The short version of this whole subject: most people eating a reasonably varied diet in the UK do not need supplements, a few specific groups definitely do, and a small number of supplements cause harm at doses sold over the counter.
That is more useful than the usual framing, which treats supplements as a general good and more as better. Neither is true.
Vitamins and minerals in food come packaged with fibre, fats and other compounds that affect how they are absorbed and used, which is why a diet built on vegetables, fruit, pulses, whole grains, dairy or its alternatives, and some fish is not easily replaced by a tablet. The practical version is less about nutrient-hunting and more about range: eating a wider set of plants across a week does more than optimising any single one.
The sister site goes deeper, from two directions. There is a general account of vitamins and the body and of minerals and the body — what each one does, how the fat-soluble and water-soluble groups differ, and where the real deficiency risks are. And there is the ingredient-by-ingredient detail of what is actually in a given food, under vegetables, fruits and the wider culinary reference.
One caution when reading them. That site is written for an American audience. The biology is the same everywhere and the explanations are sound, but the official advice is not the same, and the differences are not trivial: reference intakes are set by a different body and come out at different numbers, food is fortified differently — flour, breakfast cereals and orange juice carry additions in the United States that they do not always carry here — supplements are regulated under a quite different régime, and doses are more often quoted in international units than in micrograms. Vitamin D advice in particular does not transfer at all, because it depends on latitude rather than on nutrition. Where the two disagree, the figures on this page are the UK ones.
Vitamin D is the exception to the food-first rule, because at this latitude the sun is too weak between about October and March for skin to make any, and very few foods contain much.
Ten micrograms is 400 IU; the two units appear interchangeably on packaging and the conversion is simply forty IU to the microgram, which is worth knowing when a bottle is labelled 1,000 IU and the shelf beside it says 25 micrograms.
There is an upper limit and it is easy to exceed with high-strength products: no more than 100 micrograms (4,000 IU) a day for adults and children over ten, 50 micrograms for children aged one to ten, and 25 micrograms for babies. Sunlight cannot cause an overdose; supplements can.
400 micrograms daily from before conception until the twelfth week of pregnancy substantially reduces the risk of neural tube defects such as spina bifida. The timing is the whole point: the neural tube closes in the first month, often before a pregnancy is confirmed, so starting when a test comes back positive is already late. Anyone who could become pregnant and is not actively avoiding it has a reasonable case for taking it. A higher dose is advised where there is a previous affected pregnancy, diabetes, a raised body weight or certain medicines, and that is a conversation with a GP rather than a shelf decision.
B12 occurs naturally in almost nothing that is not animal-derived. Vegans, and anyone eating very little dairy, egg, fish or meat, need either fortified foods or a supplement, and this is not optional — a deficiency causes nerve damage that is not always reversible. It also becomes harder to absorb with age and with some long-term medicines, so older people on a restricted diet are the group most often caught out.
The UK's Healthy Start scheme provides free vitamin supplements, including vitamin D for children and folic acid and vitamin D for pregnant women, to families on qualifying benefits. It is substantially under-claimed, which is worth knowing because the people it would help most are the least likely to be buying supplements anyway.
This is the part that the shelf does not advertise.
Beta-carotene and smoking. Two large trials gave heavy smokers beta-carotene — one at 20 mg a day alone, one at 30 mg a day with retinol — expecting a reduction in lung cancer. Both found more lung cancer, not less. European safety assessment puts the threshold of concern at 15 mg a day, below which intake from food, colourings and supplements is considered safe even for heavy smokers. Anyone who smokes should check the label on any antioxidant or "eye health" formulation, because beta-carotene appears in a great many of them.
Retinol and bones. Averaging more than 1.5 mg of vitamin A a day over years makes fractures more likely later in life, which matters most for post-menopausal women and older men who already carry osteoporosis risk. Liver is extraordinarily rich in it: eating liver or liver pâté weekly means skipping vitamin A supplements altogether, and fish liver oils count. In pregnancy, liver and liver products are avoided entirely.
Iron. Useful where there is a confirmed deficiency, pointless otherwise, and iron tablets are a leading cause of serious poisoning in small children, who find them attractive. They belong out of reach rather than in a kitchen drawer.
The general shape is that supplements correct deficiencies well and improve on adequacy badly. Trials of multivitamins in well-nourished populations have repeatedly failed to show reductions in heart disease or cancer, which is an unexciting result that has not reached the packaging.
"Without minerals, vitamins are worthless." A line that appears in a great deal of supplement copy and is simply wrong. Some vitamins do depend on minerals as cofactors — vitamin D metabolism involves magnesium, for instance — but vitamins are not inert without a mineral supplement, and the claim exists to sell combination products.
"Liquid vitamins absorb ten times better than tablets." No. Absorption depends on the nutrient, the dose, the rest of the meal and the individual gut, not on whether the product arrived as a liquid. Liquids are genuinely easier for people who struggle to swallow tablets, and that is the real and sufficient reason to choose one.
"The body just excretes what it doesn't need." True enough for most water-soluble vitamins at ordinary doses, and not true for the fat-soluble ones — A, D, E and K — which accumulate. It is the sentence that makes megadoses sound harmless.
The organic and inorganic distinction turns up a lot: vitamins contain carbon and minerals do not. It is chemically correct and tells a reader nothing useful about what to eat or take.
Where there is a real suspicion of deficiency — persistent fatigue, a restricted diet, a gut condition affecting absorption, heavy periods, bariatric surgery — a blood test through a GP identifies what is actually low and at what level, which is more useful than a broad supplement and cheaper than the private panels that test dozens of markers and generate findings of no clinical meaning. Several nutrients also interact with prescribed medicines, vitamin K with warfarin being the familiar one, which is another reason to raise supplements at an appointment rather than assume they are neutral.