Supplements for Women: What Changes by Life Stage

Most supplement advice aimed at women is a generic list. The useful version is organised by life stage, because the answers genuinely change — and one of them reverses completely.

The two worth testing before touching

Iron

Menstruating women lose iron monthly, and the prevalence is higher than most people assume. An analysis of national survey data covering US females aged 12 to 21 found 38.6% had iron deficiency at a ferritin threshold of 25 micrograms per litre. Iron-deficiency anaemia was only 6.3% — and here is the number that matters most: among those who were iron deficient, 83.6% were not anaemic.

Which tells you why the usual test misses it. Haemoglobin is normal in the large majority of iron-deficient women. Ferritin is the test that detects depleted stores, and it is the one to ask for.

Two honest complications. Ferritin is an acute-phase reactant, so it rises with inflammation, infection, obesity and liver disease, which can mask a real deficiency. And the cutoffs are contested — in that same population, moving the threshold from 15 to 25 to 50 micrograms per litre moves prevalence from 17% to 38.6% to 77.5%. The number your lab flags as low may not be the number a haematologist would use.

Iron is also the supplement most often taken unnecessarily. The RDA is 18mg a day for women 19 to 50, falling to 8mg from 51, with a tolerable upper limit of 45mg. High doses cause gastric upset, constipation, nausea and cramping — which in practice is the most common harm, because it drives people who genuinely need iron to stop taking it. Roughly one in ten white Americans carries a haemochromatosis mutation, for whom supplemental iron is contraindicated. And iron is a leading cause of poisoning deaths in young children, which matters if there are children in the house.

So: test, then supplement, then retest. If you do supplement, take it with vitamin C and away from tea, coffee, calcium and dairy, all of which inhibit absorption. Alternate-day dosing is often better absorbed and better tolerated than daily.

Vitamin D

Easily measured and cheaply corrected — but the “epidemic” framing is overstated. In national survey data, about 18% of women were in the inadequate range and around 5% were at risk of deficiency. The RDA is 600 IU to age 70 and 800 IU thereafter, with an upper limit of 4,000 IU. Test rather than guess at a dose, and see our fuller piece on what the big vitamin D trials actually found.

Childbearing years

Folate

This is the one genuinely non-negotiable item in the article, and the evidence behind it is the strongest here by some distance. The landmark 1991 trial randomised women at high risk of a recurrence and found neural tube defects reduced by around 72%. Notably, the other vitamins tested showed no significant effect — it was folic acid specifically, not prenatal vitamins generically.

The recommendation is 400 to 800 micrograms daily for anyone planning or capable of pregnancy, starting at least a month before conception and continuing through the first two to three months. The timing is the entire point and the most under-communicated fact in the field: the neural tube closes by about day 28, usually before a pregnancy is recognised. Started at a positive test, it is largely too late for this particular benefit. Roughly half of pregnancies are unplanned, which is the actual reason the advice is framed around anyone who could become pregnant.

A correction on MTHFR. You will be told that if you carry an MTHFR variant you cannot process folic acid and need methylfolate instead. The clinical genetics guidance is that MTHFR polymorphism testing has minimal clinical utility and should not be part of a routine work-up. More decisively, the CDC recommends 400 micrograms a day of folic acid, not 5-MTHF, for people who could become pregnant — even with the 677C>T variant. There is a further practical wrinkle: conversion factors between micrograms and dietary folate equivalents have not been formally established for 5-MTHF, so a methylfolate product’s dose is not directly comparable to the studied folic acid dose. Direct-to-consumer MTHFR testing is a marketing engine, not a clinical one.

Iodine

The RDA is 150 micrograms for adult women, 220 in pregnancy and 290 while breastfeeding. National survey data put median urinary iodine in pregnant US women below the WHO adequacy range, and the American Thyroid Association recommends 150 micrograms daily as potassium iodide for women planning pregnancy, pregnant or lactating.

The gap worth knowing about: in a survey of 59 best-selling US prenatal multivitamins, only 58% contained iodine at all, with declared amounts ranging from 25 to 290 micrograms. “I take a prenatal” does not guarantee you are getting any. Intake has also drifted down as people move to non-iodised specialty salts and as plant milks — generally not fortified, unlike dairy — replace dairy.

More is emphatically not better here. Excess iodine causes real harm, and kelp supplements are notoriously variable in content and can overshoot badly.

Choline

Important in pregnancy and under-supplied by most prenatal formulas. Eggs are the best food source by a wide margin.

A note on pregnancy generally

This is the one life stage where self-directed supplementation is genuinely inadvisable. Doses interact with prenatal care, screening results and individual risk. Take the questions to your obstetric provider rather than to a label.

Perimenopause and after

This is where the advice reverses. Once menstruation stops, iron requirements fall sharply — to the same level as men. Continuing an iron supplement out of habit is a genuine mistake, and unexplained iron deficiency in a postmenopausal woman warrants investigation rather than supplementation.

Calcium and vitamin D for bone: what the trials found

This is the section most likely to surprise you. The Women’s Health Initiative randomised 36,282 postmenopausal women to 1,000mg of calcium plus 400 IU of vitamin D or placebo for seven years and found no significant reduction in hip fracture. Total hip bone density was modestly preserved, but that did not translate into fewer fractures — a density-versus-fracture disconnect worth remembering whenever a DXA number is quoted at you. Kidney stone risk went up.

A subgroup of women who took at least 80% of their doses did show around a 29% reduction in hip fracture. That is a post-hoc adherence analysis, which is exactly the kind of finding supplement marketing over-quotes; people who take their pills reliably differ from those who do not in many other ways.

The VITAL trial went higher — 2,000 IU of vitamin D in nearly 26,000 participants — and reported that supplementation did not lower the risk of total, hip or non-vertebral fractures. The US Preventive Services Task Force recommends against daily supplementation at the lower doses for primary fracture prevention in community-dwelling postmenopausal women, and its more recent draft broadens that.

None of this means calcium and vitamin D are unimportant to bone. It means that supplementing people who are already replete does not reduce fractures. The nutrients are essential; the pill is not automatically beneficial. Food first for calcium, supplement the gap, and do not chase megadoses.

Protein

The most under-recognised item on this list. Falling oestrogen accelerates loss of both bone and muscle, and older muscle shows anabolic resistance — it responds less to the same amino acid stimulus.

The most-cited recommendation, from the PROT-AGE group, is 1.0 to 1.2 grams per kilogram per day against an RDA of 0.8, with 25 to 30 grams per meal and resistance training two to three times a week. Two caveats belong with it. The sponsoring society received funding from a major nutrition company and several authors disclosed industry relationships — that does not make the numbers wrong, but the ties are almost never mentioned in the marketing built on them. And a 2024 review looking specifically at post-menopausal women concluded that the evidence is limited and not of high quality, making firm inferences difficult.

The defensible position: more protein than the RDA is a reasonable, low-risk bet in this population, especially alongside resistance training. It is not a proven, quantified menopause-specific intervention, and without the training the effect is modest at best. Our piece on how much protein you actually need has the general numbers.

Creatine: what it does and does not do

Creatine is the current darling of menopause marketing, and the specific claim being made is the one that failed.

A 2026 meta-analysis of seven randomised placebo-controlled trials in 608 postmenopausal women found a lean mass gain of 0.37kg and a leg-press strength gain of 7.5kg versus placebo. On bone, it found density unchanged overall, with the authors stating the effect on bone density remains unclear.

The best single trial says the same thing more definitively. A two-year study in 237 postmenopausal women, with resistance training three times a week plus walking, reported that creatine supplementation had no effect on bone mineral density at the femoral neck, total hip or lumbar spine. It did preserve a bending-strength measure at the femoral neck where placebo declined — a legitimate and interesting secondary finding, but a surrogate measure, and not what “creatine builds bone” implies.

The practical finding most likely to be left out of an advert: benefits appeared only when creatine at 5 grams a day or more was combined with resistance training. Lower doses without training showed no measurable effect. The training is doing most of the work.

One transparency note: most of the creatine-for-women literature comes from a small and industry-adjacent research network. That does not make the analysis wrong — its bone conclusion is negative, which cuts against interest — but it is worth knowing. More detail in our piece on what the creatine evidence actually supports.

Magnesium

The RDA is 310 to 320mg a day, and the upper limit of 350mg applies to supplemental magnesium only, not to food — a distinction people get wrong. Around 48% of Americans consume less magnesium from food and drink than the estimated average requirement, which is the real basis for its popularity.

Two things worth correcting. First, you cannot usefully test it: the NIH is explicit that serum levels do not accurately reflect total body magnesium or tissue concentrations, and that no single method is considered satisfactory. Anyone selling you a serum magnesium as a status marker is selling a test that does not do the job. Look at your diet instead.

Second, the two things magnesium is most heavily marketed for in this demographic — sleep and anxiety — are not addressed at all in the NIH professional fact sheet. That absence is itself informative. The claim with genuine backing is migraine prevention, which neurology and headache societies rate as probably effective, though the NIH calls the underlying evidence limited.

If you do supplement, note that magnesium oxide is the cheapest and most common form in mass-market products and among the worst absorbed, and that carbonate, chloride, gluconate and oxide are the forms most likely to cause diarrhoea and cramping.

Hot flushes and the botanical aisle

The Menopause Society’s 2023 position statement on non-hormonal therapy is unusually blunt, and it is the fairest summary available. It does not recommend black cohosh, evening primrose, maca, dong quai, vitamin E, omega-3, pollen extract, rhubarb, ginseng, chasteberry or milk thistle for hot flushes, citing insufficient or absent evidence in each case. Soy foods, soy extracts and S-equol are described as mixed evidence.

To show the standard is being applied evenly rather than only to supplements: the same statement also declines to recommend paced respiration, cooling techniques, trigger avoidance, exercise, yoga, mindfulness, acupuncture and dietary modification for this symptom. What it does recommend is mostly not supplements — cognitive behavioural therapy and clinical hypnosis at the highest evidence level, alongside several prescription options including a newer non-hormonal drug.

Two safety notes rather than efficacy ones. Black cohosh carries rare reports of liver injury; the guidance is to stop it and seek advice if abdominal pain, dark urine or jaundice appear. And phytoestrogens such as red clover and concentrated soy extracts have unestablished long-term safety and may not be appropriate for women advised to avoid oestrogen — a genuine conversation for a doctor, not a label.

Collagen

Collagen deserves its own note because it is the cleanest illustration of funding bias in this whole field.

A 2025 systematic review and meta-analysis of 23 randomised trials in 1,474 participants concluded there is currently no clinical evidence to support collagen supplements for preventing or treating skin ageing. The mechanism of the illusion is the interesting part: the pooled analysis initially showed improvements in hydration, elasticity and wrinkles — but trials without pharmaceutical or supplement company backing showed no benefit, while industry-funded studies reported positive effects. Restricting to higher-quality studies, no significant effect remained in any category.

Industry bodies have disputed the methodology, alleging misclassified funding sources and reporting errors, and that pushback deserves to be on the record. But the funding-source gradient is the finding that would have to be explained away, and misclassification claims are testable.

Separately, and more practically for anyone trying to hit a protein target: collagen peptides are digested to amino acids like any other protein, and collagen is an incomplete protein — it contains no tryptophan. As a protein source it is a worse choice than whey, dairy or soy, not a better one. We have written that comparison up in collagen versus whey.

Bone: what works versus what is sold

What is sold: calcium megadoses, vitamin D megadoses, collagen, and “bone support” blends of boron, silica, vitamin K2 and strontium.

What the evidence supports, in order: progressive resistance and impact-loading exercise, which has the most consistent evidence for bone density and, more importantly, for falls — and which is free. Then adequate rather than maximal calcium and vitamin D, preferentially from food. Then protein. Then, for women with established low bone density, a prescription conversation, because the most consequential thing a woman with a low DXA score can do is not a supplement.

Two specific warnings on that blend list. Vitamin K2’s evidence is mixed and dominated by Japanese trials using a different compound at pharmacological doses than the MK-7 in Western supplements, and it has a clinically important warfarin interaction. And strontium is worth avoiding: strontium ranelate had cardiovascular safety problems leading to restriction in Europe, the citrate salt sold as a supplement has no comparable evidence, and — this is the striking part — strontium is heavier than calcium, so it artefactually inflates DXA readings. You can buy a better bone density score that does not reflect more bone.

Older women: the B12 case

This is one of the few places where a supplement recommendation is on genuinely solid ground, and where the form matters for a real physiological reason.

Atrophic gastritis affects an estimated 10 to 30% of people over 60. Reduced stomach acid means B12 is never released from the food protein it is bound to — but the free, crystalline form found in fortified foods and supplements is still fully absorbed. That mechanism is why the Food and Nutrition Board recommends adults over 50 get most of the 2.4 microgram RDA from fortified food or supplements rather than from food-bound B12 alone.

Add the drug interactions, which affect an enormous number of women: long-term proton pump inhibitors and metformin both impair B12 status.

And a callback to the folate section: high folic acid intake can correct the anaemia of B12 deficiency but not the neurological damage — masking a progressing problem. That is why the folate upper limit exists, and why “more B vitamins” is not a free action in older adults.

Across all stages

Omega-3. Worth it if you do not eat oily fish twice a week. Read the EPA and DHA figures rather than the front of the bottle — our guide to the numbers on the back of the bottle explains why.

Protein distribution. Many women eat protein in a heavily skewed pattern — almost none at breakfast, most at dinner — which fails the per-meal threshold even when the daily total looks fine. Redistributing it costs nothing and probably matters more than any product on this page.

What to skip

Detox teas and cleanses. “Hormone balancing” blends, which have no defined mechanism. Iron in a general multivitamin taken after menopause, which is a common and avoidable error — check the label of whatever you are already taking. Direct-to-consumer MTHFR and CYP1A2 gene tests. And serum magnesium, which does not measure what it is sold as measuring.