Micronutrients: A Practical Guide to Vitamins and Minerals

A varied diet built on vegetables, fruit, pulses, whole grains, dairy or its alternatives, and some combination of fish, eggs, meat or soy will supply nearly every vitamin and mineral a healthy adult needs, in a form the body handles better than any capsule. The shortfalls that persist in well-fed populations are few and fairly specific: vitamin D at high latitudes, iron in menstruating women, iodine in some regions, B12 on plant-based diets, folate before and during early pregnancy, and calcium in people avoiding dairy without replacing it.

That list is short enough to be worth learning, which is more useful than the alternative approach of taking a broad supplement against the possibility of a deficiency you probably do not have. Micronutrients also behave differently from the macronutrients most nutrition advice concentrates on. There is no benefit to exceeding requirement, several carry genuine toxicity above it, and the shape of the response curve is a plateau, not a slope.

What the reference numbers mean

The alphabet of reference values causes more confusion than it resolves, and the distinctions matter for interpreting any claim about your intake.

The Estimated Average Requirement is the intake meeting the needs of half a given population. The Recommended Dietary Allowance, the figure on food labels, is set roughly two standard deviations above that, so it covers about 97% of people. It is deliberately generous. Falling below the RDA on a given day is not a deficiency; it means you are below a threshold set to cover almost everybody, which most people are on most days.

The Tolerable Upper Intake Level is the ceiling, the highest chronic daily intake unlikely to cause harm. Supplements are the main route to exceeding it, since reaching the upper level for most nutrients through food alone is difficult or impossible.

The Institute of Medicine set these values across a series of Dietary Reference Intakes volumes published between 1997 and 2011, covering the B vitamins and folate, antioxidant vitamins, the trace elements, and calcium and vitamin D. The European Food Safety Authority has published its own values, which differ in places, notably for vitamin D and sodium. Where two authorities disagree, the disagreement is usually a reflection of genuine uncertainty in the underlying evidence rather than an error by one of them.

One further limitation deserves flagging. Most reference values are set against the prevention of deficiency disease, not against optimal long-term health, because the trials needed to establish the latter mostly do not exist. Claims that a particular nutrient's optimal intake is several times the RDA are usually extrapolations rather than findings.

The shortfalls that are real

Vitamin D

The one nutrient where supplementation is broadly defensible in temperate countries. Vitamin D is synthesised in skin exposed to ultraviolet B radiation, and above roughly 40 degrees of latitude that synthesis effectively stops for several months of the year. Dietary sources are limited to oily fish, egg yolks and fortified products.

The IOM's Dietary Reference Intakes for Calcium and Vitamin D (2011) set the RDA at 600 IU daily for adults up to 70 and 800 IU beyond, with an upper level of 4,000 IU. Many national bodies now advise a supplement through winter months, and UK guidance recommends 400 IU daily for the general population year-round. Higher doses have been tested extensively for effects beyond bone health, on cancer, cardiovascular disease and mortality, and the large trials have been mostly disappointing: the VITAL trial reported by Manson and colleagues in the New England Journal of Medicine (2019) found no reduction in cancer or cardiovascular events from 2,000 IU daily. Correcting a genuine deficiency matters; pushing intake far above sufficiency does not appear to.

Iron

Requirements diverge sharply by sex and life stage. Menstrual blood loss makes iron deficiency the most common nutritional deficiency worldwide, and iron-deficiency anaemia remains common in women of reproductive age, in endurance athletes with high losses, and in anyone with occult gastrointestinal bleeding.

Haem iron from meat and fish is absorbed considerably more efficiently than non-haem iron from plants, and non-haem absorption is enhanced by vitamin C eaten in the same meal and inhibited by tea, coffee and phytates. That said, iron supplementation should follow a blood test rather than a suspicion. Iron overload is genuinely harmful, haemochromatosis is not rare, and the symptoms of iron deficiency, principally fatigue, are shared by dozens of unrelated conditions. Diagnosis and dosing here belong with a clinician.

Vitamin B12

B12 occurs naturally only in animal foods and in fortified products. Anyone eating a fully plant-based diet needs a supplement or reliable fortified sources; this is not a matter of opinion or optimisation. Deficiency develops slowly because the liver stores several years' worth, which means people often feel fine for a long period while the deficit accumulates, and the neurological damage from prolonged deficiency can be irreversible. Absorption also declines with age as stomach acid production falls, making older adults a second group at elevated risk regardless of diet.

Folate, iodine and calcium

Folate matters most in a narrow window. Neural tube defects form in the first weeks of pregnancy, often before pregnancy is recognised, which is why supplementation is advised from before conception rather than after a positive test. Our guide to nutrition during pregnancy covers this in more detail.

Iodine is needed for thyroid hormone synthesis and requirements rise in pregnancy. Salt iodisation has largely solved population deficiency where it exists, but intakes have drifted down in some countries as processed food replaced iodised table salt, and plant-based milks are typically not fortified with it.

Calcium requirements are met easily on a dairy-containing diet and less easily without one. Fortified plant milks, tofu set with calcium salts, tinned fish with bones, and leafy greens other than spinach all contribute. High-dose calcium supplements have been associated with cardiovascular signals in some analyses without a settled conclusion, and food sources avoid the question.

Where supplements do nothing, and where they harm

The record of high-dose single-nutrient supplementation in healthy, well-nourished people is poor, and in a few instances actively bad.

Beta-carotene is the cautionary example. The ATBC trial, reported in the New England Journal of Medicine (1994), gave beta-carotene and vitamin E to Finnish male smokers and found higher lung cancer incidence in the beta-carotene arm. The CARET trial, reported by Omenn and colleagues in the same journal (1996), tested beta-carotene with retinol in smokers and asbestos-exposed workers and was stopped early for the same reason. Both were built on strong observational evidence that people eating more carotenoid-rich vegetables had less cancer. That evidence was real; the inference that the isolated compound was responsible was wrong.

Vitamin E has followed a similar arc, with the SELECT trial reported by Klein and colleagues in JAMA (2011) finding a higher incidence of prostate cancer in men taking high-dose vitamin E. Antioxidant supplementation around training has also been shown to blunt some of the adaptive signalling that exercise depends on, which is a plausible mechanism for why more is not better.

Multivitamins occupy a milder position. Large trials have generally found no effect on mortality or major disease outcomes in well-nourished populations. A low-dose multivitamin is unlikely to harm and functions as cheap insurance for someone with an erratic diet, but it does not correct a poor diet and should not be treated as doing so.

The fat-soluble vitamins, A, D, E and K, accumulate rather than being excreted, so upper limits apply meaningfully. Vitamin A toxicity from supplements is well documented, and high preformed vitamin A intake in early pregnancy is teratogenic, which is why liver is generally avoided then. Anyone taking anticoagulants needs to keep vitamin K intake consistent rather than variable, and interactions between supplements and prescribed medication are a pharmacist's or doctor's question, not something to resolve from an article.

Calories and micronutrients are linked

Micronutrient adequacy is harder to achieve on fewer calories, for the arithmetic reason that a smaller volume of food carries less of everything. This makes aggressive dieting a nutritional risk as well as an adherence one.

The practical consequence is that the tighter your calorie budget, the more the composition of it matters. Someone eating 1,500 kcal has considerably less room for foods that supply energy without much else than someone eating 3,000 kcal. Our TDEE calculator will give you a maintenance estimate, and the macro calculator will divide a target into protein, carbohydrate and fat, but neither says anything about micronutrient density. That is a food-choice question rather than a numbers question.

Very low-carbohydrate and other restrictive patterns need particular attention, since eliminating a whole food group removes its micronutrient contribution along with its macronutrients. Excluding fruit and most vegetables removes the main dietary sources of potassium, vitamin C and much of the folate supply. Our guide to reading nutrition labels covers how to check what a packaged food actually contributes, and the dietary fat guide covers the fat-soluble vitamins that ride along with dietary fat and are poorly absorbed without it.

Testing, and the limits of it

Blood tests are useful for a handful of nutrients and misleading for many. Serum 25-hydroxyvitamin D, ferritin with a full blood count, B12 with methylmalonic acid or holotranscobalamin where the result is borderline, and ferritin in athletes with unexplained fatigue are all reasonable measurements with established interpretations.

For most other nutrients, serum concentration is a poor proxy for status. Blood levels of calcium and magnesium are held tightly regardless of intake, so a normal result says little about tissue adequacy. Broad micronutrient panels sold direct to consumers generate results with unclear reference ranges and no established action threshold, which invites treating a number rather than a person.

A workable position

Eat a wide range of whole foods, with vegetables and fruit across several colours, some pulses, some whole grains, and a protein source at most meals. That covers the great majority of requirements without any counting.

Supplement deliberately and narrowly. Vitamin D through winter at temperate latitudes. B12 if you eat no animal products. Folate if you might become pregnant. Iron only after a blood test that shows you need it. Iodine if you have removed both dairy and iodised salt from your diet.

Treat high-dose single nutrients as pharmacology rather than nutrition, because that is what they are, and the trial record shows the risk of assuming otherwise. If you have a diagnosed condition, an absorption disorder, are taking prescribed medication, or are pregnant, your micronutrient plan belongs with a clinician who can see your bloods and your medication list. Nothing in a general guide, including this one, substitutes for that.