Vitamin D Isn't Bad for You. Megadosing Is.

Vitamin D toxicity is a real, documented condition, and nearly every case traces back to a supplement rather than to sunlight or food.^[1] One case report describes two teenagers hospitalized after a creatine supplement was found to contain 425,000 IU of vitamin D per serving, about 100 times the tolerable daily dose.^[1] That is the actual shape of the risk: a manufacturing error or a megadose habit, far removed from a multivitamin taken as directed.

The bigger problem for most adults runs the other way. Low vitamin D is far more common than toxic vitamin D, and that gap is the real story this article is here to settle.

The dose that matters: 4,000 IU a day, and even that isn’t risk-free

Regulators didn’t pick the upper limit at random. A meta-analysis of 22 randomized trials, covering 12,952 participants, states plainly that the upper tolerable intake level for vitamin D in the general population has been set at 4,000 IU a day.^[2]

Here’s the part most explainers skip: that ceiling isn’t a hard line between safe and dangerous. The same meta-analysis found that even daily doses of 3,200 to 4,000 IU, taken for six months or longer, modestly raised the risk of:

• Hypercalcemia, with about 4 extra cases per 1,000 people^[2]
• Falls
• Hospitalization^[2]

Bottom line: the “safe” upper limit is the point where risk becomes measurable in trials. Risk itself can start well before that line.

Meanwhile, low vitamin D is the far more common problem. A pooled analysis of 308 studies and nearly 7.95 million participants across 81 countries found that about 48% of people worldwide have blood vitamin D levels below 50 nmol/L, a commonly used deficiency threshold.^[3] Toxicity cases, by contrast, are rare enough to show up mainly as individual case reports rather than population statistics.

What the strongest trials and cohorts found

The best randomized evidence on vitamin D supplementation doesn’t come from a case report. It comes from VITAL, a placebo-controlled trial of nearly 26,000 adults that gave half the group 2,000 IU of vitamin D3 a day for a median of about five years. It found no significant reduction in invasive cancer or major cardiovascular events.^[4]

That null result sits next to a very different kind of evidence: observational cohorts that keep finding low vitamin D alongside worse cardiovascular outcomes.

• A cohort study of 1,828 adults, followed for 54 months, found vitamin D deficiency independently predicted major adverse cardiovascular events. Chronic inflammation also accounted for more of the link between body fat and those events specifically among people with low vitamin D status.^[5]
• A meta-analysis of 25 prospective cohort studies linked lower circulating vitamin D to a 44% higher combined risk of cardiovascular disease and death, and a 54% higher risk of cardiovascular death specifically. The incidence-only estimate on its own wasn’t statistically significant.^[6]

44% higher combined cardiovascular risk linked to low vitamin D across 25 cohort studies PMID 31699030

Two different kinds of evidence, two different answers. That gap is the whole debate.

Why the cohort link isn’t proof that vitamin D causes heart disease

A randomized trial and an observational cohort answer different questions. VITAL tested whether adding vitamin D changes cardiovascular risk, in people who weren’t selected for deficiency, and found that raising vitamin D levels with a daily supplement did not reproduce the lower cardiovascular risk linked to higher vitamin D levels in cohort studies.^[4]

There are two honest explanations for why the cohort link exists without the trial confirming it causes anything:

• Reverse causation through body fat. Researchers have hypothesized that the fat solubility of vitamin D causes it to be sequestered in adipose tissue, one proposed explanation, still under investigation, for why obesity correlates with lower circulating vitamin D.^[7] Low vitamin D, in other words, might be a marker of carrying more body fat rather than a cause of anything downstream of it.
• Weak genetic evidence for causation. A 2026 review triangulating genetic, observational and trial evidence found that vitamin D’s associations with cardiovascular disease and metabolic disorders are not consistently supported by Mendelian randomization or randomized trial evidence, the kind of study design that approximates a randomized trial using inherited genetic variation. The same review found vitamin D’s role in skeletal health holds up fine under that stricter test. Heart disease and metabolic disorders don’t.^[8]

None of that means deficiency is harmless. It means a cohort study can show association, while only a randomized trial tests causation, and this is the clearest case in the whole vitamin D literature where the two disagree.

How megadosing causes harm

Vitamin D’s day job is managing calcium. Push too much of it into the body for too long, and the mechanism that makes it useful is the same one that makes it dangerous.

Vitamin D toxicity produces hypercalcemia alongside suppressed parathyroid hormone and very high circulating 25-hydroxyvitamin D.^[9] Because vitamin D is stored in fat tissue rather than cleared quickly, its effects can persist for months after supplementation stops, even once the source is removed.^[9] That slow clearance is what makes megadosing different from, say, overdoing a water-soluble vitamin like C, which the kidneys just flush out. Vitamin D shares its fat-soluble, slow-clearance toxicity profile with vitamin A, another nutrient people rarely overdose on by accident, and rarely overdose on gently when they do.

Sustained hypercalcemia has real downstream consequences. Case reports document vitamin D toxicity causing hypercalcemia severe enough to impair kidney function, with one patient’s acute kidney injury and elevated vitamin D levels taking roughly 14 months to resolve after the vitamin D was stopped.^[10]

Sunlight doesn’t carry this risk, and the reason is mechanical rather than magical. Once ultraviolet light converts a compound in skin into vitamin D3, continued sun exposure photodegrades that vitamin D3 into other compounds. It’s a self-limiting process, and it’s why sunlight has never been documented to cause the toxicity seen with supplements.^[11] Your skin, in effect, has a shutoff valve that a pill doesn’t.

What overdose looks like, in the cases on record

The theoretical mechanism is one thing. The documented cases are more instructive, because they show how far from normal use it takes to get there.

Vitamin D doses in context

From the trials and case reports cited in this article.

A few things stand out in that range:

• A single annual megadose backfired. A randomized trial giving older women at high fracture risk one 500,000 IU dose of vitamin D a year, instead of a smaller daily amount, found it increased the rate of falls by 15% and fractures by 26% compared with placebo.^[12] More vitamin D, delivered the wrong way, made the exact outcome it was meant to prevent worse.
• Even a bone-health combination has a tradeoff. A Women’s Health Initiative trial of 36,282 postmenopausal women given calcium plus vitamin D3 daily for an average of seven years found a 17% higher risk of kidney stones in the supplement group compared with placebo.^[13]
• Documented toxicity cases skew extreme and sustained. One man took large daily doses of vitamin D for more than three years, intending to boost immunity during the COVID-19 pandemic, and developed acute kidney injury from oxalate crystal deposits in his kidneys, requiring temporary dialysis before a partial recovery.^[14]

What this means if you take a daily supplement

None of this is an argument against a normal multivitamin dose. A small pilot trial giving healthy adults 1,000 to 2,500 IU of vitamin D3 a day for 60 days found no excess hypercalcemia or other adverse safety signal at those doses.^[15] The doses behind every toxicity case on record are a different order of magnitude: tens of thousands of IU a day, sustained for weeks, or hundreds of thousands in a single contaminated serving.

What closes the gap between “probably fine” and “definitely fine” is a measured number instead of a guess.

The one test that settles it: a blood test for 25-hydroxyvitamin D is how vitamin D status is determined. Deficiency symptoms are nonspecific and common enough that guesswork alone leads to both missed deficiency and overprescribed supplementation.^[16]

A label alone won’t tell you if a supplement’s stated dose matches what’s inside it, which is exactly the kind of thing worth checking before you add a daily pill on top of whatever you already get from the sun and your diet, covered in more detail in our breakdown of which foods supply vitamin D. Use EatMatrix on your iPhone to scan a supplement label and see exactly what dose and form you’re getting.

For most adults, the practical math isn’t close. Documented toxicity needs a sustained or contaminated megadose, and it shows up in case reports precisely because it’s rare enough to be notable. Low vitamin D shows up in national survey data because it affects roughly half the people measured. Worry about the right one.

Get EatMatrix for iPhone

This article summarises publicly available regulatory assessments and peer-reviewed research about food ingredients. It is general information and is not intended to diagnose, treat, cure, or prevent any disease, and it is not personalised dietary or medical advice. Regulatory positions change; the EatMatrix app reflects the current assessments in our database. If you have an allergy, a medical condition, or are pregnant, talk to a qualified clinician or dietitian before changing what you eat.