Magnesium in food: which foods supply it, and who falls short
Where magnesium really comes from in a British diet, what the UK reference intakes are, and who the survey data says is falling short.
On this page
Magnesium mainly comes from plant foods: seeds, nuts, wholegrains, pulses and dark green vegetables. The NHS puts the UK reference intake at 300 mg a day for men aged 19 to 64 and 270 mg for women, and says most people should be able to get what they need from food alone (NHS). In national survey data, average intakes sit close to those figures for men and below them for women (NDNS). A minority fall well short, and the gap is widest among teenagers.
The short answer
- Seeds and nuts are the most concentrated everyday sources in a British diet. Thirty grams of pumpkin seeds supplies well over half an adult's daily reference intake, more than any other everyday snack in the table below.
- National survey data suggests most UK adults are not short of magnesium, but a persistent minority are. The proportion below the lower threshold is much higher in girls aged 11 to 18 (NDNS).
- Food first. Supplements are a question for a pharmacist or GP rather than a guess, particularly for anyone on long-term medication.
This sits in our nutrition section, which covers the nutrients that come up most often in arguments about snacking.
What does magnesium do?
Most of the body's magnesium sits in bone. Much of the rest is inside cells, where it is involved in energy metabolism, protein synthesis and normal muscle and nerve function (NHS). Those are on the authorised health claims list that Great Britain carried over from the EU, along with normal psychological function and the reduction of tiredness and fatigue (EFSA health claims; GB labelling rules).
That list is narrower than it sounds. An authorised claim means the nutrient is needed for a normal process. It does not mean more of it improves that process in someone who already has enough. Absorption is partial too. Only a fraction of the magnesium in a meal is taken up, and the kidneys adjust how much is retained.
How much do adults need?
The reference nutrient intake is 300 mg a day for men aged 19 to 64 and 270 mg for women. The NHS notes that high doses from supplements can cause diarrhoea (NHS).
Two other numbers often cause confusion. The European Food Safety Authority set adequate intakes on a different basis (EFSA), and the figure on a pack's nutrition panel is the labelling reference value, one number for all adults rather than a personal target (food labelling rules). A pack claiming to be a source of magnesium is measured against that, not the NHS figure.
Surveys use a third number: the lower reference nutrient intake, an intake low enough that only a small proportion of people would be adequately supplied at it. Being below it for a day means little. Being below it habitually is the signal researchers look for.
Which foods supply the most magnesium?
Seeds and nuts lead on density. Wholegrains, pulses and green vegetables supply less per 100 g but often more in practice, because portions are larger.
| Food | Typical portion | Magnesium per 100 g | Per portion |
|---|---|---|---|
| Pumpkin seeds | 30 g | ~550 mg | ~165 mg |
| Brazil nuts | 30 g | ~380 mg | ~114 mg |
| Sunflower seeds | 30 g | ~350 mg | ~105 mg |
| Almonds | 30 g | ~270 mg | ~81 mg |
| Dark chocolate, 70% cocoa | 25 g | ~200 mg | ~50 mg |
| Peanuts | 30 g | ~190 mg | ~57 mg |
| Porridge oats, dry weight | 40 g | ~120 mg | ~48 mg |
| Spinach, boiled | 80 g | ~80 mg | ~64 mg |
| Wholemeal bread | 70 g (2 slices) | ~75 mg | ~53 mg |
| Brown rice, boiled | 180 g | ~45 mg | ~81 mg |
| Chickpeas, cooked | 120 g | ~45 mg | ~54 mg |
| Semi-skimmed milk | 200 ml | ~11 mg | ~22 mg |
Figures are rounded from McCance and Widdowson's Composition of Foods Integrated Dataset (CoFID, gov.uk). Values vary with variety and processing, so treat them as orders of magnitude.
Two points matter. Seeds are in a class of their own, especially pumpkin seeds. And refining matters as much as the food itself: white flour and white rice lose most of the magnesium held in the germ and bran, which is why wholemeal bread carries several times what white does.
Who in the UK falls short?
The National Diet and Nutrition Survey tracks this. Across recent published rounds, the share of adults aged 19 to 64 with intakes below the lower reference threshold has sat at roughly one in ten, and has run higher in women than men. The standout group is adolescents: among girls aged 11 to 18 the share has been several times the adult figure (NDNS, gov.uk).
Diet is not the only route to low magnesium. Some medicines, heavy alcohol use and conditions that impair absorption all matter too. UK medicines regulators have flagged low blood magnesium as a possible effect of long-term proton pump inhibitor treatment, which is a question for a pharmacist rather than one to settle at home (Drug Safety Update, gov.uk). Hard water adds a little in some areas, but it is a minor share of intake for most (WHO).
Is magnesium linked to type 2 diabetes?
There is a link, but it is weaker than the headlines suggested. Prospective cohort studies repeatedly find that people with higher magnesium intakes develop type 2 diabetes at a lower rate, and the association has held across several meta-analyses (PubMed).
The problem is what magnesium tends to come with. It is found in wholegrains, nuts, pulses and vegetables, and people who eat more of those foods differ from people who eat fewer in ways that are hard to adjust away: fibre, diet quality, body weight, activity, smoking. An association like this cannot separate the mineral from the diet it arrives in. More on how we weigh evidence in how we write.
Trials of magnesium supplements on blood glucose markers exist, but they are small, short and inconsistent. The whole-food version of the question is better supported: the effect of nuts on blood sugar after a meal rests on more direct evidence than anything about the mineral in isolation. Anyone managing diabetes should take this to their own clinical team (Diabetes UK).
What does the evidence not show?
This is the part that usually gets left out.
- Magnesium is required for normal function. That is not a therapeutic effect, and nothing in the survey or cohort data supports a stronger reading.
- The diabetes association is not causal evidence. Cohorts show correlation. The supplement trials that would test causation are too small and too varied to settle it.
- There is no good routine test. Serum magnesium is the usual measurement and reflects total body stores poorly, because the body defends blood levels at the expense of tissue. So the honest answer to how many people are actually short is that nobody knows: the survey figures measure intake, not status.
- Survey intakes are self-reported. Food diaries under-record, so the true distribution is uncertain in both directions.
- Claims about sleep, cramps and anxiety are mixed at best. They are common, rarely supported by the sort of evidence that would justify them, and not on the authorised claims list a UK pack is allowed to draw on (EFSA register).
Should anyone take a supplement?
Food first is the defensible position, and for most people on a reasonable mixed diet it is enough. Seeds, nuts, wholegrains, beans and greens move intake more than a tablet does, and they bring fibre with them. Roasted chickpeas and a handful of seeds are among the cheapest levers available.
Supplements are a different question, and not one to settle from an article. High doses can cause diarrhoea, magnesium interacts with some medicines, and people with kidney problems need specific advice (NHS). A pharmacist or GP can answer it with the history in front of them, and a registered dietitian can look at the diet as a whole (BDA).
Where this came from
- NHS: vitamins and minerals
- National Diet and Nutrition Survey, gov.uk
- CoFID food composition dataset, gov.uk
- EFSA: dietary reference values
- EFSA: health claims
- Food labelling and packaging, gov.uk
- Drug Safety Update, gov.uk
- WHO: drinking water
- PubMed: magnesium and type 2 diabetes
- Diabetes UK
- British Dietetic Association