Vitamin D: What the Biggest Trial Actually Found

Vitamin D occupies an unusual position: it is the supplement most likely to be recommended by a doctor, and it has one of the largest, cleanest randomised trials of any supplement ever tested. Those two facts sit less comfortably together than most people realise.

The trial

VITAL was designed to answer the question directly rather than by inference from observational data — which is important, because the observational data on vitamin D look considerably more encouraging than the trial data, for reasons covered below.

It enrolled 25,871 adults in the United States: men aged 50 and over, women aged 55 and over, all free of cardiovascular disease and cancer at baseline. Participants received 2,000 IU of vitamin D₃ daily or matching placebo, and were followed for a median of 5.3 years.

That is a large trial, at a meaningful dose, for a meaningful duration, in a general population. It is close to the best test the question is ever likely to get.

What it found

On the two primary endpoints, nothing.

• Invasive cancer of any type: no significant difference between vitamin D and placebo.
• Major cardiovascular events — heart attack, stroke, or cardiovascular death: no significant difference.

The investigators’ own summary was that daily high-dose vitamin D for five years among initially healthy US adults did not reduce the incidence of cancer or major cardiovascular events.

There were secondary signals that generated headlines: reduced cancer risk among participants with a normal BMI, and reduced cancer deaths when the first two years of follow-up were excluded. These are worth knowing about and worth treating carefully. Secondary and post-hoc analyses of a trial that missed its primary endpoints are hypothesis-generating. They are the reason to run another trial, not the reason to start taking something.

Why testing is graded “we do not know”

Alongside the treatment question sits a testing question, and the Task Force has looked at it.

Screening asymptomatic adults for vitamin D deficiency carries a Grade I: the current evidence is insufficient to assess the balance of benefits and harms. That grade means genuinely undetermined — not “it helps,” and not “it is useless.”

The reasons are more interesting than the grade. There is no agreed threshold defining deficiency; different laboratories and bodies use different cut-offs, so the same blood sample can be normal in one system and deficient in another. And no trial has shown that finding and treating low levels in people without symptoms improves any outcome that matters.

Which is why vitamin D testing appears on the list of things deliberately left off the men’s screening schedule, alongside routine testosterone testing and whole-body MRI.

Deficiency is a real thing

None of the above means vitamin D does not matter. It means supplementing healthy people did not produce the benefits the market promises.

Genuine, severe deficiency causes real disease — osteomalacia in adults, rickets in children — and it is properly treated. Certain groups are at meaningfully higher risk: people with very little sun exposure, darker skin at high latitudes, malabsorption conditions, some medications, older adults in residential care, and those who cover most of their skin outdoors.

If you are in one of those groups, or you have bone pain or proximal muscle weakness, that is a clinical question with a real answer. The evidence above is about supplementing the general population, not about treating people who are actually deficient.

The wider lesson

Vitamin D is the clearest available illustration of a pattern that recurs across this site.

Observational studies consistently show that people with low vitamin D have worse health outcomes across an enormous range of conditions. That finding is real. The inference — that raising vitamin D will improve those outcomes — is the part that keeps failing when tested.

The most likely explanation is that low vitamin D is substantially a marker rather than a cause. People who are chronically ill, obese, sedentary or housebound have lower levels because of how they live and what is wrong with them. The vitamin is reporting the problem, not creating it.

That is exactly the structure of the alcohol and mortality puzzle, and of the observational finding that moderate drinkers have fewer erection problems. Whenever an association looks too good and too broad, “the marker is not the mechanism” is the first hypothesis to test.

It is also why supplement categories keep producing the same disappointing arc: a strong observational signal, a compelling story, and a null result when someone finally randomises it.

What to do

If you have a specific risk factor for deficiency, or symptoms, ask a doctor — that is a legitimate test with a legitimate treatment.

If you are a generally healthy man taking vitamin D because it seemed sensible, the largest trial of that exact proposition found no reduction in cancer or cardiovascular events. It is inexpensive and safe at ordinary doses, so this is not a warning. It is a recalibration of what you should expect from it, which is: not much.

And if the underlying goal was to do something meaningful for your long-term health, the interventions with actual evidence are dull and free. Blood pressure is the highest-value number you can act on. Sixty minutes of resistance training a week is associated with a 26% lower all-cause mortality. Neither comes in a bottle.

This article is for information only and is not medical advice. It cannot account for your individual circumstances. Talk to a doctor about your own situation, particularly before starting or stopping any supplement or prescription medication.