In one pass Your skin makes its own vitamin D when sunlight hits it.
Educational content, not medical advice — consult a clinician.
UVB arrives Only UVB sunlight (280–315 nm) can penetrate the upper epidermis and reach the basal layer.
D3 reaches the liver D3 binds DBP in the blood and is escorted all the way to the smooth endoplasmic reticulum of the hepatocyte.
25(OH)D enters proximal tubule After 25(OH)D bound to DBP is filtered at the glomerulus, proximal tubule cells recapture the whole complex through the megalin and cubilin receptors.
Calcitriol enters enterocyte nucleus 1,25(OH)₂D (calcitriol) flows through the gut, entering duodenal epithelial cells (the most active site of calcium absorption).
Macrophage meets pathogen Macrophages patrol the body — when they meet bacteria / virus / fungi, they recognise pathogen-associated molecular patterns (PAMP) via Toll-like receptors (TLR).
Your skin makes its own vitamin D when sunlight hits it. Ultraviolet B light (UVB) in sunlight strikes a molecule in the skin called 7-dehydrocholesterol and breaks open one of its rings, turning it into a vitamin D precursor, which then converts into by itself. No enzyme is involved; the energy comes from the light.
Three conditions have to be met. The wavelength must be right: only UVB works, and sunlight through glass makes none. Darker skin has more melanin, which blocks more of the UVB. And the sun has to be high enough: in winter at high latitudes the sun sits so low that the atmosphere absorbs almost all of the UVB.
So ten minutes of sun can produce very different amounts of vitamin D in different people and different seasons. No single number of minutes fits everyone.
A CLOSER LOOK
From UVB to D3 in the skin
UVB starts the ring opening; skin temperature then drives rearrangement. Neither step needs an enzyme.
1Light supplies energy
A UVB photon opens a ring in 7-DHC, forming previtamin D3.
2Skin temperature drives rearrangement
Previtamin D3 spontaneously rearranges into D3 at skin temperature.
3From skin into blood
D3 enters dermal capillaries, binds vitamin D-binding protein (DBP), and is carried to the liver.
Illustration for understanding; not to scale. Saved figures include explanations and sources.
Mechanism · The photochemical step in skin
The basal layer of the skin holds a molecule called 7-dehydrocholesterol (7-DHC). It is an intermediate on the pathway that makes cholesterol, and the body also puts it to use as the precursor of vitamin D. When UVB photons (280-315 nm) strike the B ring of 7-DHC, a photochemical reaction breaks the C9-C10 σ bond, the ring opens, and previtamin D3 forms.
This step needs no enzyme; it runs purely on the energy of the light. But there are three hidden conditions:
The wavelength must be right: UVA (315-400 nm) does not work, visible light does not work, and glass blocks UVB.Skin pigment gets in the way: melanin in darker skin absorbs UVB, so under the same sun synthesis is roughly 50% less efficient.The sun must be high enough: above 35°N in winter, the sun is so low that the atmosphere absorbs almost all of the UVB. That is why people at high latitudes generally have lower vitamin D levels in their blood in winter. Over the following minutes to hours, previtamin D3 converts by itself at skin temperature into (cholecalciferol), the starting point of the body's own vitamin D production.
In practice · How much sun is enough
"How much sun a day is enough?" has no single answer. How much you actually make depends on four variables:
Skin pigment (Fitzpatrick types I-VI): at the same UV intensity, types I-II (fair, Northern European skin) need 10-15 minutes, while types V-VI (deep brown or black skin) need 30-60 minutes.Area exposed: face, arms and lower legs make up about 25% of the body surface and are the usual baseline; exposing the whole body is 4-6 times as efficient as exposing just the face and hands.Height of the sun: UVB is most effective when the sun is > 50° above the horizon; at winter noon above 35°N the sun is so low that the real UVB is close to 0.Season, latitude, altitude and air pollution. By the rough estimate in Holick's review, 10-15 minutes of spring or summer sun on 25% of the skin lets fair skin make 1000-3000 of ; in winter in northern cities, that figure approaches 0.
A few practical points:
Around midday in spring and summer, 10-20 minutes of ordinary outdoor activity is enough; there is no need to bake in strong sun.In winter, in rainy spells or when you seldom go out, you usually need food plus a D3 supplement; a common amount is 1000-2000 IU/day.People with dark skin who live at high latitudes find it hard to make enough from sunlight at almost any time of year, and a supplement is often needed. Under test conditions, sunscreen of SPF 30 or more blocks 95-98% of UVB, and strict long-term sun protection clearly reduces vitamin D synthesis. That is no reason to skip sunscreen (skin cancer costs far more than low vitamin D), but it does mean that people who rarely go outside and always wear sunscreen should pay more attention to vitamin D from food and supplements.
The other route for vitamin D is food: smaller amounts, but steady.
(cholecalciferol) comes from animals: fatty fish (wild salmon, about 600 per 100 g; see Salmon · see Sardines), egg yolk (see Eggs), fish-liver oil and fortified milk (see Milk).D2 (ergocalciferol) comes from fungi: mushrooms exposed to ultraviolet light (UVB turns their ergosterol into D2; see Mushrooms). Both dissolve in fat, and both need fat to be absorbed. In the small intestine they are wrapped, together with the fat from food, into tiny particles (mixed micelles) that the gut cells take up; the vitamin is then packed into chylomicrons, the particles that carry fat, and travels through the lymph rather than the portal vein. So a vitamin D supplement taken on an empty stomach is absorbed less well, and taking it with a meal that contains fat works better. People who have had their gallbladder taken out, or who have fat malabsorption or Crohn's disease, run short of vitamin D more easily.
D3 raises in the blood more effectively than D2, so supplements usually use D3.
Clinical · How much to take, and how often
A vitamin D dose cannot be chosen from one blood-test result alone. NIH ODS cites < 30 nmol/L as associated with deficiency, 30–50 as potentially inadequate, and ≥ 50 as enough for most people. Above 75 is not a universal target. A clinician should assess testing and supplementation in context; these ranges are not a self-dosing chart.
Large, infrequent doses are not necessarily better. Sanders 2010 JAMA studied 2256 community-dwelling women aged ≥70, using an annual oral dose of 500,000 . Compared with placebo, the incidence rate ratio (IRR) was 1.15 for falls and 1.26 for fractures. A post hoc analysis found a larger increase in falls during the first 3 months after dosing. This finding concerns that population and regimen, not every intermittent schedule.
An upper limit is not a target. The adult tolerable upper intake level () is 4000 IU/day from food and supplements combined. It is not a recommended personal dose or a guarantee of safety. Excess supplementation can cause hypercalcemia and, in severe cases, kidney damage. Taking it for less than a year does not ensure safety. A clinician should decide and monitor therapeutic high doses.
Nutrients working together does not make a supplement bundle necessary. Magnesium participates in D metabolism, and calcium supplies bone mineral. Start with diet and actual need. supports osteocalcin and matrix Gla protein, but this mechanism does not establish that K2 supplements prevent osteoporosis or arterial calcification. People taking warfarin should not add K2 or suddenly change vitamin K intake without medical advice.
Chapter 3
The liver turns it into storage form
Vitamin D made in the skin or taken in from food is still only raw material. It first goes to the liver, which adds one hydroxyl group and turns it into 25-hydroxyvitamin D, written . That is the number a blood test measures: when a doctor judges whether your vitamin D is enough, this is what they read.
Once in the blood, rides on vitamin D-binding protein () to the liver, where two enzymes in the liver cells, CYP2R1 (the main one) and CYP27A1 (the backup), add the hydroxyl group.
This step is barely held back by feedback: the more vitamin D you eat or make in the sun, the more 25(OH)D rises in the blood. Its half-life in the blood is about 2-3 weeks, so one test reflects the past few weeks, not whether you were in the sun today. Most of the vitamin D circulating in the blood is in this form.
A CLOSER LOOK
Two changes before vitamin D becomes active
The liver converts D3 to 25(OH)D; the kidney is the main source of active calcitriol in the circulation.
1The first change in the liver
Liver CYP2R1 adds a hydroxyl group to D3, producing 25(OH)D.
2The circulating form
25(OH)D travels in blood bound to DBP and is the main marker used to assess vitamin D status.
3The second change in the kidney
Kidney CYP27B1 adds another hydroxyl group to form calcitriol, which binds VDR to regulate genes.
Illustration for understanding; not to scale. Saved figures include explanations and sources.
Clinical · What a vitamin D blood test measures
A vitamin D blood test measures : not , and not the later active form, 1,25(OH)₂D.
This site reads the result by the NIH ODS bands (the 2011 report from the US Institute of Medicine, IOM, agrees):
< 30 nmol/L (< 12 ng/mL): deficiency, directly linked to rickets and osteomalacia30-50 nmol/L: possibly inadequate50-125 nmol/L (20-50 ng/mL): enough for most people> 125 nmol/L: no extra health benefit, and possible adverse effects; > 375 nmol/L is a toxic level that can cause hypercalcemia Some professional societies favor a higher target. That is the view of some societies, not the NIH ODS bands.
So the single 25(OH)D number on a lab report is enough; there is no need to test "active D". Active D is tightly controlled by parathyroid hormone and other signals, and in people who are short of vitamin D it often reads normal or even high, so it says nothing about stores.
Myth · Three claims about the liver, doses and D2
"A bad liver means you can't convert vitamin D": in most liver diseases the 25-hydroxylation pathway still works, because CYP2R1 and CYP27A1 back each other up and rarely fail together. The step that more often causes trouble is the kidney's (1α-hydroxylation).
"If the test is low, take 50000 in one go": vitamin D dissolves in fat, can be stored and has a long half-life, so small regular daily doses are steadier than one big hit. At 1000-4000 IU a day (within the ), taken with a meal that contains fat, it usually takes 8-12 weeks before a retest shows the new level; aim for the NIH ODS bands, with no need to chase 75. A mega-dose like 500,000 IU once a year actually led to more falls and fractures in women over 70 in the Sanders 2010 trial.
"D2 is the same as ": it is not. D3 raises blood more and keeps it up longer. Choose D3 (cholecalciferol) supplements, not D2 (ergocalciferol).
Vitamin D is only truly activated in the kidney. In the cells of the kidney tubules, the enzyme 1α-hydroxylase (CYP27B1) adds the last hydroxyl group and makes calcitriol, also written 1,25(OH)₂D. It can enter the cell nucleus and regulate genes, and it works like a steroid hormone rather than merely a vitamin.
This step is tightly controlled. When blood calcium is low, parathyroid hormone () switches it on; when blood phosphate is high, FGF23, released by bone cells, switches it off; blood calcium and phosphate themselves also feed back on it; and magnesium is a cofactor the enzyme cannot do without. Active D lasts only a few hours in the blood; the body does not let it pile up.
Besides the kidney, many tissues, including immune cells and skin, can activate a little vitamin D on the spot, for local use only.
Mechanism · How the kidney turns D into a hormone
In the kidney tubule cells, CYP27B1 (1α-hydroxylase) adds one more hydroxyl group to and produces 1,25-dihydroxyvitamin D, or 1,25(OH)₂D, also called calcitriol. This is the active hormone that actually does the work.
The step is finely regulated: parathyroid hormone () turns it up when blood calcium is low; FGF23, released by bone cells, turns it down when blood phosphate is high; blood calcium and phosphate feed back on it directly; and magnesium is a cofactor of CYP27B1, so the step stalls when magnesium is short.
Its half-life in the blood is only 4-15 hours (compared with weeks for 25(OH)D); the body does not want it circulating freely. It works by binding the vitamin D receptor () in the cell nucleus. VDR is a transcription factor: once it binds active D, it directly regulates which genes are switched on. By one estimate, about 3% of human genes are regulated by it, directly or indirectly.
So vitamin D is not just "a vitamin"; it works more like a steroid hormone. But that does not mean "low vitamin D leaves you a little low in everything". A widely spread receptor only shows that vitamin D may play a part in many places; how big its effect is in each one has to be tested in people, one place at a time.
Mechanism · Activation outside the kidney
"Only the kidney can activate vitamin D" is an old view that was overturned over the past 20 years. We now know that CYP27B1 is present in many tissues, which can turn into active 1,25(OH)₂D locally and use it on the spot (autocrine and paracrine action).
Tissues outside the kidney known to carry CYP27B1 include:
Macrophages and dendritic cells, one of the routes by which vitamin D takes part in immune regulationKeratinocytes in the skinBreast tissue, the prostate, the lining of the colon, and the placentaThe parathyroid glands (forming a negative feedback loop)The beta cells of the pancreatic islets This has a few clinical consequences:
It explains why vitamin D can affect many seemingly unrelated organs: it is activated and acts on the spot in those tissues, not relying entirely on the kidney's central control.In sarcoidosis and some lymphomas, the macrophages' CYP27B1 runs out of control, and so much active D is made that it spills into the blood and raises blood calcium (hypercalcemia). That is a cause of high calcium other than "vitamin D toxicity".In people with chronic kidney disease (), the kidney's CYP27B1 falls, whole-body 1,25(OH)₂D runs low, and secondary hyperparathyroidism and bone disease follow. That is why doctors use active vitamin D analogues (such as calcitriol and paricalcitol) rather than ordinary . Calcipotriol, the cream used for psoriasis, is itself an active vitamin D analogue. It acts directly on the in skin cells and does not need to be activated.
An analogy: the kidney is the central mint, and the money it prints enters the blood and circulates through the whole body. Many tissues also have small local presses whose money is spent only locally and usually does not change the level in the blood. Sarcoidosis is what happens when a local press runs out of control and prints money into the blood.
Active vitamin D's best-known job is getting calcium into the body. It does three things for bone and calcium:
1. It opens the door in the small intestine. In the cells lining the duodenum, it switches on a whole active-absorption system for calcium (a calcium channel, a calcium-binding protein and a calcium pump) that moves calcium from the gut into the blood. When vitamin D is severely short, this active route shrinks and only passive diffusion is left, so only about 10% of the calcium you eat gets absorbed. 2. It cuts losses at the kidney. In the kidney tubules, it pulls calcium back from the urine into the blood. 3. It works on bone in two directions. Normally it helps bone-building cells mineralize new bone; when blood calcium falls very low, it does the opposite and mobilizes bone-dissolving cells to borrow calcium from bone, protecting blood calcium first and treating bone as the reserve of last resort.
So when vitamin D is severely short, most of the extra calcium you take cannot get in. Magnesium and vitamin K also take part in bone metabolism, but sharing a process does not mean everyone needs all of these as supplements. Nor should 's mechanism be read as a guarantee that calcium will not be deposited in the arteries.
A CLOSER LOOK
Vitamin D signals; calcium crosses
Active D acts through VDR to regulate calcium-absorption machinery. The calcium route shown runs from the gut lumen to blood.
1The cell receives the D signal
Active D binds VDR to regulate proteins involved in intestinal calcium absorption. Vitamin D does not itself carry calcium.
2Active and passive routes
The central route shows active transport through the enterocyte; calcium can also be absorbed passively between cells.
3Calcium enters blood
The route shown moves calcium from the enterocyte into blood, completing absorption from the gut lumen.
Illustration for understanding; not to scale. Saved figures include explanations and sources.
Clinical · Rickets and osteomalacia
First, a little molecular detail. The active calcium-absorption system in the small intestine consists of the TRPV6 calcium channel, the calcium-binding protein calbindin-D9k and the PMCA1b calcium pump, all switched on in the nucleus by active vitamin D acting through the . In the kidney tubules the equivalents are TRPV5 and calbindin-D28k. When blood calcium falls very low, calcium is borrowed from bone through the pathway.
Severe vitamin D deficiency is not a "subclinical concept". It is a disease that happened throughout history and still happens today.
In children it appears as rickets: bone fails to mineralize, and the long bones soften and bend. Signs include bow legs or knock knees, a pigeon chest, bead-like bumps along the ribs, a soft skull, and teeth that come in late and decay easily. In 19th-century London during the industrial revolution, smoke and crowded city housing meant children almost never saw the sun, and rickets was very common among city children. That is where the vitamin D story begins: McCollum named vitamin D in 1922, and from the 1930s milk fortified with vitamin D sharply reduced rickets. Cases still occur today, and they are not rare. Most are infants with several risk factors at once: darker skin, a mother who covers most of her skin for religious reasons, a family that has moved to a northern latitude, and long breastfeeding without a vitamin D supplement.
In adults the equivalent is osteomalacia: the bone matrix is poorly mineralized, and a scan () may not show it. It causes bone pain, muscle weakness and stress fractures. Unexplained bone pain all over, weakness in the thighs and upper arms, and repeated fractures from minor injuries, together with a severely low (< 25 nmol/L), should bring this diagnosis to mind. It is easily mistaken for fibromyalgia, depression or "just getting old".
For prevention, several public-health practices are standard:
Infants start 400 of a day soon after birth, including babies who are fully or partly breastfed and babies who are not yet drinking much formula; the American Academy of Pediatrics (AAP) strongly recommends this.Children and teenagers take 400-600 IU a day, alongside diet and outdoor play.In pregnancy, 600 IU a day (fortified foods plus a supplement).Food fortification policies (fortified milk, fortified cereals) are the main public-health tool. One reminder: calcium supplements cannot replace vitamin D. When vitamin D is severely short, the gut's active absorption route cannot open, and most of the calcium you add cannot get in. When both are low, both need replacing, and a doctor should decide how to combine them.
Almost every immune cell carries a vitamin D receptor (). In cell and animal studies, active vitamin D can induce antimicrobial peptides, push inflammatory T cells toward a "regulatory" type, and damp down excessive inflammatory signals. A plausible mechanism, however, is not proof that taking vitamin D prevents infection or cures everything.
Human trials give a much narrower answer. Martineau 2017 pooled individual data from 25 randomized trials: vitamin D cut the overall risk of acute respiratory infection by only about 12%, and the benefit was concentrated in people who took it daily or weekly and were severely deficient to begin with (< 25 nmol/L). People whose baseline was above 25 nmol/L gained much less, and large doses given at long intervals showed no protection.
So taking vitamin D is mainly about correcting a deficiency, not about "boosting" immunity.
Mechanism · Vitamin D receptors on immune cells
Macrophages, T cells, B cells, dendritic cells: almost every immune cell carries the . What active vitamin D does on the immune side is fairly clear in cell and animal studies:
It induces antimicrobial peptides such as cathelicidin (LL-37) and defensins, which can act directly against bacteria, fungi and viruses. In macrophage experiments this helps the cells fight the tuberculosis bacterium. Some people use it to explain the sunlight treatment of old sanatoriums, but in human trials of vitamin D added to tuberculosis treatment, the main outcome did not improve (Martineau 2011, covered on the page about claims and evidence).It steers T-cell development, pushing the inflammatory Th1 and Th17 types toward the regulatory Treg type. In epidemiological studies, the rates of autoimmune diseases such as multiple sclerosis, type 1 diabetes and psoriasis are associated with vitamin D levels (an observed association, which cannot show that vitamin D is the cause).It restrains excessive inflammation, turning down the pathway and reducing the release of and . The clinical evidence in people needs to be read in layers. Martineau 2017 (BMJ) is an individual-participant-data that pooled 25 randomized trials and 11,321 people:
Across all participants, vitamin D lowered the risk of acute respiratory infection by about 12% (adjusted 0.88).Among people taking it daily or weekly, those with a baseline < 25 nmol/L benefited most, about 70% (odds ratio 0.30); those with a baseline ≥ 25 benefited less (odds ratio 0.75).Regimens that gave a large dose at long intervals showed no protection. So vitamin D helps at the level of correcting a deficiency. There is no evidence for "topping up" once you already have enough.
Evidence · What vitamin D can and cannot prevent
Vitamin D has been one of the most over-promoted nutrients of the past decade, and it is also one of the few with a genuine immune mechanism, so the evidence needs to be sorted layer by layer.
Backed by randomized trials or clear causation:
Vitamin D reduces respiratory infections in people who start out deficient. Martineau 2017 (BMJ, 25 randomized trials, n=11,321): the largest reduction (about 70%) was in people with a baseline < 25 nmol/L who took it daily or weekly; across all participants the reduction was only about 12%.Preventing rickets in children and osteomalacia in adults: the causal link is clear.Secondary hyperparathyroidism caused by vitamin D deficiency: correcting the deficiency is itself part of the treatment. Limited or mixed evidence:
Add-on treatment for tuberculosis, downgraded: in the Martineau 2011 (Lancet) randomized trial, the main outcome was negative: median time to a negative sputum culture was 36.0 versus 43.5 days, adjusted 1.39 (95% 0.90-2.16), p = 0.14. Only the subgroup with the TaqI tt genotype converted faster.Multiple sclerosis (MS) and vitamin D deficiency are linked in epidemiology: MS is more common at high latitudes, and adequate vitamin D may go with slower progression, but the treatment effect in randomized trials is still weak.Preventing type 2 diabetes: the D2d trial (NEJM 2019, n=2,423) enrolled adults with prediabetes. At 4000 a day, their risk of progressing to diabetes was about 12% lower, but the difference was not statistically significant and may have been chance. Insufficient evidence or negative results:
Treating or preventing COVID-19: several randomized trials disagree, and there is no evidence so far that vitamin D prevents COVID-19 in ordinary people who are not deficient.Cancer prevention: the VITAL trial (NEJM 2019, n=25,871, median 5.3 years) enrolled ordinary middle-aged and older Americans (men from age 50, women from age 55). With 2000 IU of a day versus placebo, the rate of invasive cancer did not differ. Cancer deaths, a secondary outcome, were slightly lower, but not significantly so, and count only as a lead.Preventing cardiovascular disease: in the same VITAL trial, major cardiovascular events did not differ either.Preventing falls in older adults, downgraded: early saw a benefit, but NIH ODS now holds that vitamin D does not reduce falls in community-dwelling adults aged 65 and over; it matters only when a confirmed deficiency is being corrected (covered in detail in the chapter on muscle).Treating depression: in the D-Vitaal randomized trial (de Koning 2019, AJCN), 1200 IU a day did not improve depression in older adults."Boosting immunity" has no measurable definition; it is a marketing phrase. Putting it together: correcting deficiency (< 30 nmol/L) brings real health benefits; 30-50 nmol/L is possibly inadequate; above 50 is enough for most people; once you have enough (50-125 nmol/L), more vitamin D shows no extra benefit on most hard outcomes; and above 125 nmol/L there is no benefit and there may be adverse effects. "Vitamin D is a miracle molecule" is marketing. "Correcting a deficiency is a basic health investment" is the reality.
Low vitamin D can leave the legs weak and make falls more likely, especially in older people. Muscle contraction relies on the calcium that vitamin D manages, and skeletal muscle cells carry vitamin D receptors () themselves.
In people who are severely deficient, vitamin D improves muscle weakness. But vitamin D prevents falls only when it corrects a deficiency that really exists. For community-dwelling older adults who are not short of vitamin D, large trials and guidelines since 2018 agree that taking more does not reduce falls, and the US Preventive Services Task Force (USPSTF) goes further and recommends against taking vitamin D to prevent falls. What does have evidence is exercise: strength and balance training.
Using high-dose vitamin D to improve athletic performance has no evidence behind it either.
Clinical · Can vitamin D prevent falls?
In the United States, falls are the leading cause of injury-related illness and death among older adults (USPSTF 2018). Vitamin D was once thought to have some of the strongest clinical evidence here, but large randomized trials after 2018 revised that conclusion substantially. It now has to be told as two separate cases: real deficiency and already enough.
In theory, vitamin D might affect falls through several routes. Most of this comes from animal, cell and observational studies and has not been confirmed step by step in people:
Type II fast-twitch fibers (the ones for explosive power) waste first when vitamin D is short, and they decide whether you can recover quickly when you suddenly lose your balance.The cerebellum and the vestibular nuclei contain , and they are part of the balance system.Maintaining the myelin of the nerves that sense body position may partly depend on vitamin D.Adequate vitamin D is associated with slightly better cognition (an observed association), and alertness affects fall risk. The timeline of the evidence matters:
Bischoff-Ferrari 2009 (BMJ): 700-1000 of a day reduced falls by about 19% ( 0.81), and trials that raised blood to ≥ 60 nmol/L reduced falls by about 23% (RR 0.77); below either threshold there was no reduction. These were the core data behind guideline recommendations more than ten years ago.USPSTF 2018 (JAMA): after re-reviewing the evidence, it concluded that vitamin D has no benefit for preventing falls in community-dwelling adults aged 65 and over and recommended against taking it for that purpose (for people without known osteoporosis or vitamin D deficiency); the same statement recommends exercise.VITAL falls analysis (LeBoff 2020): in 25,871 middle-aged and older Americans taking 2000 IU a day for a median of 5.3 years, the share with two or more falls, and the share whose falls led to a doctor or hospital visit, did not differ. Participants in the trial started at an average 25(OH)D of about 77 nmol/L, so most were not deficient, and people with a baseline below 50 showed no difference either.D-Health (Waterhouse 2023, Lancet Diabetes Endocrinol): in 21,315 Australians aged 60-84 taking 60,000 IU a month for up to 5 years, fracture risk did not differ, and there was no sign that a large monthly dose increased fractures.The position of NIH ODS's 2024 fact sheet: vitamin D supplements do not reduce the number of falls or fall injuries in community-dwelling adults aged 65 and over.Very large doses are worse: in Sanders 2010 (JAMA), women over 70 given 500,000 IU once a year had more falls, not fewer. One guess is that the sudden peak in blood levels disturbs calcium signaling between nerve and muscle, but this has not been measured directly. The current consensus can be summed up this way: vitamin D for falls makes sense only when a test confirms deficiency (< 30 nmol/L), or a level in the possibly inadequate range (30-50 nmol/L) that needs correcting. It is not "all older adults should take vitamin D to prevent falls". For community-dwelling older adults who already have enough, more vitamin D does not prevent falls, and the opportunity cost is the strength training, balance training and medication review that did not happen.
In practice:
Not every adult over 65 needs routine 25(OH)D testing; for people with osteoporosis, a history of falls, malabsorption or very little sun, a doctor decides whether to test. By the NIH ODS bands, ≥ 50 nmol/L is enough for most people.For confirmed deficiency: 800-2000 IU of D3 a day, 1000-1200 mg of calcium (food plus supplements), and strength training (whose evidence is stronger than vitamin D's).For people at high risk of falling (a fall in the past 1 year, an unsteady walk, complicated medicines): a full assessment, physical therapy, changes at home, an eye check and a medication review. Vitamin D is not the core intervention. Remember that falls have many causes acting together: eyesight, medicines, muscle strength, balance, the home, getting up at night, low blood pressure and dementia all play a part. In people who already have enough vitamin D, strength and balance training plus a medication review are far better value than more vitamin D.
"High-dose vitamin D boosts athletic performance" has weak evidence. The Owens 2018 review puts it this way: vitamin D is not an ergogenic aid but a physiological requirement, and whether to take it depends on your status. Data on muscle function in athletes are mixed, and the strength gain people usually quote comes mainly from adults with 25(OH)D below about 25 nmol/L (most of them older), not from athletes who already have enough.
National Institutes of Health, Office of Dietary Supplements. (2024). Vitamin D — Fact Sheet for Health Professionals. Fact sheet (updated June 27, 2025; Wayback snapshot 20 September 2026): 25(OH)D below 30 nmol/L (12 ng/mL) is associated with deficiency, 30 to below 50 nmol/L is generally considered inadequate, 50 nmol/L (20 ng/mL) or more is sufficient for most people, and above 125 nmol/L (50 ng/mL) can be associated with adverse effects; RDA 600 IU (15 mcg) to age 70 and 800 IU (20 mcg) above 70; adult UL 4,000 IU (100 mcg); older people and people with darker skin (more melanin) make less vitamin D from sunlight (fact sheet). ods.od.nih.gov/factsheets/VitaminD-HealthProfessional
Bikle, D. D. (2014). Vitamin D metabolism, mechanism of action, and clinical applications. Chemistry & Biology, 21(3), 319–329. 10.1016/j.chembiol.2013.12.016
Holick, M. F. (2007). Vitamin D deficiency. The New England Journal of Medicine, 357(3), 266–281. 10.1056/NEJMra070553
EFSA Panel on Dietetic Products, Nutrition and Allergies. (2016). Dietary reference values for vitamin D. EFSA Journal, 14(10), 4547. 10.2903/j.efsa.2016.4547
Uwitonze, A. M., & Razzaque, M. S. (2018). Role of magnesium in vitamin D activation and function. Journal of the American Osteopathic Association, 118(3), 181–189. 10.7556/jaoa.2018.037
Institute of Medicine. (2011). Dietary Reference Intakes for Calcium and Vitamin D. National Academies Press. www.ncbi.nlm.nih.gov/books/NBK56070
National Institutes of Health, Office of Dietary Supplements. (2021). Vitamin K — Fact Sheet for Health Professionals. Fact sheet (updated March 29, 2021; Wayback snapshot 19 September 2026): free phylloquinone is about 80% absorbed, but absorption from food is much lower - the body absorbs only 4% to 17% as much from spinach as from a tablet; Table 2: natto, 3 ounces, 850 mcg (as MK-7); egg, hard boiled, 1 large, 4 mcg (as MK-4) (fact sheet). ods.od.nih.gov/factsheets/VitaminK-HealthProfessional
Martineau, A. R., et al. (2017). Vitamin D supplementation to prevent acute respiratory tract infections: systematic review and meta-analysis. BMJ, 356, i6583. Individual-participant-data meta-analysis of 25 double-blind RCTs (11,321 participants aged 0-95; IPD for 10,933). Overall adjusted OR for acute respiratory infection 0.88 (0.81-0.96). Daily or weekly dosing without boluses OR 0.81 (0.72-0.91); one or more bolus doses OR 0.97 (0.86-1.10), no benefit. Among daily/weekly recipients: baseline 25(OH)D below 25 nmol/L OR 0.30 (0.17-0.53); 25 nmol/L or above OR 0.75 (0.60-0.95). Serious adverse events not increased (OR 0.98) (abstract, PMID 28202713). 10.1136/bmj.i6583
Martineau, A. R., Timms, P. M., Bothamley, G. H., Hanifa, Y., Islam, K., Claxton, A. P., et al. (2011). High-dose vitamin D3 during intensive-phase antimicrobial treatment of pulmonary tuberculosis: a double-blind randomised controlled trial. The Lancet, 377(9761), 242-250. PRIMARY ENDPOINT NULL: median time to sputum culture conversion 36.0 vs 43.5 days, adjusted HR 1.39 (0.90-2.16), p=0.14. The only positive signal is a TaqI tt genotype subgroup. Do not cite it as evidence that vitamin D speeds conversion. 10.1016/S0140-6736(10)61889-2
Girgis, C. M., Clifton-Bligh, R. J., Hamrick, M. W., Holick, M. F., & Gunton, J. E. (2013). The roles of vitamin D in skeletal muscle: form, function, and metabolism. Endocrine Reviews, 34(1), 33–83. 10.1210/er.2012-1012
Bischoff-Ferrari, H. A., Dawson-Hughes, B., Staehelin, H. B., Orav, J. E., Stuck, A. E., Theiler, R., et al. (2009). Fall prevention with supplemental and active forms of vitamin D: a meta-analysis of randomised controlled trials. BMJ, 339(oct01 1), b3692-b3692. Two separate results: supplemental vitamin D at 700-1000 IU/day cut falls 19% (RR 0.81), and achieved serum 25(OH)D >= 60 nmol/L (24 ng/mL) cut them 23% (RR 0.77). Below either threshold there was no reduction. The story welded the 19% onto the serum threshold and wrote that threshold as 30 ng/mL. 10.1136/bmj.b3692
de Koning, E. J., Lips, P., Penninx, B. W. J., Elders, P. J., Heijboer, A. C., den Heijer, M., et al. (2019). Vitamin D supplementation for the prevention of depression and poor physical function in older persons: the D-Vitaal study, a randomized clinical trial. The American Journal of Clinical Nutrition, 110(5), 1119-1130. The D-Vitaal trial. There is no "Aronson 2021" vitamin-D depression trial; this is the paper that name pointed at. 10.1093/ajcn/nqz141
Owens, D. J., Allison, R., Close, G. L. (2018). Vitamin D and the Athlete: Current Perspectives and New Challenges. Sports Medicine, 48(S1), 3-16. Its position is that vitamin D is a physiological requirement, NOT an ergogenic aid. Athlete muscle-function data are mixed; the strength benefit people quote is seen in adults with 25(OH)D below about 25 nmol/L, largely older adults, not in replete athletes. 10.1007/s40279-017-0841-9
Sanders, K. M., Stuart, A. L., Williamson, E. J., Simpson, J. A., Kotowicz, M. A., Young, D., et al. (2010). Annual High-Dose Oral Vitamin D and Falls and Fractures in Older Women. JAMA, 303(18), 1815. A single ANNUAL oral dose of 500,000 IU in women 70+ (n=2,256) increased falls (IRR 1.15) and fractures (IRR 1.26), with risk highest in the first 3 months. The story attached this result to a monthly 50,000 IU regimen, which is a different dose and a different interval. 10.1001/jama.2010.594