Dietary Supplements: Overdose, Interactions and What Actually Works
Half your patients are taking something you didn't prescribe — a multivitamin, a fish-oil capsule, a herbal remedy a friend swore by. They don't think of it as medicine, so they don't mention it, and you don't ask. That is a mistake. A supplement is a drug: it has a dose, it has adverse effects, and it interacts with the very medicines you carefully chose. Most of them do nothing measurable in a well-nourished person. A few genuinely save lives when there is a real deficiency. And a handful — St John's wort at the top of the list — can quietly wreck a treatment plan. The skill is telling the three apart.
A 34-year-old woman comes to renew her combined oral contraceptive. She feels well; her drug history, as recorded, is a single line — "COCP." Six weeks later she is pregnant. Nothing failed mechanically. What her chart never captured was the small brown bottle in her bag: St John's wort, bought over the counter for low mood, taken every morning. It is a potent inducer of the liver enzymes that clear the pill, and it had quietly dropped the hormone levels below the threshold that prevents ovulation. She never mentioned it because, to her, it wasn't a drug — it was "just a herb." The lesson is not about one plant. It is that the medicines you don't ask about are exactly the ones that harm your patient, and a supplement is a drug whether or not the patient calls it one.
A supplement is a drug — treat it like one
The word "natural" carries no pharmacological meaning. Every principle from the Principles of Pharmacology chapter applies to a supplement exactly as it does to a prescription: it has a dose–response curve, a therapeutic window, adverse effects at the top of that window, and pharmacokinetic interactions with other drugs. The one thing it usually lacks is regulation. In most countries supplements are sold as foods, not medicines, so they escape the proof-of-efficacy and proof-of-safety demanded of a licensed drug; batch-to-batch content can vary, and contamination or mislabelling is not rare. That combination — real pharmacological activity plus weak oversight plus a patient who doesn't think it counts — is exactly what makes supplements a blind spot. The practical rule is simple and non-negotiable: always ask about supplements, herbals and over-the-counter products as part of every drug history. If you only record what you prescribed, you are reading half the chart.
Where supplements genuinely help
The honest indication is replacement — you give what is truly missing. Supplements earn their keep when they correct a defined deficiency or meet a defined physiological need — a targeted, evidence-based use, not a hopeful one. The classics every clinician should know: folic acid before and in early pregnancy to prevent neural-tube defects; vitamin D and calcium in documented deficiency and in osteoporosis (with a bisphosphonate); vitamin B12 in pernicious anaemia, malabsorption, or a strict vegan diet; iron in iron-deficiency anaemia; thiamine in the alcohol-dependent or malnourished patient — given before glucose, never after; and the specific AREDS formulation (antioxidant vitamins plus zinc) which slows progression in intermediate age-related macular degeneration. Add the situations of enforced dependence: total parenteral nutrition, major malabsorption, and pregnancy, where deficiency is predictable and replacement is standard care. In all of these the logic is the same — there is a measurable hole, and you are filling it.
Think of a supplement like topping up a car's oil. If the dipstick shows the level is genuinely low, adding oil is exactly right and the engine runs better. But pouring in extra oil on a car that is already full does not make it run faster — it overflows, fouls the engine, and causes problems it never had. Vitamins and minerals behave the same way: replacing a true deficiency restores function, but pouring more into someone already full does not "boost" anything — the surplus either spills out (water-soluble vitamins in the urine) or, worse, accumulates and poisons (the fat-soluble ones and the minerals).
Where they don't — the multivitamin myth
The mass market rests on a belief that more micronutrient is always better and that antioxidant "boosting" prevents disease. Large trials have dismantled it. In well-nourished people, routine multivitamins do not reduce cancer, cardiovascular disease, or overall mortality — the effect on hard outcomes is essentially nil. Worse, several antioxidant megadose experiments did not merely fail; they harmed. Beta-carotene supplements increased lung cancer in smokers (the CARET and ATBC trials — a result so clear the studies were stopped early). High-dose vitamin E has been associated with increased mortality and, in some data, prostate cancer. Even calcium tells a cautionary tale: large supplemental doses (as opposed to dietary calcium) have been linked to kidney stones and, in some analyses, a possible small rise in cardiovascular events. The recurring theme, developed fully in the Vitamins A/D and antioxidant chapters, is that the antioxidant story that looks so tidy in a test tube repeatedly fails — and sometimes reverses — in a whole human being.
The single most useful reframing you can give a patient: for a well-fed body, a supplement can only ever move you toward normal or past it — never above-normal-is-better. Once the tank is full, the extra has nowhere good to go. That is why the evidence for supplements is strongest exactly where there is a deficiency and weakest exactly where there isn't — and why "I take it just in case" is, for most healthy adults, at best money spent on expensive urine and at worst a real risk.
- Ask about supplements, herbals and OTC products in EVERY drug history — patients rarely volunteer them.
- A supplement is a drug: dose, adverse effects and interactions — but usually under-regulated.
- Real benefit = targeted replacement: folate preconception, D/calcium in osteoporosis, B12/iron/thiamine in deficiency, AREDS in intermediate AMD, TPN/pregnancy/malabsorption.
- Routine multivitamins do NOT prevent cancer or cardiovascular disease in well-nourished people.
- Some megadoses HARM: beta-carotene → lung cancer in smokers; high-dose vitamin E; excess calcium supplements.
Vitamin and mineral toxicity
The dividing line is solubility: what you can't excrete, you accumulate. Water-soluble vitamins (B and C) are largely forgiving — excess is filtered into the urine — though even here megadoses are not free (high vitamin C promotes oxalate stones, high B6 can cause a sensory neuropathy). The real danger sits with the fat-soluble vitamins, which accumulate in tissue. Vitamin A in excess is teratogenic (retinoid embryopathy — a key reason for strict contraception on oral retinoids), hepatotoxic, and can raise intracranial pressure. Vitamin D toxicity drives hypercalcaemia — stones, bones, groans and psychiatric moans. These are covered mechanism-first in the Vitamins A/D chapter. Minerals carry their own overdose signatures: acute iron overdose is a leading cause of poisoning death in young children (and a haemochromatosis-like burden with chronic excess); zinc in excess induces copper deficiency (a sideroblastic anaemia and neuropathy); selenium is toxic in a surprisingly narrow margin above requirement (hair and nail loss, garlic breath); and excess iodine can paradoxically cause either hypo- or hyperthyroidism. The unifying idea: for these nutrients the gap between "enough" and "too much" can be narrow, and a determined supplement-taker can cross it.
The high-yield supplement–drug interactions
If you memorise one supplement interaction, make it St John's wort. St John's wort (Hypericum), taken for low mood, is a potent inducer of CYP3A4 and of P-glycoprotein — the same induction mechanism explained in the Principles of Pharmacology chapter (CYP450 induction). By speeding their metabolism it lowers the blood levels, and therefore the effect, of a long list of critical drugs: warfarin (loss of anticoagulation), ciclosporin and tacrolimus (risk of transplant rejection), combined oral contraceptives (breakthrough pregnancy — the opening scene), and several antiretrovirals (loss of viral control). And it carries a second, pharmacodynamic danger: because it also raises serotonin, combining it with an SSRI can precipitate serotonin syndrome. It is the classic, dangerous, exam-favourite supplement interaction for a reason — it manages to be both an inducer and a serotonergic agent. The warfarin angle also belongs to the Hematology section, alongside vitamin K.
The rest cluster into a few memorable patterns. Binding interactions: calcium, iron, magnesium and antacids form insoluble complexes in the gut that block absorption of levothyroxine, bisphosphonates, tetracyclines and fluoroquinolones — the fix is simply to separate the doses by several hours, a point that links to the Endocrine chapter (levothyroxine) and to infection therapy. Vitamin K, whether from supplements or a sudden green-vegetable binge, antagonises warfarin and destabilises the INR — consistency, not avoidance, is the message. Potassium supplements and "low-sodium" salt substitutes (which are potassium chloride) added on top of an ACE inhibitor or ARB can drive dangerous hyperkalaemia, a direct link to the Cardiovascular section. Grapefruit juice earns honorary supplement status as the mirror image of St John's wort: it inhibits intestinal CYP3A4, raising levels of statins, some calcium-channel blockers and others toward toxicity. And a general bleeding caution: ginkgo, high-dose garlic and fish oil all have mild antiplatelet effects that add to aspirin, clopidogrel or an anticoagulant. You do not need to memorise every herb — you need the reflex to check any herbal against the patient's actual drug list.
St John's wort → CYP3A4 / P-gp inducer → ↓ warfarin, ciclosporin/tacrolimus, contraceptives, antiretrovirals; + serotonin syndrome with SSRIs. Calcium / iron / antacids → bind levothyroxine, bisphosphonates, tetracyclines, fluoroquinolones (separate by several hours). Vitamin K (supplements or leafy greens) → antagonises warfarin (keep intake steady). Potassium / salt substitutes + ACE inhibitor or ARB → hyperkalaemia. Grapefruit juice → CYP3A4 inhibitor → ↑ statins, some calcium-channel blockers. Ginkgo / garlic / fish oil → added bleeding risk with antiplatelets/anticoagulants.
- St John's wort is a CYP3A4/P-gp INDUCER → lowers warfarin, ciclosporin, contraceptives, antiretrovirals; + serotonin syndrome with SSRIs.
- Calcium/iron/antacids bind levothyroxine, bisphosphonates, tetracyclines, fluoroquinolones — separate the doses.
- Vitamin K antagonises warfarin; potassium/salt substitutes + ACE inhibitor → hyperkalaemia.
- Grapefruit juice is a CYP3A4 INHIBITOR — the mirror image, raising statin and calcium-channel-blocker levels.
- Fat-soluble vitamins (A, D) accumulate and poison; minerals (iron, zinc, selenium, iodine) have narrow safety margins.
- Take-home: ask, treat real deficiencies, avoid megadoses, and screen for interactions — especially St John's wort.
- Recording only prescribed medicines and never asking about supplements — then being blindsided by an interaction like St John's wort dropping contraceptive or warfarin levels.
- Recommending a multivitamin or antioxidant megadose "to prevent disease" in a well-nourished patient — no benefit, and beta-carotene in smokers or high-dose vitamin E can cause harm.
- Giving levothyroxine, a bisphosphonate or an oral antibiotic at the same time as calcium or iron — the binding blocks absorption and the drug silently fails; separate the doses.
A 38-year-old woman on a stable combined oral contraceptive becomes pregnant. She also takes warfarin for a mechanical valve, and her INR has recently drifted subtherapeutic despite an unchanged dose. On questioning, she started an over-the-counter herbal remedy for low mood a few weeks ago. Which supplement most likely explains both problems?
- Supplements are drugs — with doses, adverse effects and interactions but weak regulation and poor disclosure; always ask about them in the drug history.
- They genuinely help as targeted replacement (folate, D/calcium, B12, iron, thiamine, AREDS, TPN/pregnancy/malabsorption) but routine multivitamins and antioxidant megadoses do not prevent cancer or CVD — and some (beta-carotene, high-dose E, excess calcium) cause harm.
- Fat-soluble vitamins (A teratogenic/hepatotoxic; D hypercalcaemic) accumulate; minerals (iron, zinc→copper deficiency, selenium, iodine) have narrow safety margins.
- Know the interactions — St John's wort (CYP3A4/P-gp inducer + serotonin syndrome) above all, plus calcium/iron binding, vitamin K vs warfarin, potassium + ACE inhibitor, and grapefruit juice.
- Rang & Dale's Pharmacology — Vitamins, minerals and dietary supplements; drug interactions.
- Katzung. Basic & Clinical Pharmacology — Dietary supplements & herbal medications; vitamins.
- British National Formulary (BNF) — St John's wort interactions; potassium, calcium and iron interactions.
- The Omenn GS et al. (CARET) and ATBC Study Group trials — beta-carotene and lung cancer in smokers. New England Journal of Medicine.
- US Preventive Services Task Force — Vitamin, Mineral, and Multivitamin Supplementation to Prevent Cardiovascular Disease and Cancer.
- NICE guidance — Antenatal care (folic acid); Osteoporosis (calcium and vitamin D); Alcohol-use disorders (thiamine, Wernicke's prophylaxis).

