science

Vitamin D Supplementation: What Dose Is Actually Backed by Science?

Everyone is deficient in vitamin D. At least, that's what the internet will tell you. The truth is more interesting — and more nuanced than "take 10,000 IU and feel amazing." Some people genuinely need supplementation. Others are taking doses that would make an endocrinologist's eye twitch. Let's sort it out.

What Even Is Vitamin D?

Vitamin D is a fat-soluble hormone precursor your skin makes when exposed to UV-B light. You can also get small amounts from food — fatty fish, egg yolks, fortified dairy. The form that matters in your blood is called 25-hydroxyvitamin D (25(OH)D). That's what gets measured in lab tests.

It works throughout your body: regulating calcium absorption, supporting immune function, influencing mood and inflammation. The problem is that modern life — desk jobs, sunscreen, northern latitudes, darker skin tones — means a meaningful chunk of people run low.

How low? U.S. NHANES data from 2011–2014, covering over 16,000 people, found about 5% of Americans had serum levels in the deficient range (<30 nmol/L), and another 18% were at risk of inadequacy.^[1] Non-Hispanic Black Americans had deficiency rates of 17.5%, compared to 2.1% for non-Hispanic white Americans — a stark gap driven by melanin's effect on UV absorption. That's not nothing, but it's also not "everyone."

What Are People Actually Claiming?

Here's what's circulating on wellness TikTok and Instagram right now:

"You need at least 10,000 IU a day for immune function." Dr. Berg, a popular YouTube/TikTok figure, has called 10,000 IU a "conservative maintenance dose" and suggests people with autoimmune conditions may need far more.
"The RDA is too low — it was set for bones, not optimal health." Some naturopaths and functional medicine practitioners push for blood levels of 60–100 ng/mL, roughly double what most guidelines recommend.
"You can't overdose unless you take hundreds of thousands of IU for months." This one has a kernel of truth — but only a kernel.
"Sun exposure gives you 10,000–25,000 IU, so supplementing that is totally natural." Technically true for peak-summer, full-body sun exposure. Not true for the average person taking a pill in January in Minnesota.

The high-dose movement got enough traction that UK regulators flagged it in 2025. An investigation found 10,000 IU vitamin D3 supplements — 25 times the NHS daily recommendation — being sold on TikTok Shop, Amazon, and eBay.

So… Does It Actually Work?

Yes, with caveats. Vitamin D supplementation clearly works if you're actually deficient. The evidence for benefit in people with already-sufficient levels is much weaker. And the evidence for mega-doses in generally healthy people is essentially nonexistent.

That said, there are some genuinely impressive findings in specific populations. Immune function. Autoimmune disease. Outcomes in critically ill patients. The data here is real — it just doesn't support the "everyone take 10,000 IU" conclusion.

The Real Studies

Autoimmune disease prevention (VITAL trial, 2022). The biggest and best-designed vitamin D trial to date. Hahn et al. enrolled 25,871 U.S. adults (mean age 67) in a randomized, double-blind, placebo-controlled trial. Participants taking 2,000 IU/day for a median of 5.3 years had a 22% reduction in confirmed autoimmune diseases — including rheumatoid arthritis, psoriasis, and autoimmune thyroid disease — compared to placebo (HR 0.78, 95% CI 0.61–0.99).^[2] That's a real, statistically significant effect at a modest dose.

High-dose vs. standard-dose in hospitalized COVID patients (2022). A randomized, double-blind, placebo-controlled pilot trial by De Niet et al. gave 50 patients with COVID-19 and vitamin D deficiency either 25,000 IU/day for 4 days then weekly, or placebo. Hospital stay dropped significantly — 4 days vs. 8 days in the placebo group (p=0.003).^[3] Useful context: these were hospitalized, deficient patients — not healthy people supplementing preventatively.

10,000 IU vs. 2,000 IU in hospitalized COVID patients (2022). A multicenter Spanish RCT by Entrenas-Castillo et al. ran 85 hospitalized patients on either 10,000 IU or 2,000 IU daily for 14 days. The higher dose did raise serum D levels faster (especially in overweight patients) and was associated with shorter hospital stay. Both doses were well-tolerated, with no difference in adverse events between groups.^[4] Again — sick, hospitalized, deficient patients.

Sepsis and mechanical ventilation (2024). Ashoor et al. ran a small RCT of 80 critically ill sepsis patients on mechanical ventilation, all with known vitamin D deficiency. Those receiving 50,000 IU (versus 5,000 IU) had significantly better procalcitonin reduction, improved organ failure scores, and shorter ICU stays.^[5] Impressive — and completely irrelevant to a healthy person scrolling TikTok.

Safety at 3,200–4,000 IU/day (meta-analysis, 2023). A systematic review by Zittermann et al. analyzed 22 RCTs (12,952 participants) taking 3,200–4,000 IU/day for at least 6 months. The risk of hypercalcemia was more than double in the vitamin D group vs. control (RR 2.21, 95% CI 1.26–3.87). Risk of falls and hospitalizations was also modestly elevated.^[6] The effect was small in absolute terms — about 4 cases of hypercalcemia per 1,000 people — but it punctures the claim that doses up to 4,000 IU are universally safe for everyone.

Here's the Fine Print

Most of the compelling high-dose studies involve deficient, sick people. That's an important distinction. When researchers give someone who's at 12 ng/mL a loading dose, they're correcting a real deficit. When a healthy person with 45 ng/mL starts taking 10,000 IU a day "for immune support," they're pushing into uncharted territory without a lab guiding them.

Vitamin D is fat-soluble. Unlike vitamin C, you can't just pee out the excess. It accumulates. Toxicity — called hypervitaminosis D — causes hypercalcemia: nausea, vomiting, excessive thirst, kidney stones, muscle weakness, confusion, and in severe cases, kidney failure and cardiac arrhythmia. It's not common at reasonable doses. But it becomes meaningfully more likely as doses climb past 4,000 IU/day, especially without blood monitoring.

The current tolerable upper intake level from the National Academy of Medicine is 4,000 IU/day. The Endocrine Society's 2024 guidelines went further, stating that healthy adults under 75 generally don't need vitamin D supplementation at all.

Who's Actually Going to Notice a Difference

If you're in one of these groups, supplementation is well-supported and probably worth discussing with a doctor:

People with confirmed deficiency (serum 25(OH)D below 20 ng/mL)
Adults over 70 — absorption decreases with age
People with limited sun exposure — night-shift workers, people in northern climates, those who cover skin for cultural or medical reasons
People with darker skin tones — melanin reduces UV-driven synthesis
People with malabsorption conditions — Crohn's, celiac, gastric bypass
Pregnant women — fetal bone development and immune programming both benefit

If you're a healthy adult in your 30s who gets reasonable sun exposure and eats a varied diet, your need for high-dose supplementation is genuinely unclear. The VITAL trial's autoimmune finding at 2,000 IU is interesting — but it's not a mandate.

Let's Kill Some Myths

"The RDA of 600 IU is just for preventing rickets." Partially true — the RDA was calculated for bone health. But that doesn't mean higher doses automatically produce more benefit. For most outcomes studied, the curve flattens well below 10,000 IU.

"You can't tell if you're deficient without a test." Correct. Symptoms of deficiency — fatigue, muscle weakness, bone pain — are nonspecific. Get a 25(OH)D test before supplementing at high doses. It's a simple blood test and takes the guesswork out.

"Sun exposure proves 10,000 IU is safe to supplement." Sun exposure triggers photodegradation of excess vitamin D in the skin — a feedback loop that oral supplements bypass entirely. Your body self-regulates solar D production. Pills don't have that mechanism.

"High-dose D3 is safe because Dr. [whoever] has 'thousands of before/afters'." Before-and-after photos are not clinical trials. Selection bias, placebo effect, and regression to the mean explain most of those stories.

If You Want to Try It

Here are places to find it without overpaying or getting something sketchy:

Search Vitamin D3 + K2 on Amazon — Most experts recommend pairing D3 with K2 (MK-7 form) to help direct calcium properly. Look for 2,000–5,000 IU D3 unless your doctor advises otherwise.
Browse on iHerb — Wide selection, competitive pricing.
Fullscript — Professional-grade brands.

Heads up: The Amazon link is an affiliate link — we earn a small cut if you buy through it, at no extra cost to you.

Verdict

Vitamin D matters. Deficiency is real and undertreated in certain populations. The evidence for supplementation in people who are genuinely deficient is solid. The VITAL trial's autoimmune data at 2,000 IU is legitimately interesting.

What the evidence does not support: routine 10,000 IU dosing for healthy people, treating megadoses as a cure-all, or supplementing without ever checking your levels. The wellness influencer who says "I take 10,000 IU every day and feel amazing" is offering you an anecdote, not a study.

If you're going to supplement: get your levels tested, aim for 30–50 ng/mL in your blood, use 1,000–2,000 IU to maintain (or up to 4,000 IU if you're correcting a deficit), pair with K2, and recheck in 3 months. That's not exciting. It's just what the data says.

Quick Answers

How much vitamin D should I take daily?

For most healthy adults, 1,000–2,000 IU/day is a reasonable maintenance dose. If you have confirmed deficiency, your doctor may recommend a short-term higher dose (up to 4,000 IU or a weekly 50,000 IU prescription dose) to restore levels, then dropping back to maintenance. Get your levels tested first.

What blood level of vitamin D is "optimal"?

Most mainstream guidelines define sufficiency as 25(OH)D ≥ 20 ng/mL (50 nmol/L). Many clinicians aim for 30–50 ng/mL for general health. Levels above 100 ng/mL are associated with toxicity risk. The claim that 60–100 ng/mL is universally "optimal" is not well-supported by RCT data.

Can you overdose on vitamin D?

Yes. It's rare at doses below 4,000 IU/day, but meaningful risk begins to appear above that threshold — especially without blood monitoring. Toxicity causes hypercalcemia: nausea, kidney stones, confusion, and in extreme cases, cardiac arrhythmia or kidney failure. A 2023 meta-analysis found 3,200–4,000 IU/day more than doubled hypercalcemia risk compared to placebo.

Should I take D3 or D2?

D3 (cholecalciferol) — the animal-derived form — raises serum 25(OH)D more effectively than D2 (ergocalciferol), especially at higher doses. D3 is the standard recommendation. Pair it with K2 in MK-7 form to help shuttle calcium to bones and away from arteries.

Do I need a test before supplementing?

If you're planning to take more than 2,000 IU/day, yes — testing is worth it. Without knowing your baseline, you can't tell whether you're correcting a deficiency or pushing already-adequate levels higher. A 25(OH)D blood test is inexpensive and often covered by insurance. Ask your doctor or use a direct-to-consumer lab service.

References

Herrick KA, et al. "Vitamin D status in the United States, 2011–2014." American Journal of Clinical Nutrition, 2019. https://doi.org/10.1093/ajcn/nqz037
Hahn J, et al. "Vitamin D and marine omega 3 fatty acid supplementation and incident autoimmune disease: VITAL randomized controlled trial." BMJ, 2022. https://doi.org/10.1136/bmj-2021-066452
De Niet S, et al. "Positive Effects of Vitamin D Supplementation in Patients Hospitalized for COVID-19: A Randomized, Double-Blind, Placebo-Controlled Trial." Nutrients, 2022. https://doi.org/10.3390/nu14153048
Castillo ME, et al. "Effect of calcifediol treatment and best available therapy versus best available therapy on intensive care unit admission and mortality among patients hospitalized for COVID-19." Frontiers in Pharmacology, 2022. https://pmc.ncbi.nlm.nih.gov/articles/PMC9289223/
Ashoor T, et al. "Outcomes of High-Dose Versus Low-Dose Vitamin D on Prognosis of Sepsis Requiring Mechanical Ventilation: A Randomized Controlled Trial." Journal of Intensive Care Medicine, 2024. https://doi.org/10.1177/08850666241250319
Zittermann A, et al. "Long-term supplementation with 3200 to 4000 IU of vitamin D daily and adverse events: a systematic review and meta-analysis of randomized controlled trials." European Journal of Nutrition, 2023. https://doi.org/10.1007/s00394-023-03124-w

The content on this site is for informational purposes only and is not medical advice. Always consult a qualified healthcare professional before making decisions about your health.