline on our trusted list that recommends magnesium for bone.

How we searched

What would change this answer

• A large, long-term randomized trial of magnesium in older adults with bone density or fractures as the outcome. The authors of the 2022 review say such trials are still needed, and our search found no meta-analysis of randomized trials outside dialysis.
• A trial that selects people with low blood magnesium, since that is where the observational link with fractures shows up.
• A larger repeat of the trial in girls, which found a borderline effect on hip bone mineral in a group of unreported size.

For how much magnesium you need, where to get it and what too much does, see the magnesium guide.

Sources

1. ev-mgbd-01 · Meta-analysis or systematic review · systematic review of RCTs · n = 45 products screened
   Calcium; copper; evening primrose oil; fish oils; fluoride; magnesium; manganese; strontium; vitamin D; and black, green, and oolong tea did not meet study criteria.
   
   Who: RCTs of single natural health products in women with BMD or fracture rate as outcome, published 1966 to October 2004
   Effect: magnesium did not meet study criteria; no pooled estimate for magnesium
   Certainty: Review from 2006; some foods (calcium, vitamin D) were excluded because reviews already existed, so the reason for excluding magnesium is not spelled out in the abstract.
   Ann Pharmacother, 2006 · checked 2026-10-07 · we read the abstract
2. ev-mgbd-02 · Meta-analysis or systematic review · meta-analysis of RCTs · n = 309
   Mg supplementation alone produced a negative effect on serum PTH levels (WMD = -236.56; 95% CI -349.71 to -123.41) and CIMT (WMD = -0.18; 95% CI -0.34 to -0.01).
   
   Who: hemodialysis patients with CKD mineral and bone disorder, oral or dialysate magnesium
   Effect: serum PTH WMD -236.56 (95% CI -349.71 to -123.41); CIMT WMD -0.18 (-0.34 to -0.01)
   Certainty: Surrogate markers, BMD and fractures were not measured; PTH units not stated in the abstract; dialysis only.
   J Ren Nutr, 2022 · checked 2026-10-07 · we read the abstract
3. ev-mgbd-03 · Meta-analysis or systematic review · Cochrane systematic review of RCTs · n = 360
   There were no clear differences seen between high-dose magnesium sulphate regimens compared with low-dose magnesium sulphate regimens for the secondary infant health outcomes of fetal death; neonatal death; and rate of hypocalcaemia, osteopenia or fracture; and secondary maternal health outcomes of rate of caesarean birth; pulmonary oedema; and maternal self-reported adverse effects.
   
   Who: women in preterm labor given IV magnesium sulfate for tocolysis, high-dose vs low-dose regimens
   Effect: no clear differences in infant hypocalcemia, osteopenia or fracture; no long-term outcomes reported
   Certainty: Medication IV, not food and not a supplement; doses are compared, not magnesium versus nothing; no study of high quality.
   Cochrane Database Syst Rev, 2015 · checked 2026-10-07 · we read the abstract
4. ev-mgbd-04 · Observational data · systematic review and meta-analysis of observational studies · n = 12 studies
   We found that high intakes of magnesium were not significantly associated with risk of total hip fracture (summary effect size 1.92; 95 % CI 0.81, 4.55) or total fractures (1.01; 0.94-1.07).
   
   Who: observational studies of dietary magnesium intake and fracture, searched to January 2015
   Effect: total hip fracture summary effect 1.92 (95% CI 0.81-4.55); total fractures 1.01 (0.94-1.07)
   Certainty: Observational data; not about supplements.
   Osteoporos Int, 2016 · checked 2026-10-07 · we read the abstract
5. ev-mgbd-05 · Observational data · systematic review and meta-analysis of observational studies · n = 9 effect sizes (femoral neck)
   Based on nine effect sizes, we found a marginally significant association between magnesium intake and femoral neck BMD (0.14; 0.001, 0.28). However, no significant correlation was found between magnesium intake and BMD in lumbar spine (0.09; -0.01, 0.19).
   
   Who: observational studies reporting correlation between magnesium intake and BMD
   Effect: femoral neck BMD pooled r 0.14 (0.001-0.28); lumbar spine r 0.09 (-0.01 to 0.19)
   Certainty: Correlation, not an effect of an intervention.
   Osteoporos Int, 2016 · checked 2026-10-07 · we read the abstract
6. ev-mgbd-06 · Observational data · systematic review and meta-analysis of observational studies · n = 4 studies in hip BMD meta-analysis
   Meta-analysis of four studies showed a significant positive association between magnesium intake and hip BMD (pooled beta: 0.03, 95% CI: 0.01-0.06, p < 0.05).
   
   Who: adults aged 60 years or more; six cohort, one case-control and five cross-sectional studies
   Effect: hip BMD pooled beta 0.03 (95% CI 0.01-0.06); no associations with other sites or fractures
   Certainty: Observational data; the authors found no randomized studies in people 60+.
   Bone, 2022 · checked 2026-10-07 · we read the abstract
7. ev-mgbd-07 · Observational data · systematic review and meta-analysis of observational studies · n = 12 studies
   Next, large and long-term randomized controlled trials in older adults are needed to determine whether an increase in magnesium (supplementation) intake can improve bone health.
   
   Who: adults aged 60 years or more
   Effect: no RCTs in older adults identified; need for long-term RCTs
   Certainty: A statement of a gap; our PubMed search of 2026-10-07 (37 records) also found no MA of RCTs of magnesium on MCFA.
   Bone, 2022 · checked 2026-10-07 · we read the abstract
8. ev-mgbd-08 · Observational data · systematic review and meta-analysis of observational studies · n = 119755
   We found that lower serum Mg concentrations were associated with a significantly higher risk of incident fractures (RR = 1.579; 95%CI: 1.216-2.051; p = 0.001; I2 = 46.9%).
   
   Who: observational studies of serum magnesium and incident fractures, searched to May 2022
   Effect: RR 1.579 (95% CI 1.216-2.051), I2 = 46.9% for lower serum Mg
   Certainty: Serum magnesium is a marker of health status, not of intake; causality not established.
   Nutrients, 2023 · checked 2026-10-07 · we read the abstract
9. ev-mgbd-09 · Randomized controlled trial(s) · randomized double-blind placebo-controlled trial · n = small cohort (from over 120 screened)
   Significantly increased accrual (P = 0.05) in integrated hip BMC occurred in the Mg-supplemented vs. placebo group.
   
   Who: healthy Caucasian girls aged 8-14 with dietary Mg below 220 mg/d; 300 mg elemental Mg (oxide) daily vs placebo for 12 months
   Effect: greater integrated hip BMC accrual (P = 0.05); lumbar spine not significantly different
   Certainty: Borderline significance, small sample; number randomized not stated in the abstract.
   J Clin Endocrinol Metab, 2006 · checked 2026-10-07 · we read the abstract
10. ev-mgbd-10 · Observational data · controlled trial with matched untreated controls (indexed as RCT) · n = 20
    Serum osteocalcin levels were significantly increased (p < 0.001) and urinary deoxypyridinoline levels were decreased (p < 0.001) in the Mg-supplemented group.
    
    Who: postmenopausal osteoporotic women; magnesium citrate 1,830 mg/day for 30 days vs matched controls without medication
    Effect: serum osteocalcin up (p < 0.001), urinary deoxypyridinoline down (p < 0.001), iPTH down (p < 0.05)
    Certainty: Matched control, no placebo; PubMed labels it an RCT, but the abstract does not describe randomization. ODS describes this same study as "versus placebo," a discrepancy with the abstract. Markers, not MSCT.
    Biol Trace Elem Res, 2010 · checked 2026-10-07 · we read the abstract
11. ev-mgbd-11 · Randomized controlled trial(s) · randomized double-blind placebo-controlled crossover trial · n = 26
    RESULTS: Increasing Mg intake from the usual level (11 mmol/d) to 22 mmol/d for 28d increased urinary excretion of Mg by 36% and erythrocyte Mg content by 5% but had no effect on serum Mg, Ca, PTH, osteocalcin or bone-specific alkaline phosphatase (biomarkers of bone formation), urinary pyridinium crosslinks of collagen (biomarkers of bone resorption), or on blood pressure.
    
    Who: healthy women aged 20-28; usual diet (about 11 mmol Mg/d) vs diet plus 10 mmol/d Mg(OH)2 for 28 days
    Effect: no effect on serum Mg, Ca, PTH, osteocalcin, bone-specific alkaline phosphatase or urinary pyridinium crosslinks
    Certainty: Short, young women without deficiency.
    Eur J Clin Nutr, 1999 · checked 2026-10-07 · we read the abstract
12. ev-mgbd-12 · Randomized controlled trial(s) · randomized double-blind placebo-controlled trial · n = 78
    Combined supplementation was not associated with short-term improvements in glycemic indices or bone turnover markers in participants who were overweight and obese in our study.
    
    Who: adults with overweight or obesity; vitamin D3 1000 IU plus 360 mg magnesium glycinate vs vitamin D vs placebo for 12 weeks
    Effect: no differences in osteocalcin or other bone turnover markers between groups
    Certainty: The primary aim is glycemia; markers, not MCT.
    Nutr Res, 2023 · checked 2026-10-07 · we read the abstract
13. ev-mgbd-13 · Observational data · prospective cohort · n = 73684
    Baseline hip BMD was 3% higher (P < 0.001), and whole-body BMD was 2% higher (P < 0.001), in women who consumed >422.5 compared with <206.5 mg Mg/d. However, the incidence and RR of hip and total fractures did not differ across quintiles of magnesium.
    
    Who: postmenopausal women, Women's Health Initiative Observational Study; magnesium from food plus supplements
    Effect: hip BMD +3%, whole-body BMD +2% (highest vs lowest intake); hip and total fracture RR not different; lower-arm/wrist fracture HR 1.23 (1.07-1.42) in top quintile
    Certainty: The authors link the increase in wrist fractures to greater activity and falls.
    Am J Clin Nutr, 2014 · checked 2026-10-07 · we read the abstract
14. ev-mgbd-14 · Observational data · prospective cohort · n = 3765
    After adjusting for fourteen potential confounders at baseline and taking those with lower Mg intake as reference (Q1), men (hazard ratio (HR) 0·47; 95 % CI 0·21, 1·00, P=0·05) and women (HR 0·38; 95 % CI 0·17, 0·82, P=0·01) in the highest quintile reported a significantly lower risk for fracture.
    
    Who: Osteoarthritis Initiative participants, mean age 60.6; dietary plus supplemental Mg by FFQ
    Effect: highest vs lowest quintile: men HR 0.47 (0.21-1.00), women HR 0.38 (0.17-0.82)
    Certainty: Fractures by self-report; a population at risk of osteoarthritis.
    Br J Nutr, 2017 · checked 2026-10-07 · we read the abstract
15. ev-mgbd-15 · Observational data · prospective cohort · n = 2245
    There was no evidence of an association of dietary magnesium intake with risk of any fractures.
    
    Who: men aged 42-61, Kuopio Ischemic Heart Disease cohort
    Effect: bottom vs top quartile serum Mg: adjusted HR 1.80 (1.10-2.94) for total fractures; no association for dietary Mg
    Certainty: The gap between serum and dietary magnesium is the main uncertainty of the topic.
    Eur J Epidemiol, 2017 · checked 2026-10-07 · we read the abstract
16. ev-mgbd-16 · Observational data · retrospective population-based cohort · n = 52138
    MgO use was identified as a risk factor for hip fracture in both univariate [crude HR, 1.68; 95% confidence interval (CI), 1.55-1.81; p < 0.001] and multivariate [adjusted HR (aHR), 1.66; 95% CI, 1.54-1.80; p < 0.001] Cox proportional hazards regression models.
    
    Who: adults over 65 with and without magnesium oxide use, propensity matched, Taiwan national insurance 1996-2013
    Effect: adjusted HR 1.66 (95% CI 1.54-1.80); 1.23 vs 0.74 hip fractures per 100 person-years
    Certainty: Magnesium oxide here is a laxative medication; confusion with reasons for prescribing is possible (fragility, falls).
    Osteoporos Int, 2020 · checked 2026-10-07 · we read the abstract
17. ev-mgbd-17 · Observational data · retrospective population-based cohort · n = 4092
    ABSTRACT: In 2013, the U.S. Food and Drug Administration issued a safety warning that cautioned against using magnesium sulfate (MgSO4) injections for more than 5 to 7 days to stop preterm delivery due to the bone problems subsequently observed in infants.
    
    Who: pregnant women with live singleton births, Taiwan 2012-2014; MgSO4 injection for more than 5 days vs no tocolytics
    Effect: infant fracture adjusted HR 1.48 (95% CI 0.59-3.71)
    Certainty: Medication IV in pregnancy, not food; wide CI.
    Medicine (Baltimore), 2021 · checked 2026-10-07 · we read the abstract
18. ev-mgbd-18 · Observational data · matched controlled radiographic study · n = 33
    On evaluation of the proximal humeri, six neonates had radiographic abnormalities consisting of transverse radiolucent and/or sclerotic bands; all six were in the exposed group.
    
    Who: neonates exposed to maternal IV magnesium sulfate tocolysis for more than 7 days vs gestational-age matched unexposed neonates
    Effect: 6/11 exposed vs 0/22 unexposed with radiographic humeral abnormalities (P < .001)
    Certainty: Small sample; the clinical significance of the changes has not been established.
    Obstet Gynecol, 1991 · checked 2026-10-07 · we read the abstract
19. ev-mgbd-19 · Position of an expert body · government fact sheet
    However, more research is needed to determine whether magnesium supplements can help prevent or manage osteoporosis.
    
    Who: general population
    Effect: no recommendation for magnesium supplements against osteoporosis
    Certainty: Position without its own analysis.
    NIH Office of Dietary Supplements, Magnesium Fact Sheet for Health Professionals · checked 2026-10-07 · we read the section
20. ev-mgbd-20 · Position of an expert body · government fact sheet
    The Tolerable Upper Intake Level for supplemental magnesium is 350 mg for adults, and it ranges from 65 to 350 mg for children and adolescents, depending on age.
    
    Who: adults and children; supplemental magnesium only
    Effect: UL 350 mg/day supplemental Mg in adults; 65-350 mg in children
    Certainty: UL does not apply to magnesium from food.
    NIH Office of Dietary Supplements, Magnesium Fact Sheet for Health Professionals · checked 2026-10-07 · we read the section
21. ev-mgbd-21 · Position of an expert body · government fact sheet
    Magnesium-rich supplements or medications can decrease the absorption of oral bisphosphonates, such as alendronate (Fosamax), that are used to treat osteoporosis [61].
    
    Who: people taking oral bisphosphonates for osteoporosis
    Effect: reduced bisphosphonate absorption; separate by at least 2 hours
    Certainty: Position without its own analysis.
    NIH Office of Dietary Supplements, Magnesium Fact Sheet for Health Professionals · checked 2026-10-07 · we read the section
22. ev-mgbd-22 · Position of an expert body · government fact sheet
    The risk of magnesium toxicity increases with impaired renal function or kidney failure because the ability to remove excess magnesium is reduced or lost [1,29].
    
    Who: people with impaired renal function or kidney failure
    Effect: higher risk of hypermagnesemia
    Certainty: Position without its own analysis.
    NIH Office of Dietary Supplements, Magnesium Fact Sheet for Health Professionals · checked 2026-10-07 · we read the section