Magnesium

What it is

Magnesium is a mineral the body needs, and people who eat more of it in their diet have lower rates of type 2 diabetes, stroke and heart failure in long-term studies [1]. Supplements come as different salts — such as oxide, citrate, gluconate and aspartate — at doses that vary widely between trials [2] [3].

What the research says

Blood pressure. A 2025 meta-analysis of 38 randomised trials with 2,709 people found magnesium lowered blood pressure by 2.8/2.1 mmHg compared with placebo [3]. The drop was larger in people with high blood pressure already on medication (7.7 mmHg systolic) and in people with low magnesium levels, while in people with normal blood pressure it was not significant [3].

Type 2 diabetes. Across 18 trials in people with type 2 diabetes, about 500 mg a day lowered HbA1c by an estimated 0.73 percentage points, though the authors judged the evidence not yet sufficient for clinical guidelines [4]. In 40 cohort studies with over a million people, each extra 100 mg of dietary magnesium a day was linked to a 19% lower risk of developing type 2 diabetes [1].

Heart and stroke. The same cohort analysis linked higher magnesium intake to lower risk of stroke and heart failure, but not of coronary heart disease or overall cardiovascular disease [1].

Sleep. In three small trials of 151 older adults with insomnia, magnesium shortened the time to fall asleep by about 17 minutes, but the evidence was of low to very low quality [5]. A 2026 review of 12 trials found inconsistent results and concluded that magnesium is not supported as a routine insomnia treatment [6]. A 2025 placebo-controlled trial of 155 adults with poor sleep found a larger improvement in insomnia scores with 250 mg of magnesium bisglycinate a day for four weeks, though the effect was small (Cohen's d 0.2) [7].

Migraine. A 2025 meta-analysis found magnesium reduced migraine attacks and monthly migraine days [8], while an earlier analysis of three trials (226 people) found no significant effect on severity and called the evidence insufficient [9].

Mood. In an open-label trial of 126 adults with mild-to-moderate depression, 248 mg of magnesium a day for six weeks improved depression scores by 6 points and anxiety scores by 4.5 points [10]. A systematic review found the overall evidence on magnesium and mental health mixed [11].

Testosterone. In one 4-week trial, magnesium increased free and total testosterone in athletes and sedentary men [12]. But an 8-week trial of a zinc-magnesium supplement (ZMA) in 42 resistance-trained men found no changes in anabolic or catabolic hormones [13], and a 2024 systematic review of 52 booster studies found most products did not raise total testosterone, listing no zinc or magnesium product among its few possibly effective ones [14].

Muscle cramps. A Cochrane review of 11 trials concluded that magnesium is unlikely to give clinically meaningful relief to older adults with muscle cramps [15]. In four trials with 332 pregnant women, it did not reduce leg cramps either [16].

Doses used in studies

Blood pressure trials used 82–637 mg of elemental magnesium a day (median 365 mg) for a median of 12 weeks [3]. Glycaemic benefits were estimated at around 500 mg a day [4], and the depression trial used 248 mg a day [10]. The testosterone trial used 10 mg of magnesium per kilogram of body weight a day for four weeks [12].

Safety and interactions

• Side effects: mostly digestive, especially diarrhoea, reported by 11–37% of people taking magnesium versus 10–14% on placebo; serious side effects were no more common than with placebo [15].
• Kidney problems and older age: magnesium oxide needs care with dose and blood levels in older people and people with impaired kidney function [17].
• Medicines: magnesium-containing antacids can bind tetracycline and fluoroquinolone antibiotics and reduce their absorption, so they are separated by taking the antacid 4 hours before or 2 hours after the antibiotic; bisphosphonates may be affected in the same way [18].
• Pregnancy: a Cochrane review of 10 trials with 9,090 women found not enough high-quality evidence that magnesium supplements help in pregnancy, and a possible rise in neonatal deaths in one trial, which the authors said should be read with caution [2].
• Upper limits: this page does not cover official upper intake limits; trial doses are listed above.

Caveats and open questions

Many trials are small, short and use different magnesium salts and doses [6] [3]. Benefits seem largest in people who are low in magnesium to begin with [3].