This article is for general information and is not medical advice. Vitamin D supplementation is genuinely necessary for some people. Talk to your doctor before starting or stopping any supplement, and never stop a prescribed treatment based on something you read online.
Somewhere between a quarter and a half of American adults take vitamin D, depending on which survey you believe. The logic seems airtight: blood tests show widespread deficiency, deficiency is linked to cancer, heart disease, fractures, depression and diabetes, and the correction costs about six cents a day.
Then somebody actually ran the experiment. Twice. At enormous scale. And the results were close to nothing.
VITAL: 25,871 people, five years, no effect
The VITAL trial, led by JoAnn Manson at Brigham and Women’s Hospital and published in the New England Journal of Medicine in 2019, randomised 25,871 adults to 2,000 IU of vitamin D3 daily or placebo, and followed them for a median of 5.3 years. It was designed to settle the two biggest claims at once: cancer and cardiovascular disease.
Invasive cancer of any type: 793 cases in the vitamin D group, 824 in the placebo group. Hazard ratio 0.96, confidence interval crossing 1. Major cardiovascular events: 396 versus 409. Hazard ratio 0.97. Neither result was statistically significant, and neither was close.
The trial was not underpowered, not short, and not badly run. It was the study the field had been asking for, and it came back null.
D-Health: 21,315 people, a bigger dose, still nothing
Australia’s D-Health trial, published in The Lancet Diabetes & Endocrinology in 2022, took a different approach: 21,315 older adults, a much larger monthly dose of 60,000 IU, and all-cause mortality as the primary outcome. Median follow-up was 5.7 years.
Deaths: 1,100 in the vitamin D group, 1,090 in the placebo group. The hazard ratio was 1.04. If anything, the point estimate pointed the wrong way, though the confidence interval comfortably included no effect.
Two trials, on two continents, with nearly 50,000 participants between them, using two different dosing strategies. Both null.
The threshold that created the epidemic
Here is the part that reframes everything. The reason so many people were told they were deficient is a number: 30 nanograms per millilitre. Below that, you were “insufficient.” The Endocrine Society’s 2011 clinical practice guideline is where that threshold entered routine medicine.
In 2024, the Endocrine Society published a new guideline — and withdrew it.
The 2024 guideline does not recommend routine screening of 25-hydroxyvitamin D levels in healthy adults, and does not endorse a specific target blood level for the general population. The society that popularised the threshold no longer supports using it as a population-wide diagnostic cut-off. The reasoning is that the number was derived from indirect physiological markers, not from trials showing that correcting to 30 ng/mL improves outcomes — and when those trials were eventually run, they didn’t.
The Institute of Medicine had actually flagged this in 2011, setting a sufficiency threshold of 20 ng/mL and warning explicitly that the higher number risked labelling healthy people as sick. For thirteen years, two major bodies published incompatible numbers and most clinical labs printed the higher one on the results slip.
So why do observational studies keep finding associations?
They do, consistently, and that’s what makes this interesting rather than simply embarrassing. People with low vitamin D really do get more cancer, more heart disease and more infections. The associations are large and they replicate.
The likeliest explanation is reverse causation and confounding. Vitamin D is made in skin exposed to sunlight. The people with the highest levels are the people who go outside — which means they are, on average, more mobile, less obese, less chronically ill, younger in function if not in years, and not housebound. Low vitamin D may be a very good marker of poor health without being a cause of it. It’s a thermometer, not a furnace.
Inflammation compounds this. Systemic inflammation lowers circulating 25-hydroxyvitamin D. Many of the diseases associated with low vitamin D are inflammatory. The arrow may run from disease to low vitamin D rather than the other way around.
What the trials do support
This is not an argument that vitamin D does nothing. It is an argument about who and how much.
Vitamin D unambiguously prevents and treats rickets and osteomalacia. It matters for people with genuine deficiency — those with minimal sun exposure, darker skin at high latitudes, malabsorption conditions, certain medications, or who are institutionalised. Supplementation in combination with calcium has shown fracture reduction in older adults in residential care, the population where deficiency is both common and severe.
VITAL itself found a signal worth watching: in secondary analyses, autoimmune disease incidence was lower in the supplement group. That’s a secondary outcome in a trial designed for something else, which means it’s a hypothesis, not a finding. It deserves its own trial.
What the evidence does not support is the thing most people are doing: a healthy adult with no risk factors, no symptoms, and a blood level in the twenties, taking a daily supplement to prevent cancer or heart disease.
The more useful question
The vitamin D story is a near-perfect case study in how a plausible mechanism, a cheap intervention and a widely available blood test can produce a public health movement that outruns its evidence. Nothing here involved fraud or even bad faith. The mechanism was real. The associations were real. The test was accurate. The inference was wrong.
The trials that finally checked cost hundreds of millions of dollars and took a decade, which is exactly why they’re rare, and exactly why the conclusion they reached hasn’t travelled as far as the original claim did.
If you have been told you’re deficient, that may well be accurate and worth treating. If you started taking vitamin D because you read that it prevents cancer, the two largest trials ever run on that question disagree with you — and so, as of 2024, does the organisation that set the threshold.
Sources
- Manson JE et al. Vitamin D Supplements and Prevention of Cancer and Cardiovascular Disease. New England Journal of Medicine 2019;380:33-44.
- Neale RE et al. The D-Health Trial: a randomised controlled trial of the effect of vitamin D on mortality. The Lancet Diabetes & Endocrinology 2022;10(2):120-128.
- Demay MB et al. Vitamin D for the Prevention of Disease: An Endocrine Society Clinical Practice Guideline. Journal of Clinical Endocrinology & Metabolism 2024;109(8):1907-1947.
- Institute of Medicine. Dietary Reference Intakes for Calcium and Vitamin D. National Academies Press, 2011.
