Why modern diets keep coming up short on magnesium — and what the research says to do about it

Magnesium is the fourth-most-abundant mineral in the human body and a cofactor for more than 300 enzymes — the ones that run ATP production, blood-pressure regulation, nerve conduction, muscle contraction, glucose handling, and DNA repair. It is also the mineral that, on the latest UK survey data, the largest share of the population is failing to get enough of.

The picture in modern research is consistent enough that magnesium specialists have started using the phrase “subclinical deficiency” to describe it: not the textbook hypomagnesaemia that lands people in hospital, but a steady, low-grade shortfall that does not show up on a routine blood test and quietly stacks the odds for hypertension, type 2 diabetes, poor sleep, and cardiovascular disease over decades.

What the UK survey data actually shows

The National Diet and Nutrition Survey (NDNS) is the rolling government survey of what people in the UK actually eat. Its most recent multi-year report (Years 1–9, published by Public Health England) sets a clear benchmark: the Lower Reference Nutrient Intake (LRNI) is the intake below which almost everyone is, by definition, getting too little to meet their needs. For magnesium that floor is roughly 190 mg/day for men and 150 mg/day for women.

The proportion of the UK population below that LRNI is not small:

• Teenagers (11–18): about 38% have magnesium intakes below the LRNI — 27% of boys and 50% of girls.
• Young adults (19–29): roughly one in five sit below the LRNI for magnesium.
• Women aged 19–64: 11% below the LRNI, with intakes essentially unchanged across the nine-year rolling survey.

The Reference Nutrient Intake — the target the NHS Eatwell Guide assumes for a healthy adult — sits at 300 mg for men and 270 mg for women. Average UK intake hovers a little below that for most age groups, which means a much larger share of the population is short of the target than is short of the floor.

Why magnesium falls out of a modern diet

Magnesium is concentrated in the parts of plant foods that modern processing tends to remove, and in soils that modern farming tends to deplete. Several structural shifts push intakes down at once.

Refining. Magnesium sits in the bran and germ of cereal grains. Milling wheat to white flour removes roughly 80% of the magnesium content; polishing brown rice to white rice strips out a similar share. The UK consumes more ultra-processed food, by share of energy, than any other country in Europe, and refined cereals are a large part of that.

Soil and crop trends. The 2022 UK update by McCance & Widdowson’s data custodians, building on the earlier Mayer analysis, compared the mineral content of common UK fruits and vegetables across 1940, 1991 and 2019. Magnesium content in many vegetables had fallen meaningfully across the period, alongside larger declines in copper, iron and other minerals. A US analysis by Donald Davis and colleagues (2004) made the same point from USDA data: a slow decline in mineral density across the second half of the twentieth century.

What replaces whole foods. Sugar, fat and refined starch carry essentially no magnesium. Calories that come from ultra-processed foods displace the leafy greens, legumes, nuts, seeds and whole grains that supply the bulk of dietary magnesium — so two diets with the same calorie count can be hundreds of milligrams apart on this nutrient alone.

Medications and lifestyle factors that quietly pull magnesium out

Intake is only one side of the ledger. A second set of modern factors increases how much magnesium leaves the body or prevents what is eaten from being absorbed.

Proton pump inhibitors. In March 2011 the US Food and Drug Administration issued a safety communication requiring new warnings on all prescription PPIs — omeprazole, esomeprazole, lansoprazole, pantoprazole and rabeprazole — after a series of reports of clinically serious hypomagnesaemia in patients on long-term PPI therapy. Most cases occurred after more than a year of use; symptoms included tetany, seizures, arrhythmia and abnormal QT interval. PPIs are one of the most commonly prescribed drug classes in the UK, including over-the-counter.

Diuretics. Loop and thiazide diuretics increase urinary magnesium losses; in older patients on long-term treatment, this is a recognised cause of low-level deficiency.

Alcohol and caffeine. Heavy alcohol intake causes both poor absorption and increased renal excretion of magnesium; high caffeine intake produces a smaller but measurable increase in urinary loss.

Chronic stress. Sustained sympathetic activation drives magnesium out of cells and into urine. This is not a wellness slogan: it is a measurable physiological response, and one of the reasons people under prolonged psychological load can become magnesium-depleted on what looks like a reasonable diet.

Why a normal blood test is not reassuring

The single most important point in the modern magnesium literature is how little a routine serum magnesium result tells you. Roughly 99% of the body’s magnesium sits inside cells and bone; less than 1% circulates in serum, and that compartment is held within a narrow range by the kidneys at the expense of cellular stores. Serum can read perfectly normal in someone whose intracellular and bone magnesium have been falling for years.

This is the argument DiNicolantonio, O’Keefe and Wilson made in their 2018 Open Heart review, titled Subclinical magnesium deficiency: a principal driver of cardiovascular disease and a public health crisis. Their case is that because the standard test misses it, and because the dietary and medication trends above are pulling intake down across whole populations, a large share of adults in modern societies sit in a chronic, low-grade deficit that the clinical system is not set up to detect. More sensitive measures — red-blood-cell magnesium, ionised magnesium, or the more recent magnesium depletion score that combines diuretic use, PPI use, eGFR and alcohol intake — pick up depletion that serum does not.

What the trial evidence shows when intake goes up

The most useful way to read the magnesium literature is to look at where supplementation produces a measurable effect in randomised trials. Three areas have the strongest evidence.

Blood pressure. A 2024 umbrella meta-analysis of ten review papers covering 8,610 participants (Frontiers in Nutrition) found magnesium supplementation lowered systolic blood pressure by roughly 1.3 mmHg and diastolic by 1.4 mmHg on average. The effect was larger in people who were hypertensive at baseline or who started with low intakes — a pattern that fits the deficiency-correction interpretation rather than a pharmacological one.

Glucose control in type 2 diabetes. A 2024 systematic review and dose-response meta-analysis in the British Journal of Nutrition pooled controlled trials of oral magnesium in type 2 diabetes and found modest but consistent improvements in fasting glucose and insulin sensitivity, with stronger effects at higher doses, longer durations, and in older participants. The signal is strongest in people who started with low magnesium status.

Sleep. Magnesium supplementation has a smaller evidence base for sleep than the marketing suggests, but the better-controlled trials — including a 2024 randomised trial in adults with disturbed sleep and metabolic disease — show improvements in insomnia symptoms and sleep-onset latency. The mechanism is plausible (magnesium is a natural NMDA antagonist and GABA-A modulator), but the effect sizes are real rather than dramatic.

Eating to the target before supplementing to it

The first thing to do with a magnesium gap is close it from food. Per-100 g, the densest UK-typical sources are:

• Pumpkin seeds (~540 mg), sunflower seeds (~325 mg), chia seeds (~335 mg)
• Almonds (~270 mg), cashews (~290 mg), Brazil nuts (~375 mg)
• Dark chocolate >70% (~230 mg)
• Cooked spinach (~80 mg), Swiss chard (~85 mg), edamame (~65 mg)
• Black beans, kidney beans, chickpeas (~60–70 mg cooked)
• Wholegrain oats, brown rice, wholegrain bread (~45–90 mg)
• Oily fish — mackerel, salmon (~30 mg)

A reasonable everyday rule is to make sure two of these turn up at most meals — oats and seeds at breakfast, leafy greens or legumes at lunch and dinner, a handful of nuts as a snack. That alone moves most people from the LRNI region up past the RNI without supplementation.

When a supplement makes sense, and which form

Supplementing is most clearly justified in three groups: people on long-term PPIs or diuretics, people whose diets are heavy on refined and ultra-processed food and unlikely to change in the short term, and older adults with reduced appetite. Doses in the trial literature mostly sit between 200 and 400 mg of elemental magnesium per day; the EU’s tolerable upper intake for supplemental magnesium is 250 mg/day above food, which is a cautious limit set against the threshold for osmotic diarrhoea rather than systemic harm.

Form matters less than the marketing suggests but is not irrelevant. Magnesium citrate and magnesium bisglycinate are the two with the best human-absorption data and the fewest gastrointestinal complaints; bisglycinate is the form most often used in sleep and anxiety trials because it is gentle and well-tolerated. Magnesium oxide is cheap and high in elemental magnesium per tablet but poorly absorbed and the most likely to cause loose stools. The expensive “L-threonate” form has interesting preclinical data on brain magnesium but a thin human-trial base; it is not obviously better than citrate or bisglycinate at this point.

As with any supplement category that sees high-volume cheap products, what is in the capsule is not always what is on the label. Independent batch testing — through programmes like NSF, Informed Sport, or USP Verified — confirms the dose, screens for heavy-metal contamination, and (in the case of Informed Sport) checks against the WADA banned-substances list. For a daily mineral that most people on a UK diet have a plausible reason to take, that level of verification is the cheap part of the decision.

You can browse third-party-tested magnesium products on Certwell or read more about how each certification compares.