Vitamin D: What It Does, and Why Magnesium Comes First

Vitamin D is not really a vitamin. It is a secosteroid prohormone your skin builds from cholesterol using UVB light, and your body then has to switch it on in two separate steps. Both steps are run by enzymes that require magnesium. That single fact is the part almost every vitamin D article leaves out, and it changes how you should think about the whole nutrient.

Why Is Vitamin D Called a Prohormone and Not a Vitamin?

A vitamin is something you must eat because your body cannot make it. Vitamin D fails that test. The name is a historical accident: rickets was described and treated with cod liver oil decades before anyone worked out that sunlight did the same job internally.

Structurally it is a secosteroid — a steroid whose B-ring has been snapped open, since UVB photons at 290–315 nm cleave the 9,10 carbon bond of 7-dehydrocholesterol. Functionally it is a prohormone: the end product, calcitriol, is a steroid hormone that binds a nuclear receptor and changes which genes a cell transcribes.

Its oldest and best-established job is calcium and phosphate handling. Per StatPearls, without adequate vitamin D you absorb roughly 10–15% of dietary calcium; with it, 30–40%. Phosphorus goes from about 60% to 80%. Calcitriol does this by switching on active transport in the duodenum — the TRPV6 channel, calbindin, and the PMCA1b pump. One caveat immediately: that is the deficient versus replete contrast, not a dose-response you can keep exploiting. The Linus Pauling Institute cites only about a 6% change in calcium absorption across a 25(OH)D range of 20 to 66 ng/mL.

Beyond calcium, the vitamin D receptor (VDR) appears in more than 30 tissues. Genome-wide ChIP-seq work finds roughly 2,000–3,000 VDR binding sites per cell type, a few hundred genes measurably responding, and only about 18% overlap between two different cell types. VDR runs a different program in each tissue.

How Does Your Body Turn Sunlight Into an Active Hormone?

Six steps, two organs, one photon.

1. 7-dehydrocholesterol → previtamin D3. In the deeper epidermis, UVB at 290–315 nm breaks open a cholesterol precursor. No enzyme — straight photochemistry.
2. Previtamin D3 → cholecalciferol (D3). Body heat rearranges it over hours. This is also the molecule in a D3 supplement.
3. Cholecalciferol → 25(OH)D (calcidiol). The liver adds a hydroxyl using CYP2R1, the 25-hydroxylase. Poorly regulated and substrate-driven — more input, more product — which is why 25(OH)D is what a blood test measures. It is the storage form, half-life about 15 days.
4. 25(OH)D → 1,25(OH)2D (calcitriol). The kidney's proximal tubule adds a second hydroxyl using CYP27B1, the 1-alpha-hydroxylase. Tightly controlled: parathyroid hormone up, FGF23 down. Immune cells and keratinocytes run local copies. Calcitriol's half-life is under four hours, which is why testing it tells you almost nothing.
5. Calcitriol → VDR → transcription. Calcitriol binds the vitamin D receptor, which pairs with the retinoid X receptor, lands on vitamin D response elements in DNA, and changes gene expression.
6. Shutdown. CYP24A1, the 24-hydroxylase, inactivates both 25(OH)D and calcitriol. Calcitriol induces its own destroyer.

In transit, about 85–90% of circulating 25(OH)D rides on vitamin D binding protein (DBP), 10–15% on albumin, under 1% free.

The sunlight step and the heat step happen without enzymes. Everything after that is enzymatic. That is where magnesium enters.

Why Does Magnesium Come First?

Because you cannot run any of those enzymes without it.

The canonical reference is Uwitonze AM and Razzaque MS, "Role of Magnesium in Vitamin D Activation and Function," Journal of the American Osteopathic Association, 2018;118(3):181–189. Its central sentence, exactly: "The enzymatic activity of both hepatic 25-hydroxylase and renal 1α-hydroxylase is a magnesium-dependent process." The review identifies three vitamin D-metabolising enzymes requiring magnesium, plus the transport protein.

Component	Name	Job in the pathway
25-hydroxylase	CYP2R1, liver	Cholecalciferol → 25(OH)D, the storage form
1-alpha-hydroxylase	CYP27B1, kidney	25(OH)D → calcitriol, the active hormone
24-hydroxylase	CYP24A1	Inactivates 25(OH)D and calcitriol
Binding protein	DBP / GC	Carries 85–90% of circulating 25(OH)D

All four are magnesium-dependent. Read that again and the interesting part appears: magnesium is required to build the active hormone and to break it down. It is not a gas pedal. It is closer to a thermostat.

That prediction has been tested. Dai Q, Zhu X, Manson JE and colleagues, American Journal of Clinical Nutrition, 2018;108(6):1249–1258, ran a randomized, double-blind, placebo-controlled trial nested in a colorectal cancer prevention study: 180 adults aged 40–85, 12 weeks, a personalized mean dose of 205.5 mg/day of elemental magnesium. At a baseline 25(OH)D of 30 ng/mL, magnesium increased 25(OH)D3 by 2.79 ng/mL (P = 0.03). At 50 ng/mL, it decreased it by 6.87 ng/mL (P = 0.002). The interaction was P = 0.001, surviving Bonferroni correction. Magnesium pushed people toward the middle from both directions — exactly what a cofactor for the activating and deactivating enzymes alike should do.

Population data agree. Deng X, Song Y, Manson JE and colleagues, BMC Medicine, 2013;11:187, analysed NHANES 2001–2006 and NHANES III and found magnesium intake significantly interacted with vitamin D intake for the risk of both vitamin D deficiency and insufficiency (P for interaction < 0.001 for each), concentrated in exactly the groups you would predict: winter-sampled participants, non-Hispanic Black participants, people with obesity, and women. The paper also reported that the inverse association between 25(OH)D and cardiovascular mortality held only above median magnesium intake (P for interaction = 0.03) — an observational finding in a general population, not a treatment effect.

Traffic runs both ways: calcitriol increases intestinal magnesium absorption through TRPM6 and TRPM7, so low vitamin D degrades magnesium status in turn. Two nutrients, each gating the other. We go deeper in our complete guide to what magnesium actually does and on the general pattern in nutrient cofactors and why nothing works alone.

What makes this practical rather than merely elegant: the NIH Office of Dietary Supplements reports that 48% of Americans consume less magnesium than the Estimated Average Requirement, and that magnesium is a cofactor in more than 300 enzyme systems. A shortfall that common, sitting upstream of a nutrient most people are also short of, is not hypothetical. It is the argument for building a daily formula around magnesium rather than treating it as garnish — which is why Current leads with 220 mg per stick instead of a token 20 mg.

Why Won't My Vitamin D Level Move?

The most common real-world vitamin D question, and almost nobody answers it properly. Work the list in order.

1. You retested too early. 25(OH)D has a half-life near 15 days, so steady state takes 8 to 12 weeks.
2. The dose is too small for your body. Gallagher and colleagues, Annals of Internal Medicine, 2012, found overweight participants landed 12.5 nmol/L lower and participants with obesity 17.5 nmol/L lower on the same dose. Some people need 1.5 to 3 times as much.
3. You're taking D2, not D3. Less efficient, faster-clearing, and it lowers your 25(OH)D3 fraction while raising the total.
4. You're taking it without fat. Vitamin D needs micellar incorporation; an empty-stomach capsule absorbs less reliably.
5. Your magnesium is low. The 25-hydroxylation step that produces the metabolite your lab measures is magnesium-dependent. If you are in the 48% below the EAR, you are asking a magnesium-dependent enzyme to do more work without more cofactor. This is the step nobody checks. For the symptom-level version, see the signs you might be low on magnesium.
6. You're dosing intermittently. One real-world regression analysis found daily dosing associated with a 30.69 nmol/L advantage over one-to-two-times-weekly dosing.
7. Your genotype. Variants in GC and CYP2R1 are the loci most consistently linked to baseline 25(OH)D and to response. Carlberg's personal vitamin D response index suggests roughly one in four people is a low responder at the gene-expression level.
8. You're already replete. Gallagher tested 400 through 4,800 IU/day and found 25(OH)D plateauing near 45 ng/mL. Past roughly 1,600–4,000 IU/day, more input stops producing proportional output.

Does Taking Vitamin D Deplete Magnesium?

There is a loop here, and it deserves careful language. Every hydroxylation, transport step and act of catabolism in the pathway consumes magnesium-dependent enzymatic capacity. A large dose of vitamin D increases the throughput demanded of CYP2R1, CYP27B1 and CYP24A1 at once. With a comfortable magnesium reserve, unremarkable. Sitting below the EAR already, the extra demand pulls on a reserve that was thin to begin with. Calcitriol also drives calcium into circulation, and calcium and magnesium compete at shared absorption and reabsorption routes.

So: low magnesium slows vitamin D activation, people take more vitamin D, magnesium demand rises, the shortfall deepens. Mechanistically coherent and repeatedly raised in review articles. It is also not a proposition any randomized trial has tested head-on — nobody has given people high-dose vitamin D and measured the change in magnesium status against a placebo arm. Treat it as a well-reasoned mechanism, not a demonstrated effect. The practical takeaway is modest: if you supplement vitamin D daily, keep magnesium intake in reasonable shape at the same time.

What's the Difference Between D2 and D3 — and Is Lichen D3 the Same?

D2 is ergocalciferol, from UV-irradiated fungal ergosterol. D3 is cholecalciferol, the molecule your skin makes. The gap is real but more conditional than most copy implies. The reference is Tripkovic and colleagues, American Journal of Clinical Nutrition, 2012;95(6):1357–64, pooling seven randomized trials:

Regimen	Studies	D3 advantage over D2	Significant?
All pooled	7	+15.23 nmol/L (95% CI 6.12–24.34)	Yes, P = 0.001
Bolus dosing	3	+34.10 nmol/L (16.38–51.83)	Yes, P = 0.0002
Daily dosing	6	+4.83 nmol/L (−0.98 to 10.64)	No, P = 0.10

The 2012 advantage was driven almost entirely by large intermittent doses; at daily doses it did not reach significance. Most brand copy skips that. The stronger evidence at a daily physiologic dose came later: Tripkovic 2017, AJCN 106(2):481–490, a 12-week randomized, double-blind, placebo-controlled food-fortification trial in wintertime Britain — 335 healthy South Asian and white European women at 15 mcg (600 IU) per day across five arms. Both D3 vehicles significantly beat both D2 vehicles, with incremental differences of 15.3 to 42.9 nmol/L (all P < 0.0003). A 2023 meta-analysis in Advances in Nutrition was directionally consistent, with BMI as an effect modifier.

Why? Jones and colleagues, JCEM, 2014, measured half-lives in 36 people: 25(OH)D2 13.9 ± 2.6 days versus 15.1 ± 3.1 for 25(OH)D3 (P = 0.001). Binding protein holds 25(OH)D3 more tightly, and D2 is a poorer CYP2R1 substrate. One under-discussed consequence: supplementing D2 lowers circulating 25(OH)D3 even as it raises the total. That does not make D2 useless. It makes D3 the better default.

Vegan D3 from lichen: what the evidence does and does not show

Most supplemental D3 comes from lanolin, which comes from sheep's wool. The vegan route is lichen — commercially reindeer lichen (Cladonia rangiferina) and relatives, which accumulate 7-dehydrocholesterol. Manufacture is not exotic: extract the sterol fraction, irradiate with controlled UVB, allow thermal isomerisation, purify. The same sequence your skin performs, run in a reactor. Since Current uses lichen-derived D3, we will be scrupulous.

Established: the end product is cholecalciferol, C27H44O — molecularly identical to lanolin D3 and to what your skin makes, with no structural, stereochemical or isotopic difference. Because CYP2R1 acts on the cholecalciferol molecule itself, there is no plausible biochemical mechanism by which the origin of the starting sterol would change what happens next. And because lichen D3 is D3, the D3-versus-D2 evidence above applies to it.

Not established: we could not locate a single published, peer-reviewed human bioequivalence trial comparing lichen-derived against lanolin-derived D3 on serum 25(OH)D. Plenty of brands assert bioequivalence; none we checked cite a head-to-head human study. So the defensible claim is molecular identity — not "clinically proven equivalent," and certainly not "better absorbed because it's plant-based," which is meaningless. The real differentiators between D3 sources are less romantic: potency accuracy, carrier, stability, overage control.

What Else Does Vitamin D Need to Work?

Magnesium is the strongest case by a distance. The others run from textbook mechanism with unproven payoff down to interesting-but-thin.

Vitamin K2 and where the calcium goes. Vitamin K is the cofactor for gamma-glutamyl carboxylase, which converts glutamate residues to Gla residues. Two of those proteins matter: osteocalcin, a VDR target gene that needs carboxylation before it can bind hydroxyapatite in bone, and matrix Gla protein, the most potent known endogenous inhibitor of vascular calcification, which also requires carboxylation. The clinical evidence is mixed. A 2023 meta-analysis in Frontiers in Nutrition (14 trials, 1,533 patients) found reduced coronary artery calcification progression (mean difference −17.37, 95% CI −34.18 to −0.56, P = 0.04) and substantially lower dephospho-uncarboxylated MGP (−243.31, P = 0.0001), while hedging its own conclusion. Against that, AVADEC (Circulation, 2022) gave MK-7 at 720 mcg/day plus vitamin D to men with aortic valve calcification and did not slow progression. The mechanism is real and the biomarker moves; the hard endpoint has not followed. And the fear version — vitamin D calcifies your arteries unless you take K2 — has no randomized support: VITAL ran 2,000 IU/day for 5.3 years in 25,871 people without K2, with no excess cardiovascular events and no excess hypercalcemia.

Calcium. The most direct partner, since regulating calcium is what vitamin D exists to do. In a 2024 network meta-analysis in BMC Geriatrics (35 trials, 58,937 participants), vitamin D plus calcium performed better on falls (RR 0.67, 95% CI 0.50–0.86) than vitamin D alone (RR 0.85).

Vitamin A and the RXR partner. VDR does not bind DNA productively alone; it heterodimerizes with the retinoid X receptor, whose ligand is 9-cis-retinoic acid, a vitamin A metabolite, and RXR contributes materially to transactivation. But this cuts both ways, and anyone selling vitamin A "to help your vitamin D" has read half the literature: because RXR is shared, high retinoic acid can compete VDR away from it, and retinoic acid receptors can bind vitamin D response elements directly. Observational data have linked high preformed retinol intake to lower bone mineral density on exactly that hypothesis, though the association remains contested. Beta-carotene does not carry the same concern.

Zinc. A precise structural fact and a tempting overreach. The VDR's DNA-binding domain contains two Cys4-type zinc finger motifs, each coordinating one zinc ion; without zinc the fingers do not fold and VDR cannot grip DNA. What does not follow is that taking zinc improves vitamin D signaling — structural zinc comes from a tightly buffered intracellular pool, and that leap is not established in humans. Only about 15% of US adults fall below the zinc EAR from food, against 48% for magnesium. We ranked the full set in our complete guide to trace minerals.

Boron. The weakest link, included because it is usually oversold. USDA metabolic-ward work by Nielsen and colleagues found low dietary boron appeared to raise urinary calcium and magnesium excretion and lower plasma calcium and vitamin D. Median US intake is about 0.87–1.35 mg/day; the upper limit is 20 mg/day. The NIH Office of Dietary Supplements states plainly that more research is needed to determine whether boron supplementation affects bone health in humans.

How Many Americans Are Actually Low on Vitamin D?

It depends entirely on which line you draw, which is the whole fight in miniature. Using IOM cut-points, the NIH Office of Dietary Supplements reports from NHANES 2011–2014 that 5% of Americans aged 1 and over are at risk of deficiency and 18% at risk of inadequacy. A pooled analysis of NHANES 2001–2018 in Frontiers in Nutrition gives the fuller distribution:

Category	Serum 25(OH)D	Share of US population
Severe deficiency	Below 10 ng/mL	2.6%
Moderate deficiency	10–20 ng/mL	22.0%
Insufficiency	20–30 ng/mL	40.9%
Sufficiency	Above 30 ng/mL	34.5%

About 25% of Americans sit below 20 ng/mL and about 65% below 30. The subgroup spread is wider than the average suggests:

• Skin pigmentation. Deficiency below 12 ng/mL runs 17.5% in non-Hispanic Black Americans, 7.6% in non-Hispanic Asian, 5.9% in Hispanic and 2.1% in non-Hispanic white Americans. Melanin is a competitive UVB chromophore — it absorbs the same photons 7-dehydrocholesterol needs. Reported inhibition factors range from about 1.3-fold to about 8-fold, so the popular "dark skin needs 3 to 5 times longer in the sun" is a plausible midpoint, not a measured constant.
• Season. Severe deficiency roughly doubles from summer (1.6%) to winter (4.1%).
• Age. The highest US deficiency rates sit in the 20–29 band, not the elderly. Indoor work and indoor leisure explain more than biology.
• BMI. People with a BMI of 30 or above have consistently lower 25(OH)D. The best explanation is not "trapped in fat" but volumetric dilution — Drincic and colleagues argued in Obesity (2012) that vitamin D distributes through total body volume, so the same input yields a lower concentration in a larger body.
• Latitude, counterintuitively. A pooled global analysis of 7.9 million participants found the 20–40°N band had the highest severe-deficiency prevalence (23.1%), above 40–60°N. Clothing, sun avoidance and fortification policy outrank raw latitude.

Can You Get Enough Vitamin D From the Sun?

In summer, at moderate latitude, with skin exposed: yes, easily. In winter above roughly 35–37°: no, not at all, however long you stand outside.

The foundational study is Webb, Kline and Holick, JCEM, 1988: no measurable cutaneous previtamin D3 synthesis in Boston (42.2°N) from November through February, none in Edmonton (52°N) from October through March, and year-round production in Los Angeles and San Juan. The physics fits in a sentence: UVB at 290–315 nm is preferentially absorbed by stratospheric ozone, and atmospheric path length scales with solar zenith angle, so once the sun drops below roughly 30° above the horizon effectively no vitamin D-effective UVB reaches the ground. Hence the field rule: if your shadow is longer than you are tall, you are making little or none. A 2022 analysis in Nutrients put the mean European "vitamin D winter" at 126 days, from 4 days in Nicosia (35°N) to 215 in Reykjavik, with cities below 40°N averaging around 60.

The Denver problem: altitude versus latitude

We are a Colorado company, so this one is personal. Denver sits at 39.7°N and 5,280 feet (1,609 m), which puts it in the band with a vitamin D winter of roughly two to four months, approximately November through February. The altitude genuinely helps, more than most people realise. The Hong Kong Observatory's standard figure is about 12% more broadband UV per 1,000 m of elevation — but the review by Young and colleagues in JBMR Plus (2021) reports that a kilometre of height increases irradiance at 300 nm by about 24%, roughly double the broadband gradient, because the shortest wavelengths are most attenuated by the atmospheric column you just climbed above. The same review cites in vitro work finding roughly fourfold higher previtamin D3 production at Everest base camp (5,300 m) than in Agra, India (170 m). Applied to Denver: roughly 35–40% more UVB than sea level at the same latitude and sun angle.

Now the counterpoint, which is the important half. Altitude increases UVB intensity at a given solar angle. It does not change the solar angle. At noon in December at 39.7°N the sun sits about 27° above the horizon, and the atmospheric path is long enough that a 35–40% intensity bonus does not reliably lift D-UVB above the synthesis threshold. Altitude makes summer sun more efficient — and more dangerous; Colorado has among the highest melanoma rates in the country. It barely helps in the months when it might, because that is precisely when Coloradans are outside in jackets and gloves. Better summer, same winter. "We get 300 days of sunshine" is not a vitamin D strategy from November to February.

Sunscreen, glass, and age

Sunscreen. The scary version does not survive contact with data. A 2017 study by Libon and colleagues found SPF 50+ under narrowband UVB reduced cutaneous vitamin D production by 75.7% to 92.5% — but reduced circulating 25(OH)D3 by only 7.7% to 13.2%. People apply roughly a quarter to half the tested 2 mg/cm² thickness, and the previtamin D photoreaction saturates at very low, sub-sunburn UV doses. "SPF 30 blocks 97% of UVB, so it blocks 97% of your vitamin D" is a true first clause bolted to a false conclusion.

Glass. Ordinary window glass, and especially a laminated windshield, transmits UVA but blocks essentially all UVB below about 320 nm. Since synthesis needs 290–315 nm, you make no meaningful vitamin D through a window, in a car, or in a sunroom.

Age. The classic 1985 study by MacLaughlin and Holick reported aging could more than halve the skin's capacity to produce previtamin D3 — the origin of "a 70-year-old makes 75% less than a 20-year-old." A 2024 study in Nutrients compared 11 adults aged 18–40 with 10 aged 65–89 and found no significant difference in baseline skin 7-dehydrocholesterol (0.22 versus 0.25 mcg/mg) and no significant difference in the serum vitamin D3 rise after UV exposure, attributing the original result to methodological limitations. Older adults do have lower status on average, but the evidence increasingly points to less time outdoors, more clothing coverage and lower dietary intake rather than intrinsically incapable skin.

What Did the Big Vitamin D Trials Actually Find?

This is where credibility is won or lost. The largest randomized trials, in tens of thousands of people over years, were null on their primary endpoints. Everything below is a research finding about supplementation in a study population, reported as research.

Trial	Design	Primary result
VITAL (Manson, NEJM 2019)	25,871 US adults, 2,000 IU/day D3, median 5.3 y	Cancer HR 0.96 (0.88–1.06); major cardiovascular events HR 0.97 (0.85–1.12). Both null. No excess hypercalcemia.
VITAL — fractures (LeBoff, NEJM 2022)	1,991 incident fractures	No reduction in total, non-vertebral or hip fractures.
VITAL — autoimmune (Hahn, BMJ 2022)	Same cohort, ancillary endpoint	HR 0.78 (0.61–0.99), P = 0.05 — the one positive signal.
VITAL — depression (Okereke, JAMA 2020)	~18,353 participants	Null for incident depression and mood scores.
D-Health (Neale, Lancet D&E 2022)	21,315 Australians 60–84, 60,000 IU monthly, 5 y	No reduction in all-cause, cancer or cardiovascular mortality.
DO-HEALTH (Bischoff-Ferrari, JAMA 2020)	2,157 adults, mean age 74.9, 2,000 IU/day, 3 y	Null on all six primary endpoints.
D2d (Pittas, NEJM 2019)	2,423 adults with prediabetes, 4,000 IU/day	No significant reduction in progression to type 2 diabetes.
Jolliffe (Lancet D&E 2021)	43 randomized trials, respiratory infection	OR 0.92 (0.86–0.99) — significant but modest.

The autoimmune result deserves more than a row. Hahn and colleagues reported a roughly 22% lower rate of incident autoimmune disease over 5.3 years at 2,000 IU/day, with vitamin D plus omega-3 at HR 0.69 (0.49–0.96). It is the most interesting positive vitamin D result of the last decade — and it is an ancillary endpoint in a trial whose primary endpoints were null, with a P value sitting exactly at 0.05, and a 2024 follow-up found the effect attenuated two years after supplements stopped. That is the profile of a result that sometimes fails replication. Promising, not proven.

The respiratory story changed too. Martineau's 2017 individual-participant meta-analysis (25 trials, 11,321 participants) found an adjusted odds ratio of 0.88, with a striking gradient: 0.81 for daily or weekly dosing versus 0.97 for bolus, and 0.30 in participants starting below 10 ng/mL. Jolliffe's 2021 update put it at OR 0.92. Then the 2025 update, at 46 trials and 64,086 participants, reported OR 0.94 (0.88–1.00), P = 0.057 — no longer statistically significant overall. Subgroups held: children, daily dosing, the 400–1,000 IU/day range. The general-population claim did not.

Why daily dosing beats big occasional doses

The cleanest dosing evidence in the field, and it contains a trap. For raising your blood level, bolus dosing is not inferior: a 2023 Bayesian network meta-analysis in Frontiers in Nutrition (116 trials, 11,376 participants) found no statistically significant difference between daily and intermittent regimens at equivalent total dose. Anyone claiming daily raises your number better is overstating.

For what happens to people, daily wins consistently. Martineau: 0.81 daily or weekly versus 0.97 bolus. Jolliffe 2025: 0.84 daily, nothing significant weekly or monthly. The 2024 falls network meta-analysis: RR 0.78 for daily, no preventive effect for intermittent — and a 2023 analysis in Osteoporosis International found a borderline harm signal for falls with intermittent high doses (RR 1.06, 0.99–1.15, P = 0.051). The leading explanations: large boluses spike calcitriol and strongly induce CYP24A1, accelerating catabolism into a post-peak trough, and the local CYP27B1 systems in immune cells depend on steady substrate rather than peaks. The 2024 Endocrine Society guideline recommends daily over intermittent high doses accordingly. A daily stick is, structurally, a daily-dosing habit — a formulation argument rather than a health claim, and the reasoning behind treating minerals as a daily foundation rather than an occasional intervention.

Is 20 ng/mL Enough, or Do You Need 30?

Two respected bodies drew the line in two places, and then one of them moved it again.

Body	Sufficiency threshold	Recommended intake
IOM / NASEM (2011)	20 ng/mL (50 nmol/L)	RDA 600 IU (1–70), 800 IU (71+); upper limit 4,000 IU
Endocrine Society (2011)	30 ng/mL (75 nmol/L)	Often 1,500–2,000 IU/day for at-risk adults
Endocrine Society (2024)	Declines to define a threshold	See below

The IOM's position is that 20 ng/mL covers the skeletal requirements of 97.5% of the population — and, a point almost always missed, that because an RDA is set against a population distribution, the median requirement corresponds to only about 16 ng/mL. The Endocrine Society's 30 ng/mL came from parathyroid hormone suppression and calcium absorption studies plus observational data on non-skeletal outcomes.

Then came Demay, Pittas, Bikle and colleagues, "Vitamin D for the Prevention of Disease: An Endocrine Society Clinical Practice Guideline," JCEM, 2024;109(8):1907–1947. It abandoned the 30 ng/mL target and the deficiency/insufficiency/sufficiency nomenclature, stating there is no clear evidence defining an optimal target level for disease prevention. It suggested against routine 25(OH)D testing in every population it considered, including older adults, pregnant people, and people with obesity or darker skin. It suggested against empiric supplementation above the DRI in healthy adults under 75, while recommending it for ages 1–18, adults 75 and over, pregnancy, and high-risk prediabetes. And it recommended daily rather than intermittent high dosing.

This was the same society that in 2011 told clinicians to screen at-risk patients and target 30 ng/mL, driving a decade-long explosion in testing. Reversing it landed hard, and it has drawn substantive criticism: it recommends supplementation without specifying doses; by superseding the 2011 document it orphaned populations it does not address, such as malabsorption, kidney disease and post-bariatric patients; and it says almost nothing about sunlight. The honest reading is that both things are true. The guideline is about disease prevention in the general population, not about treating deficiency. It does not say vitamin D is unimportant. It says routine testing in healthy people does not change outcomes, and no universal target number is supported by randomized evidence.

How Much Vitamin D Should You Take Daily?

For most adults not being treated for a diagnosed deficiency: a daily dose between the 600 IU RDA and the 4,000 IU upper limit, taken every day with food.

Heaney's rule of thumb is that serum 25(OH)D rises about 1 ng/mL per 100 IU/day of D3 at steady state. So 1,000 IU/day works out to roughly a 10 ng/mL increase over 8 to 12 weeks — moving someone at 18 ng/mL to around 28. Someone at 32 moves less reliably, because the curve flattens: Gallagher found 25(OH)D plateauing around 45 ng/mL even at 4,800 IU/day, and found 800 IU/day was enough to lift 97.5% of participants above 20 ng/mL, the empirical basis for the IOM's RDA.

Why 25 mcg / 1,000 IU is a sensible daily maintenance dose. It sits above the 600 IU RDA for adults under 71 and the 800 IU RDA for 71 and over, so it more than covers the requirement. It sits at 25% of the 4,000 IU upper limit, leaving headroom for the vitamin D you also get from food, sun and anything else you take. It is daily rather than bolus, the regimen the outcome data favour. And it is maintenance, not therapy — repleting a diagnosed deficiency is a conversation with a clinician, not a job for a drink mix.

Why megadosing is not better. The dose-response curve flattens; past roughly 1,600–4,000 IU/day you raise disposal flux more than tissue concentration, since calcitriol induces its own catabolic enzyme. The outcome data do not reward higher doses — the falls network meta-analysis found 800–1,000 IU/day performed best, with both doses under 500 IU/day and doses above 1,000 IU/day faring worse. And toxicity is real: a 2018 review of case reports in Nutrients found documented toxicity involved intakes near 50,000 IU/day, presenting 25(OH)D of 150 to 1,220 ng/mL and serum calcium of 11.1 to 23.1 mg/dL, from manufacturing errors of up to 4,000-fold, self-overdosing, and prescribed megadoses. Reaching 150 ng/mL from 1,000 IU/day is not physiologically plausible. The 4,000 IU limit is a precautionary ceiling, not a toxicity threshold — but the region far above it is genuinely hazardous.

Timing. With a meal containing fat. Beyond that, morning versus evening does not meaningfully change 25(OH)D. If you take magnesium in the evening — and the magnesium and sleep research is a legitimate reason people do — that is fine. Consistency beats clock time.

To compare how daily formulas actually handle vitamin D and magnesium, we ran the label numbers in our comparison of the leading electrolyte powders, covered the category-wide shortfall in the magnesium gap, and went head-to-head in Current versus LMNT and Current versus Liquid I.V. Most of the category carries no vitamin D at all, which is worth knowing before assuming your hydration mix covers it; the full Current panel is published in the same detail.

Where the Evidence Is Weak

• The depletion loop. Mechanistically sound, never tested head-on in a randomized trial measuring magnesium status after high-dose vitamin D.
• Lichen D3 bioequivalence. Molecular identity is certain. A published head-to-head human trial against lanolin D3 does not appear to exist.
• Vitamin K2 and arteries. The carboxylation biochemistry is textbook and the biomarker moves reliably. Hard endpoints have not followed; AVADEC was null.
• Zinc and VDR function. The zinc fingers are structurally certain. The nutritional inference is not established in humans.
• Boron. Small USDA metabolic-ward studies, and the NIH's own fact sheet saying more research is needed.
• Muscle. Severe deficiency causes a real proximal myopathy that reverses with repletion. But VDR expression in mature skeletal muscle is genuinely disputed, with early positive results later attributed to poor antibody specificity.
• Who responds. Carlberg's response index suggests roughly one in four people is a molecular low responder, and gene-expression response often does not track serum 25(OH)D. It is the best explanation on offer for why the megatrials keep coming up null — and it is not actionable, since no validated assay exists.

Myths Worth Retiring

• "Vitamin D regulates 3,000 genes." Genome-wide data support roughly 2,000–3,000 binding sites per cell type and a few hundred responsive genes; the citable range is 100 to 1,250.
• "You can't use vitamin D without magnesium." An overstatement of the very paper this article is built on, copied endlessly from a 2018 press release. The randomized trial shows a bidirectional normalising effect, not an on-off gate. Magnesium matters. It is not a key in a lock.
• "Vitamin D prevents cancer, heart disease, fractures and depression." VITAL, D-Health and DO-HEALTH: null, null, null.
• "Vitamin D prevents colds and flu." The 2025 update across 46 trials reported OR 0.94, P = 0.057 — not significant overall. What survives is daily, low-dose, in children, and in the profoundly deficient.
• "SPF 30 blocks 97% of UVB, so it blocks your vitamin D." Measured effect on circulating 25(OH)D3: 7.7% to 13.2%.
• "Everyone should get tested." The 2024 Endocrine Society guideline suggests against routine testing in every population it considered.
• "More vitamin D means more calcium absorbed." Absorption plateaus — about a 6% change across 20 to 66 ng/mL.
• "Older adults make 75% less vitamin D from sunlight." Challenged by a 2024 study finding no difference in skin 7-dehydrocholesterol or UV response between adults 18–40 and 65–89.
• "Take a big monthly dose — same thing." Same blood level, worse outcomes, borderline harm signal for falls.
• "D2 is useless." It raises 25(OH)D. It is just less efficient and shorter-lived, and it lowers the 25(OH)D3 fraction.
• "Colorado's altitude and sunshine mean you're covered." Altitude raises UVB roughly 24% per kilometre at 300 nm. It does not change December's solar zenith angle at 39.7°N.
• "Vitamin D is toxic above 4,000 IU." That is a precautionary ceiling. Documented toxicity involves 25(OH)D above 150 ng/mL, typically from 50,000+ IU/day or manufacturing errors.

The thread through all of these runs through the difference between electrolytes, minerals and trace minerals and the eight B vitamins and their forms too: the specific, named mechanism is usually more interesting than the exaggerated version, and it holds up better. If you are reaching for vitamin D as the single answer to feeling flat, the fuller list of reasons people run tired is worth reading first.

Frequently Asked Questions

What does vitamin D actually do?

Its best-established job is calcium and phosphate absorption — roughly 10–15% of dietary calcium absorbed without adequate vitamin D, 30–40% with it. Once activated to calcitriol it binds the vitamin D receptor, which pairs with the retinoid X receptor and changes gene transcription in more than 30 tissues, with a different gene set in each.

Should you take magnesium with vitamin D?

There is a good mechanistic reason to have adequate magnesium if you supplement vitamin D daily: the enzymes that convert vitamin D to its storage form and then to its active hormone are both magnesium-dependent, as are the enzyme that breaks it down and the protein that carries it. Roughly 48% of Americans fall below the Estimated Average Requirement for magnesium. What you should not believe is that vitamin D is useless without magnesium — the randomized evidence shows a normalising effect, not a gate.

Why is my vitamin D still low after taking supplements?

In order: you retested before 8 to 12 weeks; the dose is too small for your body size, since people with obesity often need 1.5 to 3 times as much; you are taking D2 rather than D3, or taking it without food containing fat; your magnesium intake is low; you are dosing weekly rather than daily; or variation in GC and CYP2R1 is limiting your response.

What is the difference between D2 and D3?

D2 (ergocalciferol) comes from irradiated fungal ergosterol; D3 (cholecalciferol) is the form your skin makes. Tripkovic's 2012 meta-analysis found D3 raised 25(OH)D by 15.23 nmol/L more overall, though that advantage was driven by bolus dosing and was not significant in the daily subgroup; the 2017 Tripkovic trial in 335 women at 600 IU/day showed a clear D3 advantage at a daily dose.

Is vegan D3 from lichen the same as regular D3?

Chemically, yes — lichen-derived D3 is cholecalciferol, molecularly identical to lanolin D3 and to what your skin produces, and the activation enzymes act on that molecule regardless of the starting sterol's origin. What does not exist, as far as we can find, is a published head-to-head human trial. "Chemically identical" is defensible; "clinically proven bioequivalent" is not.

Do you need K2 with vitamin D3?

The mechanism is real: vitamin K activates osteocalcin and matrix Gla protein, which direct calcium into bone and inhibit vascular calcification. The clinical evidence is mixed — trials improve the dp-ucMGP biomarker, but AVADEC found MK-7 at 720 mcg/day did not slow aortic valve calcification. The fear version has no randomized support; VITAL ran 2,000 IU/day for 5.3 years without K2 and found no excess cardiovascular events or hypercalcemia.

How much vitamin D should I take daily?

The RDA is 600 IU for ages 1–70 and 800 IU for 71 and over, with an upper limit of 4,000 IU/day for ages 9 and up. A daily 1,000 IU sits above the RDA and at a quarter of the upper limit, raising serum 25(OH)D by roughly 10 ng/mL over 8 to 12 weeks.

Can you take too much vitamin D?

Yes, but the region of real risk sits far above normal supplemental doses. Documented toxicity in a 2018 review of case reports involved intakes near 50,000 IU/day, with 25(OH)D of 150 to 1,220 ng/mL. VITAL used 2,000 IU/day in 25,871 people for over five years with no excess hypercalcemia or adverse events.

Can you get enough vitamin D from the sun?

In summer, at moderate latitudes, with skin exposed — usually yes. In winter above roughly 35–37° latitude, no: the sun is too low for UVB at 290–315 nm to reach the ground in useful amounts. If your shadow is longer than you are tall, you are making little or none. Glass blocks UVB, so a sunny window does not count.

What is the best time of day to take vitamin D?

Whenever you will actually take it, with a meal containing some fat, since vitamin D is fat-soluble. There is no convincing evidence that morning versus evening changes serum 25(OH)D — consistency matters far more, because daily dosing outperforms intermittent dosing on outcomes.

Sources

1. Uwitonze AM, Razzaque MS. "Role of Magnesium in Vitamin D Activation and Function." Journal of the American Osteopathic Association, 2018;118(3):181–189. Link
2. Dai Q, Zhu X, Manson JE, et al. "Magnesium status and supplementation influence vitamin D status and metabolism: results from a randomized trial." American Journal of Clinical Nutrition, 2018;108(6):1249–1258. Link
3. Deng X, Song Y, Manson JE, et al. "Magnesium, vitamin D status and mortality: results from US NHANES 2001–2006 and NHANES III." BMC Medicine, 2013;11:187. Link
4. NIH Office of Dietary Supplements. "Vitamin D — Health Professional Fact Sheet." Link
5. NIH Office of Dietary Supplements. "Magnesium — Health Professional Fact Sheet." Link
6. NIH Office of Dietary Supplements. "Boron — Health Professional Fact Sheet." Link
7. Linus Pauling Institute Micronutrient Information Center. "Vitamin D." Oregon State University. Link
8. StatPearls. "Vitamin D." NCBI Bookshelf NBK441912. Link
9. Demay MB, Pittas AG, Bikle DD, et al. "Vitamin D for the Prevention of Disease: An Endocrine Society Clinical Practice Guideline." Journal of Clinical Endocrinology & Metabolism, 2024;109(8):1907–1947. Link
10. Tripkovic L, Lambert H, Hart K, et al. "Comparison of vitamin D2 and vitamin D3 supplementation in raising serum 25-hydroxyvitamin D status: a systematic review and meta-analysis." American Journal of Clinical Nutrition, 2012;95(6):1357–1364. Link
11. Tripkovic L, Wilson LR, Hart K, et al. "Daily supplementation with 15 µg vitamin D2 compared with vitamin D3 to increase wintertime 25-hydroxyvitamin D status in healthy South Asian and white European women: a 12-wk randomized, placebo-controlled food-fortification trial." American Journal of Clinical Nutrition, 2017;106(2):481–490. Link
12. Jones KS, Assar S, Harnpanich D, et al. "25(OH)D2 half-life is shorter than 25(OH)D3 half-life and is influenced by DBP concentration and genotype." Journal of Clinical Endocrinology & Metabolism, 2014;99(9):3373–3381. Link
13. Manson JE, Cook NR, Lee I-M, et al. "Vitamin D Supplements and Prevention of Cancer and Cardiovascular Disease" (VITAL). New England Journal of Medicine, 2019;380:33–44. Link
14. Hahn J, Cook NR, Alexander EK, et al. "Vitamin D and marine omega 3 fatty acid supplementation and incident autoimmune disease: VITAL randomized controlled trial." BMJ, 2022;376:e066452. Link
15. LeBoff MS, Chou SH, Ratliff KA, et al. "Supplemental Vitamin D and Incident Fractures in Midlife and Older Adults." New England Journal of Medicine, 2022;387:299–309. Link
16. Okereke OI, Reynolds CF, Mischoulon D, et al. "Effect of Long-term Vitamin D3 Supplementation vs Placebo on Risk of Depression or Clinically Relevant Depressive Symptoms" (VITAL-DEP). JAMA, 2020;324(5):471–480. Link
17. Neale RE, Baxter C, Romero BD, et al. "The D-Health Trial: a randomised controlled trial of the effect of vitamin D on mortality." Lancet Diabetes & Endocrinology, 2022;10(2):120–128. Link
18. Bischoff-Ferrari HA, Vellas B, Rizzoli R, et al. "Effect of Vitamin D Supplementation, Omega-3 Fatty Acid Supplementation, or a Strength-Training Exercise Program on Clinical Outcomes in Older Adults" (DO-HEALTH). JAMA, 2020;324(18):1855–1868. Link
19. Pittas AG, Dawson-Hughes B, Sheehan P, et al. "Vitamin D Supplementation and Prevention of Type 2 Diabetes" (D2d). New England Journal of Medicine, 2019;381:520–530. Link
20. Martineau AR, Jolliffe DA, Hooper RL, et al. "Vitamin D supplementation to prevent acute respiratory tract infections: systematic review and meta-analysis of individual participant data." BMJ, 2017;356:i6583 / NIHR HTA 2019;23(2). Link
21. Jolliffe DA, Camargo CA, Sluyter JD, et al. "Vitamin D supplementation to prevent acute respiratory infections: a systematic review and meta-analysis of aggregate data from randomised controlled trials." Lancet Diabetes & Endocrinology, 2021;9(5):276–292. Link
22. Jolliffe DA, Camargo CA, et al. "Vitamin D supplementation to prevent acute respiratory infections: updated systematic review and stratified meta-analysis." Lancet Diabetes & Endocrinology, 2025. Link
23. Gallagher JC, Sai A, Templin T, Smith L. "Dose Response to Vitamin D Supplementation in Postmenopausal Women." Annals of Internal Medicine, 2012;156(6):425–437. Link
24. "Effect of vitamin D, calcium, or combined supplementation on fall prevention: a systematic review and updated network meta-analysis." BMC Geriatrics, 2024;24. Link
25. "Effects of intermittent or single high-dose vitamin D supplementation on falls and fractures." Osteoporosis International, 2023;34. Link
26. "Comparison of daily and intermittent vitamin D supplementation: a Bayesian network meta-analysis." Frontiers in Nutrition, 2023;10:1168115. Link
27. "Prevalence, trend and predictors of vitamin D deficiency in the US population, 2001–2018." Frontiers in Nutrition, 2022;9:965376. Link
28. "Global and regional prevalence of vitamin D deficiency, 2000–2022 (7.9 million participants)." Frontiers in Nutrition, 2023;10:1070808. Link
29. Drincic AT, Armas LAG, Van Diest EE, Heaney RP. "Volumetric Dilution, Rather Than Sequestration, Best Explains the Low Vitamin D Status of Obesity." Obesity, 2012;20(7):1444–1448. Link
30. Webb AR, Kline L, Holick MF. "Influence of season and latitude on the cutaneous synthesis of vitamin D3." Journal of Clinical Endocrinology & Metabolism, 1988;67(2):373–378. Link
31. Young AR, Morgan KA, Harrison GI, et al. "Physical Determinants of Vitamin D Photosynthesis: A Review." JBMR Plus, 2021;5(1):e10460. Link
32. "Comprehensive Analysis of Seasonal and Geographical Variation in UVB Radiation Relevant to Vitamin D Synthesis." Nutrients, 2022;14(23):5189. Link
33. Hong Kong Observatory. "Ultraviolet radiation at high altitude." Link
34. Libon F, Nikkels-Tassoudji N, Cavalier E, et al. "Sunscreens block cutaneous vitamin D production with only a minimal effect on circulating 25-hydroxyvitamin D." 2017. Link
35. "Comparative Study of Healthy Older and Younger Adults Shows They Have the Same Skin Concentration of 7-Dehydrocholesterol and Response to UVR." Nutrients, 2024;16(8):1147. Link
36. MacLaughlin J, Holick MF. "Aging decreases the capacity of human skin to produce vitamin D3." Journal of Clinical Investigation, 1985;76(4):1536–1538. Link
37. "Development of Vitamin D Toxicity from Overcorrection of Vitamin D Deficiency: A Review of Case Reports." Nutrients, 2018;10(8):953. Link
38. "Vitamin K supplementation and vascular calcification: a systematic review and meta-analysis of randomized controlled trials." Frontiers in Nutrition, 2023;10:1115069. Link
39. Diederichsen ACP, Lindholt JS, Möller S, et al. "Vitamin K2 and D in Patients With Aortic Valve Calcification" (AVADEC). Circulation, 2022;145(18):1387–1397. Link
40. "Relationship of Structure and Function of DNA-Binding Domain in Vitamin D Receptor." Molecules, 2015;20(7):12389–12409. Link
41. "Retinoid X Receptor Is a Nonsilent Major Contributor to Vitamin D Receptor-Mediated Transcriptional Activation." Molecular Endocrinology, 2003;17(11):2320–2328. Link
42. Nielsen FH, Hunt CD, Mullen LM, Hunt JR. "Effect of dietary boron on mineral, estrogen, and testosterone metabolism in postmenopausal women." FASEB Journal, 1987;1(5):394–397. Link
43. Carlberg C, Haq A. "The concept of the personal vitamin D response index." Journal of Steroid Biochemistry and Molecular Biology, 2018;175:12–18. Link
44. Wang Y, Zhu J, DeLuca HF. "Where is the vitamin D receptor?" Archives of Biochemistry and Biophysics, 2012;523(1):123–133. Link
45. Ramagopalan SV, Heger A, Berlanga AJ, et al. "A ChIP-seq defined genome-wide map of vitamin D receptor binding." Genome Research, 2010;20(10):1352–1360. Link
46. Reider CA, Chung R-Y, Devarshi PP, et al. "Inadequacy of Immune Health Nutrients: Intakes in US Adults, the 2005–2016 NHANES." Nutrients, 2020;12(6):1735. Link
47. "The 2024 Endocrine Society Guideline on Vitamin D: Comprehensive Summary and Critical Appraisal." Nutrients, 2026;18(9):1472. Link

These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease. This article is for educational purposes and is not medical advice. Talk to your doctor before starting any supplement, especially if you are pregnant, nursing, have kidney disease, or take prescription medication.