The Case for Caution: Rethinking Synthetic Vitamin D

A hormone, not a vitamin

Vitamin D is sold like a vitamin but behaves like a steroid hormone, and that difference is the root of every concern in this post. Its active form, 1,25-dihydroxyvitamin D (calcitriol), switches genes on and off in the gut, bone, kidney and immune cells.

For years the message has been simple: test one blood marker, and if it is low, take more. The research tells a more complicated story. Higher doses have repeatedly failed to deliver what was expected, and in some trials they did harm.

This piece sets out what the studies actually show, what remains uncertain, and which popular claims go further than the evidence.

Why a pill is not the same as sunlight

Skin has a built-in brake; a capsule does not. With prolonged sun exposure, excess pre-vitamin D in the skin is converted into inactive by-products (lumisterol and tachysterol), so sunlight alone does not cause vitamin D toxicity.

Oral cholecalciferol bypasses that brake. Whatever is swallowed is absorbed and passed to the liver, so the dose, not the body, sets the ceiling.

The body does still regulate the active hormone tightly, mainly through parathyroid hormone (PTH), calcium and phosphate. But that control has limits, and very large or very sudden doses test them, as the trials below show.

More is not better: high daily doses and bone

The Calgary Vitamin D Study gave 311 healthy adults aged 55-70 either 400, 4,000 or 10,000 IU daily for three years. The researchers expected higher doses to build stronger bone. The opposite happened.

Bone loss rose with dose, and so did hypercalciuria, a risk factor for kidney stones. The authors linked the effect to raised bone-resorption markers alongside suppressed PTH. They called the finding hypothesis-generating, but it sits awkwardly with the idea that higher is always safer. source

These participants were not deficient at the start. The trial suggests that once stores are adequate, extra vitamin D brings no bone benefit and may carry a cost.

Big doses, big swings: bolus dosing

One large dose is not the same as many small ones, and the clearest harm in the trial literature comes from bolus dosing. In an Australian trial of 2,256 women aged 70 and over, a single 500,000 IU oral dose each autumn or winter led to 15% more falls and 26% more fractures than placebo. source

The excess risk was greatest in the three months after each dose, when blood levels peaked. The authors suggested the sharp rise, the fall that followed, or both, might be responsible.

Large loading doses are still used in practice. This trial is a reason to question any regimen that delivers vitamin D in surges rather than a steady, modest supply.

Stored vs active D: why one number misleads

Standard testing measures only the stored form, 25(OH)D, yet it is the active form, 1,25(OH)₂D, that does the work. Looking at one without the other can lead to the wrong conclusion.

Low stored D does not always mean deficiency. Immune cells such as macrophages can make active vitamin D outside kidney control. In granulomatous conditions like sarcoidosis this is well recognised, and it can produce normal or high active D from a low stored level. Researcher Trevor Marshall raised this point in a 2002 letter, arguing studies should give the active hormone more weight. source

There is also a feedback loop. In a 1984 study, giving active vitamin D to healthy volunteers completely blocked the expected rise in stored D after a large dose. source So a low 25(OH)D can partly reflect high active hormone, not a lack of intake.

What range looks best? A 2024 analysis of 455,062 German patients found inflammation markers (CRP and white cell count) lowest at a stored level of about 32-42 ng/mL (80-105 nmol/L) and an active level of about 60 pg/mL. The calcium-phosphate product rose sharply above 50 ng/mL (125 nmol/L). The authors proposed targets of 30-50 ng/mL and 50-70 pg/mL, while noting the findings are observational and need confirming against clinical outcomes. source

The practical point: chasing an ever-higher stored number offers no clear advantage and may raise mineral risks, and the active hormone is worth measuring when the picture does not add up.

The ratio between stored and active D, and what it may mean over time

The relationship between the two forms tells you more than either number alone, and researchers increasingly use ratios to capture it. Three are in use, each answering a different question.

A high active-to-stored ratio has more than one cause. When stored D is low, PTH rises and drives more conversion, so active D can be normal or even high. (source) Granulomatous disease and kidney disease also shift the ratio. Mangin's infection model is one explanation, not the only one, which is why PTH, calcium and kidney function belong alongside the ratio.

Long-term evidence is thin, and it cuts both ways. In a German cohort of 3,258 patients followed for a median of 7.7 years, low stored D and low active D were each independently linked to higher death rates. (source) In heart failure, a low ratio of active D to PTH strongly predicted cardiovascular death. (source) Neither study shows that a persistently high active-to-stored ratio causes harm, and none has tracked what years of supplementation do to the ratio.

The case for measuring both is strong; the case for fixed targets is not yet made. A ratio shows how the body is handling vitamin D, not just how much is stored. Thresholds such as "healthy between 50 and 70" circulating online are not validated. Following one person's ratio over time, with PTH and calcium, is more meaningful than comparing a single result with a population cut-off.

What does "deficient" actually mean?

A stored vitamin D level of 20 ng/mL (50 nmol/L) meets the needs of nearly everyone, yet many patients at that level are told they are short and advised to supplement. The gap comes from which cut-off is used, and that cut-off was never settled.

The higher cut-off spread quickly. Many commercial laboratories adopted 30 ng/mL, so results between 20 and 30 are still flagged as "low" or "insufficient" on reports. At that line, roughly 80% of Americans appear inadequate. (source)

Independent experts say this inflated the problem. A 2016 New England Journal of Medicine commentary argued that using the 20 ng/mL adequacy figure as a dividing line wrongly labels most people whose needs are met as deficient, creating the appearance of a pandemic. On the deficiency threshold, fewer than 6% of Americans were deficient. (source)

The higher cut-off came with conflicts of interest. The 2011 guideline was overseen by Dr Michael Holick, who disclosed long-standing consultancy for Quest Diagnostics, a major vitamin D testing laboratory, and payments from companies with vitamin D products. He has said this did not influence his advice. Testing for US Medicare patients rose 547% between 2007 and 2016. (source) This shows conflicted guidance that benefited industry, not proof of deliberate manipulation.

In 2024 the guideline's own society stepped back. The Endocrine Society no longer endorses any threshold for deficiency, insufficiency or sufficiency, and advises against routine testing of healthy adults. (source) Lab reports have been slow to catch up.

The number itself has limits too. Results vary between laboratories and methods, so a value near a cut-off can fall either side of it. (source) Stored D also drops during inflammation: after hip or knee replacement it fell by about a fifth within 48 hours. (source)

What this means in practice. A single result between 20 and 30 ng/mL, in someone who is well, is not by itself a diagnosis or a reason for high-dose supplements. Deficiency in the body shows up as raised PTH, low phosphate, raised alkaline phosphatase or bone symptoms, and those are what should guide treatment.

The cofactor web: magnesium, vitamin A and iron

Vitamin D never acts alone, and supplementing it in isolation ignores the nutrients it depends on and interacts with.

Magnesium. The enzymes that activate and break down vitamin D are magnesium-dependent. In a randomised trial, magnesium raised vitamin D in people with low levels and lowered it in those with high levels, suggesting magnesium acts as a regulator. source Many adults fall short of recommended magnesium intake, so a vitamin D pill taken without attention to magnesium may work poorly.

Vitamin A. Vitamin D's receptor (VDR) must pair with the retinoid X receptor (RXR), a receptor vitamin A metabolites also use. The two vitamins are known to antagonise each other: a Swedish human study found vitamin A blunted the calcium response to vitamin D. source This supports keeping the fat-soluble vitamins in balance rather than megadosing one. Whether ordinary vitamin D doses cause a functional vitamin A deficiency in people has not been shown.

Iron. Vitamin D lowers hepcidin, the hormone that controls iron release and absorption. In a small pilot trial (28 adults), one 250,000 IU dose cut hepcidin by 73% within a week, with no change in ferritin. source That could help anaemia of inflammation. It also means high doses alter iron handling in ways not yet studied long term, a gap worth naming.

Copper and ceruloplasmin. Ceruloplasmin is the copper-carrying enzyme that lets iron leave the liver and load onto transferrin. Direct human evidence that vitamin D supplementation lowers it is lacking. There is an indirect route: in rats, a high-calcium diet lowered plasma copper and ceruloplasmin activity, and high-dose vitamin D raises calcium absorption. (source) Vitamin A matters here too: retinoic acid increased ceruloplasmin production in rats. (source) When ceruloplasmin is absent, iron builds up in the liver, as copper-deficient rats showed. (source) Against this, a small pilot study in keratoconus found vitamin D supplementation increased the availability of copper inside cells. (source) The vitamin D-copper-iron chain is plausible but untested in people; the best-established cause of copper deficiency remains high-dose zinc.

Calcification: what the evidence does and doesn't show

True vitamin D toxicity causes hypercalcaemia, which can deposit calcium in kidneys and blood vessels. That is established. Whether everyday supplement doses calcify arteries is far less clear.

Animal studies show striking arterial damage, but usually at very high doses. Fred Kummerow's group at the University of Illinois found that swine fed excess vitamin D3 for three months developed coronary atherosclerosis, with lesions rising in proportion to dose and resembling human disease (source: Taura et al., Tohoku J Exp Med, 1979). Rabbits injected with cholecalciferol twice weekly developed aortic calcification with raised vitamin D receptor levels in arterial smooth muscle. source In mice with a genetic calcification disorder, even mild vitamin D plus calcium worsened arterial calcification, while normal mice were unaffected. source

In humans the short-term picture is reassuring, but long-term data are missing. A follow-up of the Calgary trial found 10,000 IU daily for three years did not change arterial calcification compared with 400 IU. Three years is short for a process that develops over decades, so this rules out rapid harm, not long-term risk. source

The honest conclusion: calcification is a real risk with toxicity and may matter more for people with kidney disease, stone history or inherited calcification disorders. It has not been shown at typical doses in healthy people, and no study has followed people long enough to rule it out.

A population-wide experiment without the long-term data

High-dose vitamin D moved from a treatment for deficiency to an everyday habit in about fifteen years, faster than the evidence on long-term safety. US national survey data show the shift.

Sources: Rooney et al., JAMA 2017; Nutrients 2018 review. The lead author of the survey study noted there's "not much research on longer-term health outcomes on high-dose supplements." (source)

Testing has not kept pace with use. Many people now self-dose from shops and online without any baseline test, while others are tested repeatedly with only the stored marker measured. Neither approach shows what the dose is doing to calcium, PTH or the active hormone.

Reports of harm have risen with use. US poison centre exposures to vitamin D rose from about 196 a year (2000-2005) to about 4,535 a year (2005-2011), most of them unintentional and many in young children. (source) A UK review found most toxicity cases were preventable and linked to high-dose over-the-counter products, unlicensed preparations and formulation errors. (source) UK clinicians have since reported adults self-dosing at 10,000 IU daily after the pandemic, and a cluster of 12 children with toxicity from mis-made prescribed products in early 2025. (source; source)

What sustained high dosing does to the body. Short of outright toxicity, the trials point to a consistent set of changes:

• Calcium handling: more calcium in the urine, rising with dose, a risk for kidney stones (Calgary trial).
• Bone turnover: PTH suppressed and bone-resorption markers raised, with dose-related bone loss (Calgary trial).
• Feedback loops: the active hormone suppresses the liver's production of stored D, so the two markers can move in unexpected directions (Bell, 1984).
• Iron regulation: hepcidin falls sharply after a large dose, altering iron handling in ways not studied long term (Smith, 2017).
• Cofactor demand: the enzymes that make and clear vitamin D depend on magnesium, which many people lack (Dai, 2018).
• At toxic levels: hypercalcaemia, with thirst, frequent urination, nausea, confusion, kidney injury and, rarely, death.

None of these is a reason to fear modest doses that correct a real deficiency. Together they are a reason to question routine high dosing without testing, and to ask for long-term studies before treating it as harmless.

Where the online claims outrun the evidence

A sceptical case is only as strong as its weakest claim, and several popular ones are not supported. Leaving them out makes the real concerns harder to dismiss.

The hypothesis that chronic infection drives high active D (associated with Meg Mangin and the Marshall Protocol) is worth knowing about. It remains outside the evidence base and has not been tested in controlled trials.

Practical takeaways

The evidence points to moderation, not avoidance: correct a genuine deficiency, then stop chasing higher numbers.

• Prefer sunlight and food where possible: oily fish, eggs, liver and cod liver oil, with sensible sun exposure. Cod liver oil supplies vitamins A and D together, as nature pairs them, but it is rich in vitamin A: a tablespoon of standard oil holds about 1,360 IU of D and 4,080 mcg of A, above the adult vitamin A upper limit of 3,000 mcg a day. A teaspoon is a more sensible amount, labels vary widely by brand, and vitamin A totals need extra care in pregnancy and alongside multivitamins.
• Avoid megadoses and bolus doses. The clearest harms came from very large single doses and from 4,000-10,000 IU daily in people who were not deficient.
• Test more than one marker when results or symptoms don't fit: stored and active vitamin D, calcium, PTH, phosphate and magnesium.
• Mind the cofactors. Adequate magnesium matters, and fat-soluble vitamins are better kept in balance than taken as isolated high doses.
• Take extra care with kidney disease, kidney stones, sarcoidosis or other granulomatous conditions, where vitamin D can raise calcium unpredictably.
• Know the warning signs of excess: thirst, frequent urination, nausea, constipation, confusion and fatigue warrant a calcium check.

Suggested blood panel

No single result shows what vitamin D is doing; these tests read together, taken when well rather than during an illness, give a far fuller picture.

Optional, where available: ceruloplasmin oxidase activity, the closest measure of working, copper-loaded ceruloplasmin.

Anyone taking high doses, or with any of the conditions above, should review them with a clinician who will test rather than assume. And because supplements can interact with prescribed medicines, tell your GP or pharmacist about everything you take, and never stop a prescribed medicine without speaking to your doctor.

References

1. Burt LA et al. Effect of high-dose vitamin D supplementation on volumetric bone density and bone strength. JAMA. 2019;322(8):736-745. Summary, University of Calgary
2. Sanders KM et al. Annual high-dose oral vitamin D and falls and fractures in older women. JAMA. 2010;303(18):1815-1822. Abstract
3. Li X et al. Target values for 25-hydroxy and 1,25-dihydroxy vitamin D based on their associations with inflammation and calcium-phosphate metabolism. Nutrients. 2024;16(16):2679. Full text
4. Bell NH et al. Evidence that 1,25-dihydroxyvitamin D3 inhibits the hepatic production of 25-hydroxyvitamin D in man. J Clin Invest. 1984;74(4):1540-1544. Abstract
5. Marshall TG. Puzzling vitamin D results. CMAJ. 2002;167(8):849. Letter
6. Dai Q et al. Magnesium status and supplementation influence vitamin D status and metabolism. Am J Clin Nutr. 2018;108(6):1249-1258. Summary, Vanderbilt
7. Johansson S, Melhus H. Vitamin A antagonizes calcium response to vitamin D in man. J Bone Miner Res. 2001;16(10):1899-1905. Thesis
8. Smith EM et al. High-dose vitamin D3 reduces circulating hepcidin concentrations. Clin Nutr. 2017;36(4):980-985. Record
9. 1,25-dihydroxyvitamin D3 receptor is upregulated in aortic smooth muscle cells during hypervitaminosis D. Life Sciences. 2002. Abstract
10. Vitamin D and calcium supplementation accelerate vascular calcification in a model of pseudoxanthoma elasticum. Int J Mol Sci. 2022. Abstract
11. Calgary Vitamin D Study secondary analysis: high-dose vitamin D and arterial calcification (NCT01900860). Abstract