Vitamin D: Bones, Testing, Supplements, and the Risks of Megadoses
Evidence-based guide to vitamin D: cholecalciferol metabolism, calcium homeostasis, immune function, 25(OH)D blood testing, dosages, and toxicity risks.
Evidence-based guide to vitamin D: cholecalciferol metabolism, calcium homeostasis, immune function, 25(OH)D blood testing, dosages, and toxicity risks.
Vitamin D occupies a special place among nutrients. It is called a vitamin, but biologically it behaves more like a precursor to a steroid hormone: the body can synthesize it in the skin under ultraviolet exposure, then modify it sequentially in the liver and kidneys and convert it into an active molecule that interacts with nuclear receptors and influences gene expression.
Its best-established and least controversial role is maintaining normal calcium and phosphate metabolism, bone mineralization, and musculoskeletal function. Severe vitamin D deficiency causes rickets in children and osteomalacia in adults.
The situation is much more complicated when it comes to the many proposed “extraskeletal” effects. Vitamin D receptors are indeed present in many tissues, and laboratory and observational studies link vitamin D to immune function, glucose metabolism, cardiovascular regulation, inflammation, and numerous other processes. But the presence of a receptor or an elegant molecular mechanism does not prove that raising vitamin D levels with supplements will protect a person from infections, diabetes, cancer, or cardiovascular disease.
This is why the modern view of vitamin D has become much more restrained than it was a decade ago. The goal today is not to push 25(OH)D to the highest possible number, but to ensure adequate intake, identify true clinical deficiency in people who are genuinely at risk, and avoid uncontrolled high-dose supplementation.
The two main forms are vitamin D3, or cholecalciferol, and vitamin D2, or ergocalciferol.
D3 is produced in the skin from 7-dehydrocholesterol under UVB radiation and is also found in some animal-derived foods. D2 is produced, among other pathways, when ergosterol in fungi and yeast is exposed to ultraviolet radiation.
Neither D2 nor D3 is the main hormonally active form. After entering the body, vitamin D undergoes a sequence of metabolic conversions.
First, primarily in the liver, it is hydroxylated to form 25-hydroxyvitamin D — 25(OH)D, or calcidiol. This is the main circulating form and the one usually measured when vitamin D status is assessed. Its circulation time is roughly two weeks, so it reflects the combined contribution of food, supplements, and skin synthesis much better than the active hormone does.
The next step occurs mainly in the kidneys: 25(OH)D is converted into 1,25-dihydroxyvitamin D — 1,25(OH)₂D, or calcitriol. Calcitriol is the highly active hormonal metabolite that binds to the vitamin D receptor, or VDR. Local production of calcitriol also occurs in several other tissues, where it participates in paracrine and autocrine regulation.
This system is tightly controlled by calcium, phosphate, parathyroid hormone, kidney function, and multiple enzymes. So the idea that “more vitamin D means more active hormone and therefore better health” is physiologically incorrect.
Calcitriol increases the intestine’s ability to absorb calcium and phosphate. This helps maintain mineral concentrations required for normal bone formation and remodeling.
In severe deficiency, a child develops impaired mineralization of growing bone, resulting in rickets. In adults, impaired mineralization of already formed bone is called osteomalacia.
This is not the same as osteoporosis.
In osteomalacia, the main problem is inadequate mineralization of the bone matrix. In osteoporosis, bone mass decreases and bone microarchitecture deteriorates. The two conditions can coexist, but their underlying mechanisms are different.
Vitamin D also contributes to normal neuromuscular function. However, supplementation studies in people who already have adequate vitamin D status show much less dramatic results than physiology alone might suggest: additional vitamin D does not consistently reduce falls or fractures.
This is one of the main reasons vitamin D has generated so much controversy.
The VDR is present in many cell types. Vitamin D influences the expression of genes involved in immune responses, cell proliferation, inflammation, neuromuscular function, and glucose metabolism. Some immune cells can locally convert 25(OH)D into active calcitriol.
In laboratory models, this is extremely compelling.
The problem appears when we move from molecular biology to real-world disease prevention.
People with low 25(OH)D do appear to have higher rates of some chronic diseases. But an observational association does not prove causality. A person with chronic illness may spend less time outdoors, eat less well, have obesity, move less, and at the same time have a lower 25(OH)D level.
Large clinical trials have not confirmed the idea that raising vitamin D levels in most otherwise healthy adults is a universal strategy for preventing cancer, cardiovascular disease, depression, or diabetes.
That is why the presence of VDR in an immune cell cannot be translated directly into a recommendation to “take vitamin D for immunity.” Physiological involvement and proven benefit from supplementation are two different questions.
For healthy populations, the National Academies of Sciences, Engineering, and Medicine use the following recommended intake levels:
For infants during the first year of life, the reference intake is 400 IU, or 10 mcg per day.
One microgram of vitamin D equals 40 IU.
These values refer to total intake and are designed to meet the needs of most healthy people even with minimal sun exposure. They do not mean that every adult must take a tablet containing exactly 600 IU every day: some vitamin D can come from food and some can be synthesized in the skin.
In 2024, the Endocrine Society published a new clinical guideline on vitamin D that differs substantially from the approach that had been popular for many years.
For healthy adults under age 75, the Society does not recommend routinely taking vitamin D in amounts above established dietary reference intakes solely for disease prevention.
Certain groups are considered separately. Additional empirical intake may be reasonable in children and adolescents, pregnant individuals, adults over 75, and some people with prediabetes.
The more important change is this: the guideline does not support routine measurement of 25(OH)D in healthy people without specific indications and does not define a universal blood concentration that everyone should aim for.
This is a major shift away from the era when almost every level below 30 ng/mL was automatically labeled “insufficient” and treated with the goal of pushing it into the 40–60 ng/mL range.
This is where one of the most important debates begins.
For years, a common classification was:
below 20 ng/mL — deficiency;
20–30 ng/mL — insufficiency;
30–50 or 30–60 ng/mL — optimal.
Today, that table should not be presented as a universal medical standard.
The Food and Nutrition Board at the National Academies interprets the evidence differently:
below 12 ng/mL, or 30 nmol/L — a level associated with risk of true deficiency and impaired bone health;
12–20 ng/mL, or 30–50 nmol/L — a range in which some individuals may have inadequate status;
20 ng/mL and above, or 50 nmol/L and above — considered sufficient for most healthy people with respect to bone health.
At the same time, values above 50 ng/mL, or 125 nmol/L, should not automatically be considered “even better”: they may be associated with adverse effects.
The Endocrine Society went even further in its 2024 guideline and declined to establish universal thresholds for “insufficiency,” “sufficiency,” or target levels in healthy people for disease prevention because clinical trials do not allow such cutoffs to be defined reliably.
So a value of 32 ng/mL is not automatically better than 24 ng/mL, and 50 ng/mL is not a goal that everyone needs to reach.
When vitamin D assessment is actually indicated, the main laboratory marker is total serum 25(OH)D.
It reflects vitamin D from food and supplements as well as vitamin D produced in the skin, and it remains in circulation long enough to provide a useful overall picture.
Even this test is not perfect. Results can vary between assays and laboratories, which is why international programs have been developed to improve standardization. The same biological sample can produce somewhat different results depending on the laboratory method used.
That means a small difference such as 27 versus 30 ng/mL should not be treated as crossing a magical border between disease and health.
Testing 1,25(OH)₂D, or calcitriol, is indeed a poor way to diagnose ordinary vitamin D deficiency.
Its concentration is tightly regulated by parathyroid hormone, calcium, and phosphate, and its circulation time is measured in hours.
When vitamin D stores fall, the body can increase parathyroid hormone secretion and maintain calcitriol production. A person can therefore have low 25(OH)D while 1,25(OH)₂D remains normal or even elevated.
The active metabolite is usually measured not as part of a routine “vitamin D checkup,” but in selected clinical situations such as certain calcium disorders, kidney disease, granulomatous diseases, or rare inherited disorders of vitamin D metabolism.
Not necessarily.
This is another major difference between current guidance and common practice over the past several years.
The Endocrine Society recommends against routine 25(OH)D screening in healthy adults when there are no other indications. This includes otherwise healthy people with obesity or darker skin pigmentation. The USPSTF also concludes that evidence is insufficient to recommend population screening in asymptomatic adults.
That does not mean the test is never useful.
Testing can be clinically justified when there is suspected osteomalacia or rickets, hypocalcemia, certain bone disorders, significant malabsorption, selected liver or kidney diseases, some endocrine disorders, or other situations in which the result would actually change management.
The distinction is important: testing for a clinical indication is useful diagnostic medicine; annual testing of every healthy person is not supported by strong evidence.
Several conditions are indeed associated with a higher probability of lower 25(OH)D.
These include limited sun exposure, older age, darker skin pigmentation, disorders of fat absorption, certain intestinal diseases, bariatric surgery, and obesity.
But even here, oversimplification should be avoided.
For example, people with obesity tend to have lower 25(OH)D on average. This was once explained almost entirely by the idea that fat-soluble vitamin D becomes “trapped” in adipose tissue. Today, the mechanism is considered more complex and may involve volumetric dilution, tissue distribution, and metabolic differences. Obesity itself still does not mean that every asymptomatic person needs routine vitamin D testing.
Darker skin also reduces the efficiency of cutaneous vitamin D synthesis under the same UVB exposure because melanin absorbs UVB. But fixed claims such as “synthesis falls two- to fivefold” are too crude, and the clinical implications of lower total 25(OH)D across different populations remain an area of ongoing discussion.
Vitamin D3 is produced in the skin under UVB radiation, so latitude and season do matter.
During winter at higher latitudes, UVB can become insufficient for effective synthesis. Classic studies, for example, found no detectable previtamin D3 production during winter months in Boston and an even longer “vitamin D winter” in Edmonton.
But turning this into a universal rule that “north of the 37th parallel, vitamin D synthesis completely stops from October to April” is incorrect.
UVB availability is influenced by:
Because so many variables are involved, it is impossible to define a single universal amount of sun exposure that guarantees adequate vitamin D synthesis for everyone. NIH specifically notes that precise recommendations for sun exposure are difficult for this reason.
There is also a more important issue: ultraviolet radiation is a carcinogen. Deliberately increasing unprotected UVB exposure solely to raise vitamin D is therefore not an ideal strategy. Vitamin D needs can be met through food and supplements without intentionally increasing skin cancer risk.
In theory, sunscreen reduces UVB penetration and therefore can reduce cutaneous vitamin D synthesis.
In real life, however, people rarely apply sunscreen thickly and evenly enough, and reapply it consistently enough, to completely block vitamin D production.
So the claim that “SPF causes vitamin D deficiency” is too categorical. NIH notes that the real-world effect of ordinary sunscreen use on vitamin D status remains uncertain.
There is no reason to abandon sun protection in order to obtain vitamin D.
Natural food sources of vitamin D are relatively limited.
The most notable amounts are found in fatty fish such as salmon, trout, mackerel, and herring. Vitamin D is also present in fish liver oils, egg yolks, and some animal-derived foods.
In many countries, fortified foods play an important role: milk, plant-based beverages, breakfast cereals, and other foods may have vitamin D added intentionally.
This is why someone who rarely eats fish and does not consume fortified foods may have a relatively low dietary vitamin D intake even if the rest of their diet is otherwise healthy.
Both forms can raise 25(OH)D.
But comparative studies show that D3, or cholecalciferol, is on average more effective than D2 at raising total 25(OH)D and maintaining the achieved concentration over time.
That is why D3 is often the more practical choice for routine supplementation.
However, D2 is not an “ineffective” form. It also raises vitamin D status and can be used in clinical regimens.
For people avoiding animal-derived products, both D2 and plant-derived D3, for example from lichen, are available.
Vitamin D is fat-soluble, so the presence of dietary fat can improve absorption.
In one randomized study, peak D3 concentrations were roughly one-third higher when vitamin D was taken with a fat-containing meal compared with a fat-free meal. Other studies have also suggested an advantage to taking vitamin D with food.
But this does not mean vitamin D must be taken with avocado, cheese, or a specific amount of oil.
Some vitamin D is absorbed even without dietary fat, and modern oil-based formulations can reduce dependence on the exact composition of the meal. In one study of an oil-based D3 preparation, no significant difference was found between fasting and fed conditions.
The practical rule is simple: it is usually convenient to take vitamin D with a normal meal, but no special high-fat diet is required.
The relationship between vitamin D and magnesium has a real biochemical basis.
Magnesium is required for the activity of several enzymes involved in vitamin D metabolism, and experimental evidence shows that magnesium status can influence the conversion of vitamin D metabolites. A randomized trial has also shown that magnesium supplementation can influence 25(OH)D metabolism depending on baseline status.
But this is sometimes turned into another overly broad claim: “vitamin D should never be taken without magnesium.”
There is no such universal recommendation.
If a person gets enough magnesium from food and does not have magnesium deficiency, taking a separate magnesium supplement solely because they use a standard dose of D3 is not necessary.
The claim that vitamin D inevitably “depletes intracellular magnesium” and therefore causes muscle cramps is even less well supported. Muscle cramps have many possible causes.
The relationship between vitamins D and K is also biologically interesting. Vitamin D participates in calcium metabolism, while vitamin K is required for the carboxylation of osteocalcin and Matrix Gla Protein.
But the popular formula that “D3 raises calcium and K2 sends it into bones and keeps it out of the arteries” oversimplifies physiology.
There is no good evidence that every person taking a standard dose of D3 must automatically take K2 to prevent vascular calcification.
The decision to use K2 is a separate question and is especially relevant in people taking warfarin or other vitamin K antagonists.
For adults, including pregnant and breastfeeding individuals, the Food and Nutrition Board sets a Tolerable Upper Intake Level, or UL, of 4,000 IU — 100 mcg of vitamin D per day.
But this number is often misunderstood.
The UL is not a line beyond which the next unit of vitamin D suddenly becomes toxic.
It is the highest level of regular daily intake for the general population that is considered unlikely to cause adverse effects.
In clinical practice, physicians may temporarily use doses above 4,000 IU in selected conditions. So the statement that “anything above 4,000 IU is pharmacological and permitted only with laboratory-proven deficiency” is also too rigid.
A more accurate formulation is: long-term self-supplementation above the UL without a clear indication is undesirable, while higher therapeutic doses may be appropriate in specific clinical settings.
Vitamin D is fat-soluble. Unlike many water-soluble vitamins, excess vitamin D cannot simply be excreted indefinitely in the urine.
With prolonged excessive intake, intestinal calcium absorption increases and can lead to hypercalcemia and hypercalciuria.
It is this excess calcium, rather than some vague “vitamin toxicity,” that causes most of the clinical complications.
Symptoms may include:
In severe cases, kidney stones, nephrocalcinosis, kidney failure, soft-tissue calcification, cardiac arrhythmias, and even death can occur.
Vitamin D toxicity is almost always caused by excessive supplementation, not by normal food intake or ordinary sun exposure.
Here too, it is important to distinguish “undesirably high” from “toxic.”
NIH notes that overt toxicity with hypercalcemia usually occurs when 25(OH)D levels are above roughly 150 ng/mL, or 375 nmol/L.
That is considerably higher than 100 ng/mL.
So it is not accurate to say that every level above 100 ng/mL automatically represents clinical hypervitaminosis.
But 150 ng/mL should not be treated as a “safe target” either. The Food and Nutrition Board recommends avoiding concentrations above roughly 50–60 ng/mL, because there is no proven additional benefit at these levels and the probability of adverse effects may increase.
In other words:
50 ng/mL is not a target;
100 ng/mL is not an automatic diagnosis of toxicity, but it is an unnecessarily high range for most people;
around 150 ng/mL and above is the range often seen in clinical toxicity.
Vitamin D can be taken daily, weekly, or in some medical regimens at longer intervals.
But the idea that “the larger and less frequent the dose, the more convenient and effective it is” does not always hold.
In its 2024 guideline, the Endocrine Society prefers lower daily doses over large intermittent doses in adults over 50 when vitamin D supplementation is actually indicated.
This is an important practical shift. High-dose bolus regimens should not be used simply for convenience when there is no specific reason to do so.
There is no universal rule that exactly 5,000 IU requires testing while 4,000 IU can be taken indefinitely without monitoring.
But if someone plans to take doses above the established UL for a prolonged period, especially 5,000–10,000 IU or more, it is reasonable to first ask why that amount is necessary.
In the treatment of true deficiency, bone disease, malabsorption, or other medical conditions, laboratory monitoring may be justified. For self-directed “prevention” in an otherwise healthy person, such doses are usually unnecessary.
It is particularly risky to take 10,000–20,000 IU for years simply because one old blood test once showed “slightly low vitamin D.”
When treating confirmed significant deficiency, using high doses, managing malabsorption, or addressing another clinical problem, repeat testing after several months can indeed be useful.
But the rule that “everyone who starts vitamin D should repeat 25(OH)D after 8–12 weeks” is not universal.
The modern Endocrine Society guideline specifically recommends against routine follow-up testing in healthy people taking vitamin D when there is no other reason for laboratory monitoring.
A blood test should answer a clinical question. If the result will not change management, the value of testing is questionable.
Vitamin D may look like a harmless supplement, but drug interactions do exist.
Orlistat reduces fat absorption and can therefore reduce absorption of fat-soluble vitamin D.
Some glucocorticoids can negatively affect calcium balance and vitamin D metabolism.
Thiazide diuretics deserve particular attention. They reduce urinary calcium excretion, so combining them with high-dose vitamin D can potentially increase the risk of hypercalcemia, especially in people with kidney dysfunction or hyperparathyroidism.
So vitamin D bought over the counter and vitamin D used as part of a medical treatment plan are not always the same thing.
For a healthy adult without risk factors, there is no need to turn vitamin D into an endless laboratory project.
The basic approach is simple: consider age-based intake recommendations, diet, fortified foods, and lifestyle, and avoid megadoses solely to achieve an attractive number on a blood test.
When testing is genuinely indicated, 25(OH)D is the main marker, not calcitriol.
There is no need to automatically label every value below 30 ng/mL as disease. For most healthy people, NASEM considers 20 ng/mL and above sufficient for bone health, while the Endocrine Society no longer defines a universal target concentration for healthy populations.
When choosing a supplement, D3 is usually more practical than D2 and is on average more effective at raising 25(OH)D. Vitamin D can generally be taken with a normal meal.
There is no need to automatically add magnesium or K2 simply because someone takes vitamin D. Adequate magnesium, calcium, protein, and other nutrients still remain part of normal bone metabolism, but they should be considered on their own merits.
Long-term self-directed intake above 4,000 IU per day should not become routine. When genuinely high therapeutic doses are used, monitoring depends on the diagnosis, baseline status, comorbidities, and treatment plan.
With vitamin D, it is especially useful to distinguish four different concepts:
the physiological role of vitamin D;
the blood concentration of 25(OH)D;
the need to take a supplement;
the proven effect of supplementation on clinical outcomes.
Vitamin D is genuinely essential for calcium metabolism and bone mineralization. But that does not mean that the higher the 25(OH)D level, the stronger the bones.
Vitamin D receptors are genuinely present in many tissues. But that does not prove that high doses protect against every disease affecting those tissues.
A person can genuinely have severe deficiency that requires treatment. But that does not mean every healthy person should regularly test vitamin D and keep the result above 30, 40, or 50 ng/mL.
And finally, 4,000 IU is an important upper safety reference for long-term self-directed intake, not a magical border between a “vitamin” and a “drug.”
The central principle is simple: vitamin D is essential, and true deficiency has real consequences for bone and mineral metabolism. But modern evidence-based medicine increasingly rejects the pursuit of high 25(OH)D levels and routine megadose supplementation. The goal is adequacy, not the maximum possible number.
This material is provided for educational purposes only and does not replace medical consultation, diagnosis, or individualized treatment.
Serum 25-hydroxyvitamin D [25(OH)D] is the only standardized diagnostic marker. Measuring 1,25(OH)2D is uninformative for deficiency.
No. Doses of 10,000 IU daily carry a documented risk of hypercalcemia and renal damage over time. Safe daily maintenance is 600–2,000 IU.