Vitamin D: what the evidence does (and doesn't) say about supplementing

Few things generate as much apparent consensus as vitamin D: almost everyone knows someone who takes it, and almost nobody is quite sure what for. Public conversation swings between two equally unhelpful extremes: the headline that half the country is deficient, and the sneer that it does nothing at all. The reality, once you read the large clinical trials, is both duller and more interesting: the levels measured in Spain really are low, and at the same time supplementing healthy people in the general population has not been shown to prevent the illnesses it is credited with preventing. This article explains what has been measured, what has been tested in trials, and where the limits of what can be claimed today actually lie.

What is actually being measured when they test your vitamin D

The value that appears on your blood test is 25-hydroxyvitamin D, abbreviated to 25(OH)D, and it is usually expressed in nanograms per millilitre. It is often assessed alongside intact parathyroid hormone (PTH), because the two are linked in the regulation of bone and mineral metabolism. In the Spanish population cohort published in the European Journal of Clinical Nutrition in 2011, with 1,262 subjects aged between 20 and 83, the median 25(OH)D was 22.46 ng/mL and the median intact PTH was 42.29 pg/mL.

That figure needs to be read carefully, because this is precisely where the misunderstandings begin. A median value is not a collective diagnosis: it tells you where the centre of a distribution sits, not how many people have a clinical problem. What counts as insufficiency depends on the threshold you choose, and that threshold has been debated for years. The figure doing the rounds on social media about the percentage of Spaniards who are deficient does not come from this data, and we are not going to repeat it here.

The Spanish picture: low levels even where there is plenty of sun

The intuitive idea that a sunny country should not have a vitamin D problem does not square with what has been measured. A cross-sectional study carried out in osteoporosis units across Spain, in late spring, found that 76% of postmenopausal women with osteoporosis who had not yet been treated had 25(OH)D values below 30 ng/mL. The review that collects this data, published in the Revista de Osteoporosis y Metabolismo Mineral in 2014, concludes that the high prevalence of insufficiency in Spanish postmenopausal women and older people is independent of geographical area and of the cut-off point used.

Two caveats matter. First: this is a selected population, women attending osteoporosis units, not the general population. Second: the fact that the measurements were taken in late spring undermines seasonality as the sole explanation, because that is the time of year when you would expect the highest readings. In other words, the problem of low levels in Spain is real and documented in specific groups. What to do with that information is another matter entirely, and that is where the clinical trials come in.

What the large randomised trials found

Cancer and cardiovascular disease: VITAL

VITAL is the reference trial on this question: randomised, placebo-controlled, with a 2×2 factorial design and 2,000 IU daily of vitamin D3. Its main results, published in the New England Journal of Medicine in 2019, were clear: supplementation did not reduce the incidence of invasive cancer or of cardiovascular events compared with placebo.

There is one nuance that gets quoted a great deal and is worth placing properly. Secondary analyses and updated meta-analyses of VITAL found a reduction in cancer mortality when the earliest follow-up period was excluded, but not in incidence. That is an exploratory result, not a confirmatory one: it comes from looking at the data in a way that was not planned in the original design, and it generates hypotheses rather than closing them. Presenting it as proof that vitamin D prevents cancer deaths would be overinterpreting it.

Respiratory infections: an effect that keeps shrinking

This is the most instructive story of the lot, because it shows how evidence evolves. The meta-analysis in the Martineau and Jolliffe series, covering 43 randomised trials and published in 2021, found a statistically significant but very small protective effect against acute respiratory infections: OR 0.92 (95% CI 0.86–0.99). The update published in Lancet Diabetes & Endocrinology in 2025 extends the search to April 2024 and adds six further trials, and with them the effect is no longer statistically significant: OR 0.94 (95% CI 0.88–1.00).

Meanwhile, a meta-analysis of 30 randomised trials with 30,263 participants, published in Nutrients in 2022, found no significant overall effect: RR 0.96 (95% CI 0.91–1.01), a confidence interval that crosses 1 and is therefore compatible with no benefit at all. In the subgroup on daily dosing an effect did emerge: RR 0.83 (95% CI 0.73–0.95). Subgroup analyses are signals rather than firm conclusions: in this same meta-analysis the effect disappeared when only the highest-quality studies were kept, and the 2025 update found no differences by dosing frequency.

Fractures and falls: the USPSTF position

The US Preventive Services Task Force, in its December 2024 draft recommendation, concludes with moderate certainty that vitamin D supplementation, with or without calcium, provides no net benefit for the primary prevention of fractures in postmenopausal women and community-dwelling men aged 60 and over, nor for preventing falls. In a separate recommendation, it does recommend exercise to prevent falls in people over 65 at increased risk of falling.

The limits of that recommendation matter as much as its content: it does not apply to people with osteoporosis, with a previous osteoporotic fracture, with conditions that cause vitamin D deficiency or malabsorption, or with a diagnosed deficiency. We are talking about primary prevention in a healthy population, which is a completely different clinical scenario from someone with established bone disease.

What the 2024 Endocrine Society guideline says

In August 2024, the Endocrine Society published a clinical practice guideline on vitamin D for disease prevention, based on clinical trials and aimed at people with no established indication for treatment or for 25(OH)D testing. Its main conclusion is twofold and somewhat counterintuitive: healthy adults under 75 are unlikely to benefit from taking more vitamin D than the IOM's recommended daily intake, and they do not need their 25(OH)D level measured in a blood test either.

The guideline does suggest an intake above the recommended level in four specific groups: children and adolescents, pregnant women, adults aged 75 and over, and adults with high-risk prediabetes. That last group catches a lot of people's attention; if you want to understand what that intermediate state of glucose metabolism is and how it is managed, we cover it in the guide to prediabetes. It is worth remembering that suggesting a higher intake for a group is not the same as prescribing treatment for an individual: that is a matter for a doctor who knows your medical history.

More is not better: the AEMPS warning

Here is the part that almost never appears in content about vitamin D, and it is the most relevant one in the Spanish context. The Spanish Agency for Medicines and Medical Devices (AEMPS) warned in its safety notice MUH (FV) 2/2019 that serious cases of hypercalcaemia due to overdosing of medicines containing vitamin D as the sole active ingredient continue to be reported to the Spanish Pharmacovigilance System. The pattern described is specific: cholecalciferol in children and calcifediol in adults.

The same notice describes serious cases of hypercalcaemia in newborns and infants treated with cholecalciferol at daily doses far above those recommended to prevent deficiency, in some cases using presentations intended for adults. The cases are linked to errors in prescribing, dispensing or administration. This is why vitamin D is not a harmless supplement by definition: it is a fat-soluble vitamin with scope to accumulate, and the dose matters.

The myths the evidence does not support

Three widely repeated claims have no backing in the data we have reviewed. The first: that vitamin D helps you lose weight. There is nothing in this evidence to support that, and clinically grounded weight loss still relies on other routes, as we explain in the guide to what GLP-1s are and how they work. The second: that it is a treatment for depression. The third: that it prevents COVID. None of the three can be asserted.

There is a fourth, subtler myth: the headline figure for the exact percentage of Spaniards who are deficient. It circulates widely and does not come from the sources we have cited. The honest thing to do is to state what has actually been measured: a population median of 22.46 ng/mL in a cohort of 1,262 people, and 76% below 30 ng/mL in a selected population of postmenopausal women with osteoporosis. Nothing more, and nothing more is needed to take the subject seriously.

How all this fits into metabolic health

Vitamin D often turns up in blood tests ordered for other reasons, and that is where it makes sense to place it: as one more piece of a complete metabolic picture, not as an isolated marker you chase on its own. If you are looking at your blood tests because you are worried about the bigger picture — glucose, blood pressure, waist circumference, lipids — there is far more to be gained from understanding the whole picture, something we cover in the guide to metabolic syndrome and in the one on metabolic fatty liver disease.

The pattern that keeps repeating in the research of the past decade is this: isolated supplements deliver far less than expected in healthy general populations, while interventions aimed at the underlying determinants — physical activity, body composition, sleep, medical treatment where indicated — remain the ones that move the needle. The USPSTF makes the same point when it notes that exercise does prevent falls where the supplement does not.

In summary

Vitamin D is a good example of why a low laboratory value does not automatically translate into a treatment that works: the levels measured in Spain are low in certain groups, and even so the large trials have not shown that supplementing healthy people in the general population prevents cancer, cardiovascular events or fractures. Metabolic health is decided elsewhere. If what you are after is tackling your weight with clinical rigour, what exists in Spain today is medical weight-loss treatment, with prior assessment, a prescription where appropriate from a registered doctor and ongoing follow-up, which is what we do at AiON. You can start by taking a few minutes to check whether you are eligible.