Vitamin D occupies a strange position in the supplement world. On one hand, it is genuinely essential - severe deficiency causes rickets in children and osteomalacia in adults, and adequate levels are necessary for calcium absorption and bone maintenance. On the other hand, vitamin D has been hyped as a cure for everything from heart disease to depression to COVID-19, and much of that hype has not survived contact with large, well-designed clinical trials.
The result is that most people either assume they need a lot of it, or assume they don’t need any of it. Neither position is quite right. Here is what the research actually says, starting with what is settled and working toward what isn’t.
The basics: what vitamin D does and why it matters#
Vitamin D is a fat-soluble vitamin that functions more like a hormone than a typical dietary vitamin. Your body makes it when UVB radiation from sunlight hits cholesterol in your skin, converting 7-dehydrocholesterol to vitamin D3 (cholecalciferol). You can also get it from food - fatty fish, egg yolks, fortified milk - and from supplements, which come in two forms: D2 (ergocalciferol) from plant sources and D3 (cholecalciferol) from animal sources.
Once produced or consumed, vitamin D travels to the liver, where it is converted to 25-hydroxyvitamin D - the form measured in blood tests. From there it goes to the kidneys, where it is converted to the active hormonal form, 1,25-dihydroxyvitamin D. This active form regulates calcium absorption in the gut, mobilizes calcium from bone when dietary intake is low, and influences the expression of hundreds of genes through the vitamin D receptor, which is present in most tissues.
That last point is important. The fact that vitamin D receptors are found in immune cells, muscle tissue, the brain, and the cardiovascular system is the biological rationale for why vitamin D might affect all those systems. But receptor presence does not equal clinical benefit - a point that gets lost in a lot of supplement marketing.
D2 vs D3: does the form matter?#
Yes, it matters. D3 (cholecalciferol) is the form your skin produces and is consistently more effective at raising and maintaining blood levels of 25-hydroxyvitamin D than D2 (ergocalciferol). A 2012 meta-analysis published in the American Journal of Clinical Nutrition reviewed seven randomized trials and found that D3 raised serum levels significantly more than D2 at equivalent doses. D3 also appears to be more stable in storage and has a longer half-life in the body.
D2 is still used - it is the form available by prescription in high-dose formulations - and it does work, but it is less efficient. If you are buying an over-the-counter supplement, D3 is the better choice. Most quality supplements use it. If the label just says “vitamin D” without specifying, check the ingredients; if it doesn’t say, assume D2 or contact the manufacturer.
The VITAL trial and what it changed#
The VITAL study (VITamin D and OmegA-3 TriaL), published in 2019 in the New England Journal of Medicine, is the largest and most rigorous randomized trial of vitamin D supplementation ever conducted. It enrolled 25,871 participants - men aged 50 and older and women aged 55 and older - and randomized them to receive 2,000 IU of vitamin D3 daily or placebo, with a median follow-up of 5.3 years.
The results were sobering for anyone who expected vitamin D to be a panacea. The primary endpoints were incidence of invasive cancer and major cardiovascular events. Vitamin D supplementation did not reduce the risk of either. There was no difference between the vitamin D group and the placebo group for heart attack, stroke, or cardiovascular death. For overall cancer incidence, same story - no significant reduction.
There were hints of benefit in secondary analyses. Cancer mortality showed a suggestive reduction that became statistically significant in analyses that excluded the first two years of follow-up - meaning it might take time for any protective effect to emerge. An ancillary study, VITAL-DEP, examined depression and found no overall benefit for preventing depression or improving mood, though a signal toward benefit was present in certain subgroups.
The VITAL findings don’t mean vitamin D is useless. They mean it is not the broad-spectrum disease preventer that earlier observational studies suggested. And that disconnect - between strong observational associations and null randomized trials - points to a fundamental problem in vitamin D research: confounding.
The confounding problem: why low vitamin D is associated with everything#
Observational studies have consistently found that people with low vitamin D levels have higher rates of heart disease, cancer, diabetes, depression, autoimmune disease, and all-cause mortality. But people with low vitamin D also tend to be older, heavier, less active, spend less time outdoors, have poorer diets, and have more chronic illness. All of those things independently increase disease risk.
When you adjust for these confounders in observational data, the associations weaken substantially. And when you run randomized trials like VITAL - where confounding is eliminated because people are assigned to vitamin D or placebo by chance - the effects largely disappear. The low vitamin D is likely a marker of poor health rather than a cause of it. This is not a fringe position. It is the consensus interpretation of the VITAL investigators and most epidemiologists who work in this area.
What about respiratory infections and COVID-19?#
A 2021 meta-analysis published in The Lancet Diabetes and Endocrinology combined individual participant data from 46 randomized controlled trials and found that vitamin D supplementation reduced the risk of acute respiratory infection. Importantly, the benefit was concentrated in people with baseline 25-hydroxyvitamin D levels below 25 nmol/L (10 ng/mL) - that is, people who were genuinely deficient. In people with adequate levels, the effect was small and not statistically significant.
When COVID-19 arrived, observational studies quickly found associations between low vitamin D and severe outcomes. Randomized trials testing vitamin D as a treatment for hospitalized COVID-19 patients have been largely negative. The NIH COVID-19 treatment guidelines state that there is insufficient evidence to recommend vitamin D for the prevention or treatment of COVID-19 outside the context of a clinical trial.
That does not mean vitamin D plays no role in immune function - it does. The vitamin D receptor is expressed on immune cells, and vitamin D modulates both innate and adaptive immunity. But the immune benefits of correcting a deficiency are not the same as the immune benefits of supplementing a normal level.
The reference range controversy: what level is actually optimal?#
This is where things get genuinely contentious among researchers and clinicians. The Institute of Medicine (now the National Academy of Medicine) defines vitamin D sufficiency as a serum 25-hydroxyvitamin D level of 20 ng/mL (50 nmol/L) or higher. This threshold is based on bone health outcomes - calcium absorption, parathyroid hormone suppression, fracture risk - where the evidence is solid.
The Endocrine Society, by contrast, recommends levels above 30 ng/mL (75 nmol/L), citing potential non-skeletal benefits. Many functional medicine practitioners aim for 40-60 ng/mL or even higher.
Who is right? The IOM’s position has the stronger evidentiary basis because it ties the threshold to outcomes that have been demonstrated in randomized trials. The higher thresholds endorsed by the Endocrine Society are based partly on observational data and partly on the assumption that higher levels are better - an assumption that randomized trials like VITAL have not confirmed.
There is also evidence that more is not always better. A 2010 study published in the Journal of the American Medical Association found that fracture risk actually increased at very high vitamin D levels. A 2018 study in JAMA Cardiology reported a U-shaped relationship with cardiovascular mortality, with increased risk at both low and very high levels. The sweet spot for bone health and overall safety appears to be in the 20-40 ng/mL range for most people.
Who should test, and who can supplement without testing?#
Testing vitamin D levels is reasonable for people at high risk of deficiency: older adults, especially those in institutional settings; people with dark skin living at northern latitudes; people who cover most of their skin for cultural or religious reasons; people with malabsorption conditions like celiac disease or inflammatory bowel disease; people who have had gastric bypass surgery; and people with osteoporosis or osteopenia.
For a healthy adult who spends time outdoors, eats a varied diet, and has no specific risk factors, routine testing is less clearly necessary. But if you are going to take doses above 2,000 IU per day for more than a few months, testing is worth doing. You are flying blind otherwise.
If you do test, get the right test: serum 25-hydroxyvitamin D, not 1,25-dihydroxyvitamin D. The latter can be normal or even elevated in deficiency because the parathyroid hormone drives conversion, and it is not a reliable indicator of vitamin D status.
Dosing: what the numbers actually mean#
The RDA for vitamin D is 600 IU per day for adults up to age 70 and 800 IU per day for those over 70. These numbers are designed to maintain bone health in people with minimal sun exposure and assume normal absorption. They are conservative, but they are evidence-based.
For people with confirmed deficiency (levels below 20 ng/mL), typical treatment involves 2,000-5,000 IU per day for 8-12 weeks followed by retesting and a lower maintenance dose. High-dose bolus regimens - like 50,000 IU once weekly - are also used, usually under medical supervision, but there is some evidence that daily dosing is more effective than intermittent high doses.
The tolerable upper intake level set by the National Academy of Medicine is 4,000 IU per day for adults. This is not a target - it is a safety ceiling. Doses above this level, taken chronically, increase the risk of hypercalcemia. Toxicity from vitamin D itself is rare but serious when it occurs: symptoms include nausea, vomiting, weakness, confusion, and kidney damage from calcium deposition. Toxicity is almost always the result of supplement errors - people taking 10,000 IU or more daily for extended periods - rather than sun exposure or diet.
Interactions and cofactors: calcium, magnesium, K2, and more#
Vitamin D increases calcium absorption. That is its primary job. If you are taking vitamin D, you need adequate calcium intake - ideally from food. The RDA for calcium is 1,000-1,200 mg per day for adults. Dairy, fortified plant milks, leafy greens, and canned fish with bones are good sources. Supplementing calcium in addition to vitamin D is generally unnecessary if your dietary intake is reasonable and has not been shown to provide additional cardiovascular or bone benefit in most populations.
Magnesium is a cofactor for the enzymes that convert vitamin D into its active form. If you are magnesium-deficient, vitamin D metabolism can be impaired regardless of how much you take. This is one reason magnesium and vitamin D are often recommended together. The relationship is bidirectional: high-dose vitamin D can also deplete magnesium by increasing its utilization.
Vitamin K2 is frequently paired with vitamin D in supplements, based on the idea that vitamin D increases calcium absorption while vitamin K2 directs that calcium into bone rather than soft tissues like arteries. The mechanistic rationale is plausible - vitamin K2 activates matrix Gla protein, which inhibits arterial calcification, and osteocalcin, which binds calcium in bone. However, the clinical trial evidence for the D-plus-K2 combination is limited. A 2015 trial found that vitamin K2 supplementation slowed arterial calcification in people with pre-existing calcification, but most studies have been small. Adding K2 is unlikely to cause harm and may offer theoretical benefit, but the evidence for it as a necessary cofactor is thinner than many supplement companies imply.
Bottom line#
Test before you supplement if you can, especially if you are in a high-risk group. If your 25-hydroxyvitamin D level is above 20 ng/mL, routine supplementation beyond the RDA has not been shown to provide meaningful additional benefit for most health outcomes. If you are deficient, 2,000 IU per day of D3 is a reasonable starting point, with retesting in three months to confirm you have reached target levels.
The strongest evidence for vitamin D is bone health - preventing rickets, osteomalacia, and, in combination with calcium, reducing fracture risk in older adults. For cardiovascular disease, cancer, and respiratory infections, the randomized trial evidence is either negative or shows benefit only in people who were deficient to begin with.
The gap between what vitamin D was expected to do, based on observational studies, and what it has been shown to do in large randomized trials is one of the more important cautionary tales in nutrition science. Nutrient deficiencies should be corrected. Nutrient supplementation beyond sufficiency should not be assumed to provide additional protection. The two are not the same thing.
