What the Research Says
What the Magnesium Research Actually Shows
Magnesium is involved in over 300 enzymatic reactions and deficiency is common. The evidence on sleep, anxiety, and athletic performance is solid — with important caveats about form and dosing.
Why Magnesium Gets Attention
Magnesium is the fourth most abundant mineral in the body and a cofactor for ATP synthesis, protein production, and neuromuscular signalling. Population studies consistently find that a majority of adults consume below the RDA (310–420mg/day), making it one of the most widespread nutrient gaps in developed countries. Deficiency — even subclinical deficiency — has broad physiological effects.
What Replicates Strongly
Magnesium glycinate and threonate improve sleep quality in studies of deficient populations. Multiple RCTs in older adults with poor sleep (a population with commonly low magnesium) show improvements in sleep efficiency, sleep onset latency, and cortisol with supplementation. The mechanism involves GABA-A receptor activation (magnesium potentiates GABA signalling) and melatonin regulation. The effect is larger in those with lower baseline magnesium.
Magnesium reduces cortisol response to physical and psychological stressors. Stress depletes intracellular magnesium while simultaneously requiring it for adrenal function — a vicious cycle. RCTs show that magnesium supplementation blunts cortisol and adrenaline response to standardised stressors, with effects most pronounced in individuals with high baseline stress and low dietary intake. The anxiolytic effect tracks closely with cortisol modulation.
Athletic performance benefits are clearest for endurance and high-sweat activities. Sweat losses of magnesium are significant in aerobic exercise — approximately 0.5–1mg per litre of sweat. Multiple sports nutrition RCTs find that athletes in calorie-restricted or high-sweat conditions show performance and recovery improvements with 300–400mg supplementation. Effects on untrained individuals or those with adequate dietary intake are less consistent.
Magnesium L-threonate penetrates the blood-brain barrier more effectively and may support cognitive function. Animal studies showed dramatic improvements in synaptic density and learning with this form; small human RCTs show improvements in cognitive measures in older adults. This form is more expensive and evidence is thinner than for general magnesium biomarkers, but it's the best candidate for neurological applications.
Form matters significantly for bioavailability and tolerability. Magnesium oxide (common in cheap supplements) has poor bioavailability (~4%) and causes GI distress. Glycinate, threonate, and malate forms have superior bioavailability (20–35%) and are better tolerated. Most studies showing benefits use chelated forms; trials using oxide often show weaker effects.
What the Research Can't Tell You
Individual magnesium status (not easily tested via serum level, which poorly reflects intracellular stores), optimal dose, and which symptoms track deficiency vary considerably. A practical approach: 200–400mg of glycinate before bed for 4–6 weeks, tracking sleep quality, anxiety, and muscle cramping. These are the most reliable indicators of magnesium status improvement.