Short answer: my personal experiment showed that four months of 5,000 IU/day vitamin D3 changed my measured 25-hydroxyvitamin D from about 33 to 38 ng/mL. That is interesting as an N=1 observation, but it does not establish that 10,000 IU/day is safe, necessary or beneficial for other people.

Editorial note: Originally written by Mikhail Faranov, MD. Substantially revised and updated by Ilya Nikitin, Ph.D. in September–October 2026.

The previous version of this article made exactly that leap. I wrote that 10,000 IU/day “may indeed be feasible, safe and potentially beneficial.” I no longer think my data support that conclusion.

What my experiment actually tells us

A person’s response to vitamin D supplementation depends on baseline status, body size, absorption, sun exposure, assay variation, genetics, adherence and other factors. My small increase while taking 5,000 IU/day mainly tells us how my measured level behaved during that period.

It cannot establish a population dose-response curve, toxicity threshold or ideal target concentration.

What current reference values say

The NIH Office of Dietary Supplements summarizes the National Academies’ tolerable upper intake level for adults as 4,000 IU (100 micrograms) per day. The same fact sheet notes that vitamin D toxicity can cause hypercalcemia, hypercalciuria, kidney injury, soft-tissue calcification and cardiac arrhythmias.

The NIH also notes that 25(OH)D concentrations above about 50 ng/mL can be associated with adverse effects, while the optimal concentration for general health has not been firmly established.

Does that mean doses above 4,000 IU are never used?

No. A clinician may prescribe higher doses for a documented deficiency or a specific medical situation, often for a defined period and with appropriate monitoring. A tolerable upper intake level is not the same thing as a poisoning threshold. It is a population safety boundary for chronic intake without medical supervision.

The 2024 Endocrine Society guideline changed the tone

The 2024 Endocrine Society guideline advises generally healthy adults younger than 75 against routine supplementation beyond the recommended dietary intake and discourages routine 25(OH)D testing in the absence of a specific indication. In a major reversal of its 2011 position, the Society also withdrew its former 30 ng/mL (75 nmol/L) target: it no longer treats a level of 20–29 ng/mL as actionable “insufficiency” in otherwise healthy people. Randomized trials have not established a blood threshold at which supplementation reliably produces a net clinical benefit. This is not a claim that severe deficiency or established bone or absorption disorders should go untreated.

Kidney stones: what do the trials show?

Kidney stones are not solely a risk of extreme vitamin D toxicity. A 2024 systematic review of 10 randomized trials reported a small increase in kidney stones in the vitamin D intervention groups (relative risk 1.10, 95% confidence interval 1.01–1.19; roughly two extra cases per 1,000 participants). Some trials also supplied calcium, so this estimate does not isolate vitamin D taken alone. In a three-year randomized trial, higher daily doses increased hypercalciuria, or calcium excretion in urine, which is a plausible pathway to calcium stones. This does not establish that ordinary dietary-replacement doses cause stones, but it does matter when weighing long-term high-dose supplementation without a clear indication.

Bottom line

I still think the original experiment is worth publishing because real repeated measurements are useful. The corrected conclusion is narrower: 5,000 IU/day produced only a modest rise in my 25(OH)D during this period. It does not justify recommending 10,000 IU/day to readers. Higher-dose vitamin D is a medical dosing question, not a conclusion that can be drawn from one person’s laboratory results.

Sources