Vitamin D is not a substitute for any vaccine — Deplatform Disease

Edward Nirenberg
57 min readFeb 17, 2021


Because of the length of this blog post, I don’t really expect anyone to read it in its entirety (though of course anyone who wants to is welcome to), but if you want more than the short version, I recommend the section”The Controversies and Misconceptions of Optimal Levels and Supplementation,” in particular.

The Short Version: There is an incredible amount of hype regarding the importance of vitamin D in our health, and while adequate levels are undeniably important, this concept turns out to be not at all simple. Vitamin D deficiency is not nearly as prevalent as has often been claimed and studies examining supplementation for various reasons are consistently negative regarding benefits. Vitamin D levels have been suggested to be related to the seasonality of viruses (as many of these peak in the winter when vitamin D levels decline) and thus some propose that supplementation can be valuable in preventing or mitigating these infectious diseases. Despite this, evidence supporting this is scant and generally unconvincing (most studies are equivocal regarding a benefit, with the exception of those who have severe vitamin D deficiency which is quite rare). More recently, more severe COVID-19 has been associated with reduced vitamin D levels (although one publication that proposed this has had an expression of concern issued in part because the claimed trend was… not really much of a trend). However, the groups most likely to experience vitamin D deficiency include the elderly, those with comorbidities, and people of color, all of whom have disproportionately high risk for COVID-19 for known reasons that have nothing to do with their vitamin D level and furthermore studies on vitamin D levels and patient risk of COVID-19 or severe outcomes are not consistent in their results; without randomized prospective studies demonstrating efficacy it is absolutely not appropriate to claim vitamin D supplementation has any value in the treatment or prevention of COVID-19. Vitamin D does have many roles (some of which oppose each other) within the immune system across many different cells, but these are not simple to tease apart within an actual person, and it is therefore hard to predict how vitamin D supplementation would affect a person’s risk of COVID-19 (but if the history of vitamin D supplementation is any indicator, it wouldn’t have any meaningful effect). Supplementation with vitamin D is generally (but not always) safe at reasonable doses (toxicity due to vitamin D can occur and is devastating when it does) but you should confirm with your healthcare provider before starting vitamin D supplementation and also discuss what dosage is appropriate as it is not safe for everyone. Most of the claims surrounding vitamin D are hype, and by no stretch of the imagination can it be claimed to offer protection against a given infectious disease that comes in any way close to that of a vaccine. Actors who suggest that it is an appropriate substitute do not have your best interests at heart. There are a few very effective things you can do to prevent COVID-19: wear masks, limit gatherings and travel to only when absolutely necessary, remain in well-ventilated, humidified areas, and complete the series for a COVID-19 vaccine you start when comes your turn. Vitamin D supplementation may be an adjunct to that but it cannot be a replacement.

There is a pervasive notion that vitamin D deficiency is a modern epidemic, and some individuals have extended this to mean that a therapeutic level of vitamin D is seemingly panacea. It is fairly common among vaccine denialists to dismiss any adverse consequences of vaccine-preventable disease through handwaving about nutritional requirements and how all those things could have been avoided if only the patient had proper intake of nutrient X (where nutrient X is often Vitamin D, but is also frequently vitamin A in the context of measles; their hatred of vitamin K is therefore superficially puzzling). This tends to be related to some magical thinking and notions of purity and is thus is often further used as a weapon to make unhealthfulness a moral failing.

All of these ideas are poorly founded and not supportable with the evidence available to us. Before I begin though, here are things that I am explicitly not saying but I am sure if not for this next part some will try to misrepresent me as having claimed (and in fact you can treat this next part as a denial of those claims):

  • Vitamin D is unnecessary.
    Vitamin D is an important hormone and plays critical roles in the regulation of calcium and phosphate metabolism. It is obviously necessary and claiming otherwise is ridiculous and not what I am doing.
  • Diet has no meaningful role in immunological function (this was the subject of an excellent recent review, as well as here).
    The interplay between diet and the immune system is a complex and exciting area in immunology and is the subject of much active research. However, this research is still in its infancy and we are far from being able to make definitive recommendations about dietary changes and a precise immunological effect in humans. Further complicating this are the intersections between this field and the examination of interactions between the microbiota and the immune system. The latter studies are often interesting and valuable but involve so many variables and often require such contrived models that external validity of the studies is an unrealistic claim; this field too, I would argue, is still in its infancy and a ways away from being able to use its findings to make clinical guidances (at least those more granular than “don’t overuse antibiotics”). Nonetheless, the research in these fields is valuable and should be done.
    Principally, the most valuable recommendation regarding diet and immunity is simply to ensure that adequate intake of all nutrients is achieved. The immune system is an extraordinarily complex network of molecules, cells, tissues, and organs and deficiency of any nutrient can potentially be harmful for its optimal function.
  • Vitamin D supplementation is harmful.
    You should consult with your care team before initiating any supplement. However, while vitamin D is fat-soluble and toxicity can be devastating, it is difficult to attain at reasonable doses of the substance, though in particular, it may be an issue in patients who have sarcoidosis or other granulomatous conditions. I do however think that vitamin D supplementation in most circumstances is a waste of money as it doesn’t do much and is often recommended for people who are not actually deficient.
  • Vitamin D deficiency is not necessary to correct.
    There is no good reason to be deficient in any nutrient (or hormone in this case). Physiologic levels of vitamin D are important for health maintenance. Any individual who has evidence of vitamin D deficiency should have it corrected. However, people should be realistic in the benefits associated with that correction.

What is Vitamin D?

Firstly, vitamin D (sometimes called calciferol) is not a true vitamin, as vitamins have to be obtained from the diet, but rather depending on what specific form of vitamin D you refer to, either a hormone or pro-hormone. For most people, most vitamin D is produced endogenously (in the body) through their skin on exposure to UV-B radiation from the sun, although in the US it is generally thought that people obtain most of their vitamin D from their diet through fortified foods like milk, cereal, and orange juice. The sources used to obtain vitamin D depend a great deal on where a person is in the world as it depends on latitude and resultant sun exposure as well as policies regarding food fortification. Further complicating things is that vitamin D does not denote a single molecule but rather a group of molecules ( vitamers, this is the case for many vitamins however e.g. vitamin K), broadly categorized into D2 ( ergocalciferol, generally regarded as being the vitamer of fungal and vegetable origin) and D3 ( cholecalciferol, generally regarded as being the vitamer of animal origin) and their associated metabolic products. Most of the vitamin D obtained from diet is D2, but D3 and even 25-OH-D3 can also be present. Formally vitamin D is a type of compound called a secosteroid, which is related to other steroids (cholesterol-derived molecules; secosteroids have a broken ring in their steroid nucleus- see the vitamin D structures), whose precursors are generated in the body via the mevalonate pathway (the target of the cholesterol-lowering drugs known as statins; interestingly statins do not appear to exert consistent lowering effect on levels of vitamin D (note that in RCTs the opposite was actually observed) and in some studies there is actually a significant increase in levels noted. As an aside, please respect statins- they are one of the most important and life-saving medications ever brought into the pharmacopoeia and are extremely safe when used appropriately). Vitamin D is probably best regarded primarily as a fat-soluble hormone that regulates calcium levels within the body, though it does have many other roles. Additionally, the A ring of vitamin D, as shown in the diagram, can rotate about the C6-C7 bond and change conformation, which has inspired some drug design of vitamin D analogues (e.g. calcipotriene) that are selective for a particular conformation and therefore exhibit certain selective effects of vitamin D.

Cell Biology and Mechanism of Action

Vitamin D has both genomic and nongenomic actions. The active form of vitamin D, 1,25-(OH)2-D, also known as calcitriol behaves analogously to other lipophilic hormones at the cell biology level, with the principal difference being that unlike other steroid hormones, its receptor is a heterodimer (rather than a homodimer) comprising the vitamin D receptor (VDR) and the retinoid-X receptor (RXR) together in a complex. Notably, RXR may not be bound to a retinoid for calcitriol to bind to the receptor. The binding of vitamin D to VDR causes the latter to be phosphorylated and induces the heterodimerization of VDR with RXR. VDR targets vitamin D response elements (VDREs) which are hexanucleotide repeats separated by triplet spacers. When calcitriol binds, a region on the VDR known as the AF2 helix becomes exposed and can interact with coactivators, which can stimulate recruitment of the mediator complex to help transition to euchromatin. The major coactivators involved are thought to be the p160 family which include steroid receptor activator 1, 2, and 3 (SRC-1, SRC-2, and SRC-3), and these have histone acetylase activity. When calcitriol is not bound to the site, the complex is associated with corepressors, which recruit histone deacetylases that keep the chromatin transcriptionally inactive. In addition to these, some evidence shows that the VDR complexed to calcitriol is capable of recruiting general transcription factors like TFIIB and some TAFs which form a stable pre-initiation complex that can then transcribe mRNA from the associated DNA. Expression of VDR can be enhanced by estrogen, calcitonin, retinoic acid, and the SP1 transcription factor. VDR expression is reduced by parathyroid hormone (PTH).

Though vitamin D is related to steroid and classically the effects of steroids are explained by genomic actions relating to changes in gene expression, vitamin D has demonstrated nongenomic actions as well. For example, transcaltachia, the transport of calcium ions into intestinal epithelial cells and subsequently into the blood mediated by vitamin D. This process seems to prefer s-cis form of vitamin D over the s-trans form. The protein membrane-a rapid response steroid-binding protein (MARRS) is thought to be involved in transcaltachia, which occurs within as little as one minute and is therefore not explainable by a genomic effect as this is too quick to require gene transcription. The influx of calcium ions into the cell results in activation of multiple signaling pathways, including those that affect cell division and those dependent on protein kinase C. This is also thought to induce something of a positive feedback loop in the cells: the influx of calcium into the cell promotes certain signaling pathways which may lead to increased expression of VDR, promoting the genomic actions of vitamin D.

The effects of vitamin D on cells are subject to autoregulation, as the presence of calcitriol induces production of 24-hydroxylase, which initiates the catabolic pathway of vitamin D. It additionally inhibits expression of 1-hydroxylases which prevents activation of calcidiol into active calcitriol. This is an important mitigating factor against toxicity, but it should not be taken to mean that toxicity is impossible. The details are discussed in figure 9.5.

Production, Absorption, Transport, Storage, and Metabolism

Within the skin, vitamin D is produced from a UV-B light light-catalyzed reaction from 7-dehydrocholesterol to make previtamin D3. This reaction is thought to have evolved relatively early in the history of life, as it can be found in phytoplankton, where it is thought to represent a protective mechanism against UV-mediated DNA damage. Previtamin D3 can also be converted to the inactive metabolites lumisterol or tachysterol via light, but spontaneously converts to vitamin D3 in a light-independent manner. This is not the biologically active form of vitamin D, and further enzymatic processing is required (detailed shortly). Of note, UV radiation is not innocuous and indeed can be hazardous to health, placing people at risk of skin cancers including melanoma, as well as photodamage and premature aging of the skin and thus caution is warranted in how consumers obtain their vitamin D. Notably however, vitamin D toxicity is a lower risk if it is obtained via endogenous production catalyzed by UV-B, as excess previtamin D will undergo conversion into inactive metabolites. The data are incomplete regarding whether or not sunscreen use affects vitamin D levels; the American Academy of Dermatology (AAD) recommends sunscreen of at least SPF 30, and while evidence that lower SPFs do not affect vitamin D levels is fairly robust, evidence for sunscreens whose protection is measured to be SPF 30 or greater is lacking. In any case, AAD notes that vitamin D should preferentially be obtained from either diet or supplementation, rather than sun exposure given the risks to human health posed from this.

Alternatively, vitamin D3 can be obtained from diet, where the predominant source is fatty fish, but also butter, cream, egg yolk, and is present in small quantities in cheeses. Depending on where in the world you reside, fortification of certain foods with vitamin D is mandatory, usually in the form of vitamin D2 rather than D3 (details in the D2 vs D3 section). Data regarding the details of absorption of vitamin D when obtained from diet are scarce, but there is a general consensus on the basic process. Animal studies suggest that absorption of vitamin D is enhanced in animals who are deficient. Despite being a very fatty molecule, studies suggest that absorption of vitamin D is mediated at least in part by carrier proteins, which is supported by the observation that deficient animals tend to absorb more vitamin D. Cholesterol, as well as long chain fatty acids and phytosterols, have been shown to interfere with absorption of vitamin D suggesting competition for similar targets, which makes sense as these compounds are similar in terms of their properties (their electronic ones in particular, but also sterically in the case of cholesterol and its derivatives). While not very well defined for vitamin D, for cholesterol the principal proteins mediating absorption are Niemann-Pick C1-like carrier 1 (NPC1L1), Scavenger receptor class B type I (SR-BI), cluster of determination 36 (CD36), and ATP-binding cassette transporter A1. Treatment with ezetimibe, an inhibitor of NPC1L1, does reduce dietary absorption of vitamin D, suggesting that vitamin D may also make use of this receptor. Of further interest is that 25-OH-D has superior absorption to calciferol, despite being more polar and thus expected to be more difficult to absorb. It is estimated that about 50% of ingested vitamin D can be absorbed (mostly in the distal small intestine but rates of absorption are greatest at the jejunum and ileum), and of note, this does NOT seem to decrease with age. When obtained from diet, vitamin D associated with other dietary fats which are emulsified by bile salts secreted from the gallbladder and with the aid of digestive enzymes (primarily lipoprotein lipase) from the exocrine pancreas. Being a fatty molecule, vitamin D is packaged with other lipids into micelles and then structures called chylomicrons which are absorbed in the small intestine, and a significant but smaller portion of the vitamin D does enter the hepatic portal system to directly reach the liver. Systemic circulation of the chylomicrons is achieved when lymphatic vessels unite with the blood in the thoracic duct. As the chylomicrons are transported, tissues expressing lipoprotein lipase- primarily adipose (fat) and muscles- liberate small amounts of vitamin D which they absorb. After treatment with lipoprotein lipase, chylomicrons become chylomicron remnants which are enriched in cholesterol and still retain some vitamin D. These then traffic back to the liver, summarized in figure 400–2.

Alternatively, vitamin D in the circulation may be found in association with vitamin D-binding protein (DBP, formerly called Gc-globulin; incidentally one of the most polymorphic genes in the human genome), particularly if it is produced within the skin. DBP has three major isoforms and two glycosylation patterns, resulting in six configurations that differ in binding affinity and differ by race. It is the most important plasma carrier protein of vitamin D (about 88% of the body’s vitamin D is associated with it), but a smaller proportion of vitamin D can also be associated with albumin (approximately 11%), the most abundant protein in the blood, and the free form of 1,25-(OH)2-D unbound to any carrier (approximately 0.03%). DBP has a hierarchy for the affinity with which it binds vitamin D metabolites: 25-(OH)-D-23,26-lactone > 25-OH-D = 24,25-(OH)2 D = 25,26-(OH)2 D >> 1,25-(OH)2 D >> vitamin D >> previtamin D, and it has a slightly stronger binding affinity for D3 over D2 metabolites. Within the plasma, DBP’s concentration exceeds that of vitamin D metabolites by a factor of about 50. The binding affinity for 1,25-(OH)2-D (the highly active form of vitamin D) is about 100 times higher than for 25-OH-D and this is thought to reflect the free-hormone hypothesis, in that hormones (especially lipophilic hormones) cannot function unless they are free as they are unable to reach their target receptors on or within cells. Additionally, DBP also binds actin with very high affinity. Actin is a component of the cytoskeleton (cell’s skeleton) and when released from cells such as if they are damaged, it forms filaments rapidly that can impede microscopic circulation. It is thought that DBP prevents this process from occurring. Genetic studies on mice in which the DBP gene is knocked out show that the mice lose vitamin D metabolites in urine and these are more sensitive to vitamin D deficiency (but do not generally show vitamin D deficiency unless they have a diet low in it) and less sensitive to toxicity (presumably because of urinary loss of vitamin D metabolites). Taken together this supports that DBP’s major function is to help prevent loss of vitamin D metabolites and to control the amount of available (i.e. functional) vitamin D in the body at any given time. A family was found which had loss-of-function mutations in the DBP gene; the patient who had these mutations on both maternal and paternal copies (the proband) had apparently normal levels of calcium, phosphate, and parathyroid hormone, but very low levels of vitamin D and associated metabolites that did not respond to supplementation, though free 25-OH-D was nearly normal. This is compelling support for the free hormone hypothesis, which holds that for lipophilic hormones are only functionally active if unbound to other substances so that they can have productive interactions with their receptors, as it demonstrates essentially normal calcium metabolism despite the absence of DBP.

However, the free hormone hypothesis encounters a challenge with vitamin D because of the existence of megalin and cubilin, which form a complex on the cell membrane. These two proteins are found primarily in the kidney but also the parathyroid gland, placenta, and several other tissues and recognize vitamin D in complex with DBP. In mice in which these proteins are removed, osteomalacia results- a hallmark of vitamin D deficiency. It is thus thought that DBP-complexed vitamin D represents a reservoir of vitamin D that can be tapped into when there is limited endogenous production or dietary acquisition. Importantly, megalin and cubilin have complex roles beyond vitamin D metabolism and thus genetic studies where these genes are removed or expression is reduced will produce effects beyond what can be attributed solely to vitamin D metabolism. Therefore it is principally thought that free calcitriol represents the active form of vitamin D.

As summarized by Figure 59–1, D vitamers exhibit complex metabolism throughout the body in several key sites. Upon transport to the liver, mitochondrial CYP27A1 (also called 27-sterol hydroxylase) or microsomal CYP2R1, CYP2D11, CYP2D25, CYP3A4 carry out 25-hydroxylation. This process has been observed outside the liver in vitro but the liver is thought to be the principal site of 25-hydroxylation. In general, this step is not highly regulated and for this reason 25-OH-D levels are generally relied upon as the measure of vitamin D status in patients. Upon completion of this, 25-OH-D complexed to DBP is transported to the kidney where megalin-cubilin uptake may occur or the 25-OH-D may spontaneously dissociate and enter cells and undergo 1α-hydroxylation via mitochondrial CYP27B1 ( 1α-hydroxylase)in the renal proximal tubule. This enzyme has also been noted to be expressed at high levels however in keratinocytes, macrophages, bone, and other tissues, though it is thought that the kidneys are the principal site of 1α-hydroxylation. Upon completion of 1α-hydroxylation vitamin D is converted to the highly active 1,25-(OH)2-D ( calcitriol). Expression of renal 1α-hydroxylase is controlled by parathyroid hormone (PTH), 1,25-(OH)2-D levels, calcium, phosphate, and fibroblast growth factor 23 ( FGF23). From there, several pathways exist to inactivate vitamin D3 once it has exerted its biological effects whereupon it is released into the gut within bile, and defects in these pathways such as from genetic lesions may produce vitamin D toxicity. Classically, 24-hydroxylation is regarded as initiating the breakdown of vitamin D metabolites, forming 1,24,25(OH)3D or calcitroic acid, which is biologically inactive. CYP24A1 (24-hydroxylase) is expressed in all cells containing the vitamin D receptor ( VDR; explained shortly), and mutations in this gene are associated with infantile hypercalcemia. Calcitroic acid has substantial affinity for DBP though it has only about 10% of the potency of 1,25-(OH)2-D. Alternatively, active 1,25-(OH)2-D can be inactivated via the C23 oxidation pathway (initiated by 24-hydroxylase) or C26 oxidation pathway which produce an inactive lactone product. These are generally regarded as inactive products and their functional significance is not well understood. Alternatively, calcitriol can undergo sulfation and glucuronidation wherein it forms inactive metabolites that are excreted in the bile (vitamin D-bile salt conjugates are known to undergo enterohepatic cycling) and urine.

Unlike other fat-soluble vitamins, storage of vitamin D does not occur in the liver for the most part, instead occurring primarily in the muscle and adipose tissue. It is for this reason that individuals who have a higher adipose content may require higher levels of supplementation to correct deficiency. It has been observed that for individuals with high adiposity, weight loss also results in release of stored vitamin D. Skeletal muscle cells are also an important store of vitamin D, particularly during the winter months, where it is thought that uptake occurs via a megalin-cubilin mechanism.

D2 vs D3

Virtually all countries fortify food staples with vitamin D to prevent rickets, though a few don’t because of (1) deficiency being very rare e.g. countries near the equator, (2) disputes about whether the cost of fortification should be paid by industry or by the government, or (3) concerns for toxicity, particularly in the very young who are more prone to it. The metabolism of both forms of vitamin D are basically identical, and structurally the molecules are very similar except for a small side chain. In general, reviews of their potency find them to be essentially equivalent, and thus for the purposes of supplementation, in general either one is considered acceptable. However, D3 is shown to raise levels of calcitriol more rapidly than D2, but is also associated with greater toxicity. It is thought that D2 is cleared from the circulation more rapidly than D3 which it is thought helps to prevent toxicity. Despite this, in general, D2 has not been shown to be inferior in pharmacologic quantities at preventing rickets- the hallmark manifestation of vitamin D deficiency.

Vitamin D in Mineral Homeostasis

The most important “classical” role of vitamin D is as a regulator of calcium metabolism (and in less direct ways phosphate metabolism). The calcium ion has incredibly broad biological functions as it is a common second messenger in many signaling pathways, participates in regulation of electrochemical gradient across cells, and is a major component of bone together with phosphate (bone’s mineral matrix mainly comprises hydroxyapatite). Both excess and inadequate calcium levels can be medical emergencies in the acute setting. Perhaps the most important role of vitamin D is to ensure adequate absorption of calcium into the body, which is a function controlled primarily by the intestine. Calcium is a divalent cation which sticks to anionic species that may be obtained in the diet as well, and thus acidification of food matter is important for helping it to peel off. Additionally, calcium absorption can be enhanced through the presence of certain chelators that help function as ionophores, such as vitamin C (it’s really not surprising that inappropriately huge doses of vitamin C can result in calcium oxalate stone formation in the urinary tract). It is thought that about half of ingested calcium is absorbed, but of this 32.5% or so is secreted back into the intestinal lumen and released in the feces. Absorption of calcium from diet occurs primarily in the small intestine. The actual transport of calcium is fastest in the duodenum but the bulk of import occurs in the distal intestine. Classically the import of calcium through the intestine is described as occurring in 3 steps:

  1. Calcium ions enter at the apical surface of the intestinal epithelial cell through channel proteins like TRPV5 and TRPV6. Other protein channels are known to play a role here but it is not clear which specific ones.
  2. On entry into the intestinal epithelial cells, the free ions are bound up by calbindin-D9k, which shuttles the ions to the basolateral surface.
  3. The ions are released from the basolateral surface into the body via PMCA1b.

However, in addition to the transcellular route (where the calcium literally goes through the cells), which is incompletely understood as it clearly involves additional proteins given the effect of TRPV5/6 knockouts on calcium absorption, calcium may also be absorbed paracellularly (between the cells) aided by complexes of claudin-2 and claudin-12, as well as cadherin-17 and aquaporin-8. Under conditions where dietary calcium is high, the paracellular pathway predominates. This pathway is also enhanced by vitamin D.

In addition to ensuring adequate absorption of calcium, vitamin D has a critical role in preventing the loss of calcium into the urine. Within the kidney, this occurs primarily in the proximal tubule cells of the nephron. Vitamin D is transported primarily through the megalin-cubilin pathway in complex with VDR, wherein it inhibits expression of 1α-hydroxylase (the enzyme that induces formation of calcitriol, the active form of vitamin D) together with FGF23, a key phosphaturic hormone (promotes phosphate loss in the urine). This functions as a negative feedback mechanism to prevent toxicity of vitamin D. In the state of chronic renal failure, conversion to calcitriol can be disrupted and result in increased calcium loss into the urine which produces renal osteodystrophy in which the bones weaken from unbalanced resorption (the transport of calcium and phosphate from bones into the blood).

Within the proximal tubule cells, most calcium is reabsorbed (an estimated 1 to 2% of calcium filtered by the kidney actually ends up in the urine under physiological conditions). Most of this reabsorption is passive and does not depend on calcitriol but rather relies on the sodium gradient created across the nephron. Calcium ions enter via TRPV5 and then similar to the IECs, get taken up by calbindin-D9k/28k and then exported out via either PMCA1b or NCX1 (the sodium-calcium exchanger). TRPV5 is thought to represent the rate-limiting step in the reabsorption process. Both PTH and calcitriol enhance reabsorption of calcium. The regulation of TRPV5 expression also occurs in part through klotho and FGF23, wherein klotho is thought to promote apical retention of TRPV5 on the cell membrane through modification of its glycosylation, and FGF23 also appears to promote TRPV5 expression on the membrane by enhancing the abundance of the protein.

Understanding the effects of vitamin D on bone are more complex. Bone comprises a matrix of hydroxyapatite (a calcium phosphate-based mineral) and primarily 3 cell types: osteoclasts, osteoblasts, and osteocytes.

  • Osteoclasts are the major cells that break down bone, releasing calcium and phosphate from the bone mineral matrix into the blood, known as bone resorption. They are derived from monocytes and do not divide (some regard them as the tissue-resident macrophages of the bone). Mature osteoclasts have receptors for calcitonin, but not parathyroid hormone or vitamin D. To break down bone, osteoclasts bind a region and form an adhesive ring which makes contact with a specialized organelle known as a ruffled border, which is essentially a giant lysosome containing proteases like cathepsin K which break down the protein components of the bone and liberate free calcium.
  • Osteoblasts are the major bone-forming (known as bone mineralization) cell. They express receptors for both parathyroid hormone and vitamin D and can lay down matrix called osteoid that gets filled with hydroxyapatite. These cells also secrete large amounts of the alkaline phosphatase enzyme, which is required for bone mineralization. These cells are the major target for parathyroid hormone.
  • Osteoblasts that remain in the bone during the remodeling process become osteocytes. Osteocytes can regulate the functions of osteoblasts through secreted substances. Osteocytes are thought to play a role in the transfer of mineral from the interior of bone to the growth surfaces. A detailed review of osteocyte functions can be found here.

Bone represents a reservoir of calcium which can be pulled from or deposited into depending on the given needs of the body. However, the calcium content of the bone must be tightly regulated to prevent brittleness and subsequent fracture. Bone is continuously remodeled through the efforts of osteoclasts and osteoblasts in adulthood and chondrocytes (cartilage cells) ensure that bone continues to lengthen during childhood and growth. Humans and mice that lack the vitamin D receptor or 1α-hydroxylase develop rickets and osteomalacia but this effect disappears with adequate calcium supplementation. In the state of negative calcium balance, there is an inadequate supply of calcium, and thus vitamin D induces bone resorption to ensure adequate levels of blood calcium. This effect is primarily mediated through the actions of vitamin D on osteoblasts. The osteoblasts supply osteoclasts with RANKL, which binds the cognate RANK receptor on the osteoclasts that increases their formation and action to promote bone resorption. Therapies targeting this pathway exist as treatments for osteoporosis e.g. denosumab. Vitamin D is able to also enhance production of osteopontin and pyrophosphate which both inhibit bone mineralization. Alternatively, if there is a surplus of calcium, positive calcium balance, vitamin D seems to promote bone mineralization, although this anti-resorptive effect is not well understood regarding mechanisms. It is noted that there is a decreased ratio of RANKL to osteoprotegerin (OPG, a decoy RANKL receptor). Both vitamin D and PTH promote osteoclastogenesis. In mature osteoblasts however, vitamin D is shown to have anabolic and anticatabolic activity that increases bone mass via the LRP5 gene.

Vitamin D signaling also has important roles in the parathyroid, wherein the calcium-sensitive receptor (CaSR) and parathyroid hormone (PTH) genes are both under the control of vitamin D. Additionally, the parathyroid does express 1α-hydroxylase and its knockout produces hypocalcemia, reduced levels of calcitriol, even if expression in the kidney is normal.

Extraskeletal Effects of Vitamin D

In general, vitamin D supplementation for extraskeletal health are largely not substantiated (see “Controversy” Section), and it’s important to be mindful of the potential problem of reverse causation: individuals who are sedentary or have heightened adiposity are much more likely to have lower levels of vitamin D because of reduced sun exposure or retention in adipose tissue in the latter, and these states are both associated with chronic illnesses for unrelated reasons. Nonetheless, vitamin D does play a role in many functions that are unrelated to skeletal health. I won’t discuss them all because frankly a ridiculous number of things are documented (as vitamin D receptors are present in almost every cell), so I will constrain myself to just a few of the ones I feel are important.

Given the interest in COVID-19, it seems worthwhile to discuss the complex connection between vitamin D, the renin-angiotensin system (RAS), and cardiovascular health broadly. The renin-angiotensin system is a key regulator of blood pressure. Angiotensin II (ang II) is a potent vasoconstrictor, and it is produced from inactive angiotensinogen via renin to make angiotensin I (ang I) (which is generally regarded as inactive) and then into ang II via angiotensin-converting enzyme (ACE). Angiotensin II can undergo further processing, including by ACE2 into Angiotensin (1–7) (more on this shortly). Ang II can also stimulate release of antidiuretic hormone from the neurohypophysis to expand extracellular volume and raise blood pressure which acts synergistically with its vasoconstrictive effect. Signaling of Ang II occurs via angiotensin receptors- AT1 and AT2. AT1 is thought to mediate the vasoconstrictive effects, as well as promote retention of sodium (the most abundant electrolyte in the blood), and myocardial hypertrophy, while AT2 does the opposite, being involved in vasodilation, natriuresis (loss of sodium in the urine) and myocardial growth inhibition. Excessive activation of this axis can result in chronically elevated blood pressure. As it happens, vitamin D is a negative regulator of the renin-angiotensin system. More specifically, calcitriol is able to inhibit the biosynthesis of renin, thereby preventing the initiation of the cascade (although incompletely). Interestingly, if one takes this to its logical endpoint, this means less renin, which means less production of ang II and subsequently Ang (1–7) via ACE2. Ang (1–7) has actions that broadly oppose those of Ang II, and through signaling via the Mas receptor, it is shown to actually reduce lung inflammation, as well as fibrosis and pulmonary hypertension. Furthermore, higher ACE2 expression is actually associated with better outcomes in inflammatory lung disease, which would imply that vitamin D may actually be harmful in COVID-19, though of course, that evidence hasn’t borne out. I make mention of this mostly to demonstrate how malleable these ideas can be and the importance of having actual data. More broadly, there is an inverse relationship between the risk of cardiovascular disease and vitamin D levels, but this relationship is not necessarily causal (I would venture as far as saying it is almost certainly the result of confounding) because lower vitamin D levels are associated with other cardiovascular risk factors (e.g. heightened adiposity, sedentary lifestyle, suboptimal nutriture). This is supported by Mendelian randomization studies which do not show clear associations between vitamin D-related gene polymorphisms and cardiovascular risk, and further by the absence of apparent benefit from supplementation. It is also worth bearing in mind that most of the cardiovascular effects exerted by vitamin D are accomplished via paracrine and autocrine signaling, in contrast with the classical skeletal effects which are endocrine, and thus these may be harder to influence through supplementation.

Immunological Effects of Vitamin D

Vitamin D is as a molecule is related to corticosteroids, which are broadly immunosuppressive, and this property is shared by vitamin D at least partially. Vitamin D is shown to suppress the production of a number of pro-inflammatory cytokines like interleukin-1 (IL-1), IL-2, IL-6, IL-23, tumor necrosis factor-α, and interferon-γ, and also enhance production of anti-inflammatory ones such as IL-4 and IL-10. More generally, vitamin D is generally regarded as enhancing Th2 (and regulatory T cell) polarization of the immune system- the sort of immune response that is most useful for dealing with helminthic infections, while suppressing Th1-biased responses that are ideal for helping clear intracellular bacteria and viruses and also Th17-biased responses most useful for combating extracellular bacterial infections. Despite this, vitamin D is classically regarded as having an anti-mycobacterial effect because it is shown to enhance production of cathelicidins within macrophages that can be used to kill mycobacteria. This has inspired study of its value in the setting of tuberculosis, which is a mycobacterial disease, in the hopes that it could be given as therapy or prophylaxis. However, in general, studies have been negative regarding a therapeutic and prophylactic benefit (note that this study specifically examined individuals with documented vitamin D deficiency who are therefore most likely to benefit from supplementation). Because vitamin D’s immunological effects are broadly pro-tolerogenic, it has been examined in the setting of atherosclerosis. Though fundamentally because of inappropriate deposition of excess cholesterol (as well as other lipids and calcium, among other substances) beneath the vasculature, atherosclerosis is a chronic inflammatory condition as well and it is exacerbated by cytotoxic T cells, whose function is significantly enhanced with the aid of Th1 cells that vitamin D suppresses. Despite this, as above, benefits for vitamin D for cardiovascular risk have not been observed (see above). On the basis of its immunoregulatory effects, it has also been proposed that vitamin D levels may play a role in the prevention and modulation of type 1 diabetes, for which some evidence exists.

Because of the seasonality of respiratory viruses, vitamin D has been proposed to be a factor contributing to the commonly observed increased burden during the wintertime, as reviewed very thoroughly here. This is discussed in more detail in the next section, but it is important to note that this is just one of many factors that may contribute to increased burden of the viruses during the wintertime, which are also reviewed by Moriyama et al. Despite its immunoregulatory role and tendency to skew away from immunological states that would classically be regarded as being beneficial in viral infection, there is some evidence for antiviral properties of vitamin D. For example, vitamin D is shown to reduce NF-kB signaling without affecting interferon signaling significantly in RSV infection, serving to promote the antiviral state of the airway epithelium and reduce inflammation (note however that this evidence is preclinical and thus needs to be interpreted with caution). The ability of vitamin D to induce cathelicidins that suggest its value in the treatment of mycobacterial disease has lead some to extend the hypothesis that it may have value in viral infections as well. The sum total of our knowledge of the immunologic effects of vitamin D indicate that explicit clinical data are required to determine its value in specific infectious diseases, as it has highly pleiotropic functions, some of which oppose each other, making it very difficult to parse from principles which effects will dominate.

The Controversies and Misconceptions of Optimal Levels

Perhaps the most important question clinically in assessing vitamin D status and health is what levels are optimal. This question turns out to be not at all simple and in fact there exists a difference of opinion between multiple professional societies. In 2011, the Institute of Medicine (IOM, now the National Academy of Medicine, NAM) authored the report Dietary Reference Intakes for Calcium and Vitamin D, wherein it proposed that calcidiol (25-OH-D; this is the precursor to active vitamin D, calcitriol, 1,25-(OH)2-D, and is used because even in deficiency the levels of calcitriol may be normal) levels of at least 20 ng/mL (equivalent to about 50 nmolar) were sufficient for health and concluded that between diet and sun exposure 97.5% of the US would meet the requirements for vitamin D levels. The Endocrine Society, the professional body of endocrinologists in the US, set the lower limit at 30 ng/mL (75 nmolar). Naturally, if you set the lower bound of a normal range at a higher value, many more people would be considered vitamin D-deficient and this would warrant correction with supplementation; under the Endocrine Society’s definitions most of the US is vitamin D deficient, while under the IOM’s definitions, the vast majority if the US is vitamin D sufficient. So, who’s right? The IOM.

The IOM explained the rationale for its ranges here after the Endocrine Society voiced dissent. In essence, the IOM judged based on the evidence that there wasn’t convincing evidence of health benefits in the general population for calcidiol levels above 20 ng/mL and that all individuals with levels of calcidiol below 20 ng/mL were deficient. Their review of the data showed that there was no additional benefit once serum levels rose to 20 ng/mL and the effect on skeletal health seemed to plateau between 12 and 16 ng/mL (note that the active form of vitamin D, calcitriol, is tightly regulated by PTH, calcium, and phosphate levels). The Endocrine Society countered by noting that there is a reduced risk of falls at levels of 30 ng/mL and above, as well as a decline in the levels of PTH, and maximized calcium absorption. The review by the IOM disagreed with these conclusions. In a review of 59 studies, most actually found maximal suppression of PTH at levels of 15–20 ng/mL, and further several of the studies the Endocrine Society used in their assessment relied on weak statistical methods that were inappropriate for the dataset. Regarding falls, one meta-analysis noted that falls were inversely correlated with vitamin D levels in the elderly (this seems to be related to the observation that vitamin D deficiency causes muscle weakness). Except here we too see some statistical trickery as the data does not fit well with the trendline assigned to it at all. The IOM took the time to reanalyze the data from this study and found no relationship. Subsequent studies also are largely unable to show a benefit for vitamin D supplementation as a fall prevention strategy, and in fact in some studies, those receiving the highest dose of supplementation actually have the highest incidence of falls. Regarding calcium absorption, the Endocrine Society appeared to base its guideline on a study of 34 patients which is puzzling because a much larger study showed that calcium absorption is maximized between 8 and 20 ng/mL.

There is an important caveat attached to the IOM’s position:

In clinical situations in which the true requirement of an individual member of the general population cannot be known, practitioners’ interest may focus on maintaining a serum 25OHD level of 20 ng/ml for that individual. This is not the same as declaring the general population to be deficient if average serum values are less than 20 ng/ml. 2 Rather, it is reasonable to expect that for practitioners there are likely to be alternative definitions of vitamin D deficiency for a patient with a disease condition or a health-related consideration. The goal of the guideline therefore should be to provide a rationale for such a definition specific to the evidence associated with the condition of interest. The [Endocrine Society’s] guideline has not provided the rationale.

It does however, get even more complicated when you consider people with darker skin pigments, such as Black and Hispanic people. The extra pigment in their skin can interfere with production of vitamin D within the skin from UV radiation. Naturally, one would expect then that they would exhibit hallmarks of vitamin D deficiency like rickets and osteomalacia and be more prone to fracture. Except analysis shows the opposite is true. Their bone density is on average higher and they do not experience bone fractures as frequently. However, by basing its guidelines on an assumption of minimal to no sun exposure (i.e. all vitamin D would be dietary), the IOM circumvented this issue and thus its guidelines for appropriate levels are likely appropriate even for those with darker skin pigments. At one point, a possible explanation for the apparent inconsistency in skeletal health and vitamin D status in those with darker pigmented skin was the thought that these individuals have lower levels of vitamin D binding protein (DBP) which results in greater levels of free vitamin D. However, subsequent investigation has suggested that the methods used to make this conclusion were biased and the level of bioactive vitamin D is indeed lower in people with darker skin pigment compared with paler-skinned individuals. It is unfortunately unclear what the basis of this paradox is and how, if at all, it should guide the normal ranges set on vitamin D levels for subpopulations with higher melanin content (discussed in some detail here).

There is additionally a financial component here which cannot be ignored. Firstly, evaluating vitamin D levels in the population can readily cost billions of dollars depending on how many people are tested and how frequently. This issue is also raised by the IOM:

The guideline goes on to inappropriately conclude that the benefits of vitamin D for a large segment of our population occur only when serum levels are at 30 ng/ml 25OHD and above, and mistakenly concludes that all persons with serum 25OHD levels below 20 ng/ml are deficient in vitamin D. In turn, the guideline conclusion that at least half of our population requires routine testing represents a large, unnecessary cost.

The Endocrine Society guidelines are extremely broad with respect to defining groups at risk of vitamin D deficiency and those that therefore need screening. One of the authors of the Endocrine Society guideline is Michael Holick, who is a controversial figure for many reasons. For one, he has something of a reputation as a child abuse contrarian, as he is occasionally called to testify in court cases as an expert witness regarding whether or not the nature of a child’s injury is the result of abuse and in all cases in which he has been called to testify, he has asserted that the apparent injury pattern is the result of weakened bones from the Ehlers-Danlos syndrome, a connective tissue disorder mainly defined by hypermobility of the joints. Experts generally feel it cannot be diagnosed before the age of 5. Literature is largely unsupportive regarding the point that these patients are at much higher risk of bone fracture. Testifying in these hearings is wildly outside the scope of his training as an endocrinologist- the child abuse certification is a subspecialty level certification in pediatrics. It is extremely specialized and it is far beyond the scope of just anyone who has MD next to their name to opine on. Holick has also previously advocated for tanning beds, which the AAD considers to generally be dangerous and advises against, and the WHO groups to be a carcinogen in the same class as tobacco (though to be clear, these classifications are based on the certainty with which the product in question contributes to cancer rather than the extent). Holick’s expertise on vitamin D is largely unassailable though, as he did discover the active form of the vitamin and has authored hundreds of peer-reviewed publications about it since. It makes intuitive sense that he would be an author on the Endocrine Society’s guideline. Except … Holick has a direct incentive for screening as many people as possible for vitamin D deficiency because he has financial ties to the diagnostics industry. Creating a pandemic of vitamin D deficiency would be a surefire way to ensure that those diagnostics get run and he profit in the process. Ultimately, the actors are irrelevant: it’s about what the evidence says. But there is very clearly much more evidence to support the IOM’s position over that of the Endocrine Society’s, but one could understand how the Endocrine Society could wander so outside the evidence base given the circumstances.

Nonetheless, much additional data has been gathered exploring the value of vitamin D supplementation in various conditions since. It’s tempting given the vast array of biological effects attributed to vitamin D to think that vitamin D supplementation could have value in many of them, but this is unfortunately not what the data show. Perhaps the most thorough analysis comes from Autier et al who write (emphasis mine):

Although vitamin D doses were greater than those assessed in the past, we found no new evidence that supplementation could have an effect on most non-skeletal conditions, including cardiovascular disease, adiposity, glucose metabolism, mood disorders, muscular function, tuberculosis, and colorectal adenomas, or on maternal and perinatal conditions. New data on cancer outcomes were scarce. The compilation of results from 83 trials showed that vitamin D supplementation had no significant effect on biomarkers of systemic inflammation. The main new finding highlighted by this systematic review is that vitamin D supplementation might help to prevent common upper respiratory tract infections and asthma exacerbations. There remains little evidence to suggest that vitamin D supplementation has an effect on most conditions, including chronic inflammation, despite use of increased doses of vitamin D, strengthening the hypothesis that low vitamin D status is a consequence of ill health, rather than its cause. We further hypothesise that vitamin D supplementation could exert immunomodulatory effects that strengthen resistance to acute infections, which would reduce the risk of death in debilitated individuals. We identified many meta-analyses of suboptimal quality, which is of concern. Future systematic reviews on vitamin D should be based on data sharing so that data for participants with the same outcomes measured in the same way can be pooled to generate stronger evidence.

In other words, vitamin D supplementation didn’t seem to do much at all for basically any condition, with the possible exception of upper respiratory tract infections and asthma exacerbations, to the best of the analysis’s ability to render reliable conclusions. Let’s examine those in a bit more detail.

Martineau et al was the meta-analysis of RCTs which found a benefit for upper respiratory infections. Unsurprisingly the benefit was substantial for those who had profound deficiency (10 ng/mL or less) and took supplements weekly (adjusted odds ratio 0.30, 0.17 to 0.53), and indeed any vitamin deficiency should be corrected, but otherwise the results are somewhat underwhelming. The forest plot is shown in figure 2, and as you can see the data are pretty heterogeneous. The vast majority of the studies contain 1 within their confidence interval (indicating no effect) or very nearly do, and most of those that don’t contain what I would consider to be very small sample sizes which are prone to artifactual findings. I’d like to look more closely at a few of the studies showing statistically significant benefit:

  • Manaseki-Holland 2010: This was a double-blind 1:1 RCT of 453 children in an inner‐city hospital in Kabul, diagnosed with non‐severe or severe pneumonia at the outpatient clinic. The study is well designed and its principal finding is a significantly lower risk of repeat episodes of pneumonia among the group receiving vitamin D, which looks real to me (45 % in the vitamin D group vs. 58% of the placebo group). Children were excluded from the study if they had evidence of rickets or were known to have received high-dose vitamin D supplementation within 3 months of the study. The randomization looks less than ideal to me but not unreasonable. The principal limitation of this study as I see it is that its subject population is very young children (mean age about 13 months) and it appears to be in a region where vitamin D deficiency is very prevalent which makes me concerned about the external validity of the study.
  • Camargo 2012: A cluster-randomized trial examining 247 Mongolian schoolchildren randomized to vitamin D-fortified milk vs. placebo comparing the incidence of acute respiratory infections as per parental report. One of the major things that jump out in this study is the mean vitamin D level of these patients was 7 ng/mL, which would be universally considered severely deficient. Another potential issue is that it relies on parental reporting which isn’t ideal as it’s subject to a number of biases, but this is mitigated by the randomization process of the trial. The benefit here does also seem real to me, but again, it’s the same problem of external validity: this level of deficiency in places like the US is very rare, and we aren’t principally concerned with schoolchildren in the case of COVID-19.
  • Laaksi 2010: DBPCT of 164 young, Finnish men who received vitamin D supplementation vs. placebo which compared the number of days absent from daily duty. There was a clear benefit for the supplementation group. However, these groups also had pretty significant vitamin D deficiency to start with (~8 ng/mL), so that’s not particularly surprising. The principle is true with vitamins: a health benefit is seen when true deficiency is corrected.
  • Bergman 2012: This is a very interesting study because it examines specifically patients who have diagnosed or possible occult immunological deficiency that predisposes to respiratory infection, and it also followed all 140 patients for 1 year, which is nice (we love a long prospective study). The levels of vitamin D at baseline between the groups are similar but under some classification schemes, the placebo group would be considered mildly/borderline deficient while the treatment group starts out at a level that is sufficient. The major difference was that antibiotic use was rarer in the treatment group than the placebo group, which accounted for the marginally significant result obtained. However, the confidence intervals for this estimate are very wide, which speaks to the study being underpowered as it only has 140 patients. Further, it involves such a specific group of patients that I don’t think it’s very generalizable.
  • Marchisio 2013: 116 children with a history of recurrent middle ear infections were randomized to vitamin D supplementation or placebo for 4 months and the frequency of middle ear infections were monitored for 6 months. The small sample size made randomization difficult, as the gender distribution between the groups is not super well-matched, the number of children that had a symptomatic allergy in the vitamin D group was lower at baseline, and the vitamin D group was also more likely to be breastfed for > 3 months. These differences aren’t huge though. Importantly, the vitamin D levels in both groups were similar and would be considered sufficient under the IOM ranges. Interestingly, the vitamin D levels in the placebo group declined quite a bit throughout the study to a level that would be considered deficient. Vitamin D didn’t seem to affect complicated middle ear infections much, but did show significant reduction in uncomplicated and overall showed a reduction that took about 40 days to manifest. These effects are pretty big though and this meta-analysis does note a relationship between vitamin D levels and otitis media. So I think this might actually possibly be a legitimate benefit for vitamin D supplementation, but middle ear infections are generally a pediatric issue.

There is however an important counterpoint to the Martineau meta-analysis: Pham et al ran a very large randomized controlled trial focusing on older adults in Australia who received either vitamin D or placebo and compared the incidence of respiratory infections and symptoms in these groups. There were no differences noted. Now, it’s critical to note here that both groups had vitamin D levels which would be considered replete by both Endocrine Society and IOM standards, but the point here is: you aren’t going to benefit from supplementing something you already have adequate levels, and most people (in the US anyway) do have adequate levels.

You may have noticed that vitamin D supplementation for osteoporosis and fracture prevention was not identified as a benefit, which may seem unusual as this seems to be common wisdom. After all, what does vitamin D do if not strengthen bones? Well- it’s been looked at: neither vitamin D nor vitamin D with supplemental calcium seem to help with problems relating to bone mineral density (with the possible exception of the hip and femoral neck). Ironically, parathyroid hormone has actually shown value here: its effects depend on the frequency with which it is given and it can in fact coordinate remodeling and strengthening of bone if given intermittently. Note that even the Endocrine Society recommends vitamin D be used only as an adjunctive therapy (together with calcium) for osteoporosis. However, vitamin D levels in serum are not biomarkers for osteoporosis risk. The US Preventive Services Task Force’s most recent guideline notes that vitamin D supplementation with or without calcium was not associated with a reduction in the incidence of fractures, but was noted to increase the risk of kidney stones.

Additionally, I feel compelled to include a Professor Aaron Carroll’s summary here, which, while not a meta-analysis, does go through a large chunk of vitamin D clinical literature.

I do think it is worth in brief discussing what can make vitamin D levels decline, aside from inadequate synthesis or dietary intake. Vitamin D is a negative acute phase reactant, meaning that in states of acute inflammation, its levels can decline. It is further suggested that individuals who have chronic low-grade inflammation (e.g. inflammaging) are more likely to have reduced vitamin D levels as a result which is consistent with the observation of vitamin D as a negative acute phase reactant. Interestingly, vitamin D is not the only substance related to calcium homeostasis whose levels are changed with inflammatory states. Procalcitonin, the precursor to calcitonin (generally regarded as being a dispensable player in calcium homeostasis but one whose actions directly oppose those of parathyroid hormone and some of those by vitamin D), is sometimes used as a biomarker to help determine patients who are at risk for severe bacterial infections like pneumonia or sepsis where it can be used to help guide antibiotic therapy (it is generally thought to be suppressed in viral infections but this is a bit controversial). Despite this, its value as a biomarker is largely uncertain because of issues with sensitivity and specificity. The specific mechanisms and physiological meaning of these markers in states of high inflammation are not entirely clear, though there are calcium channelopathies associated with immunological dysfunction, suggesting that there might be some physiological role but it’s not at all clear what that role is (at least to me). Notably, supplementation of patients experiencing critical illness with vitamin D does not appear to have benefit.

Basically from the sum of these studies it’s clear that when a true deficiency exists, vitamin D supplementation is valuable, which makes sense. But it’s important to be mindful that most of the US is vitamin D replete per the IOM, and thus additional benefit is unlikely. The issue is also complicated in more highly melanated individuals because the appropriate physiological range of vitamin D in these individuals is less well established. As far as extraskeletal benefits, the only consistent benefit of supplementation seems to be a reduction in the incidence of upper respiratory infections, though in general the effect seems to be constrained to those with significant vitamin D deficiency.

Toxicity, Deficiency, and Supplementation

Some people seem to have this perception that if a little is good, more must be better. This is definitely not true of nutrients or hormones, and this is especially not true of vitamin D. It is absolutely possible (though generally difficult) to overdose on vitamin D and this can be lethal. Vitamin D status is typically determined by measuring calcidiol (25-OH-D) levels in the blood. Notably, the laboratory assessments for calcidiol levels can vary significantly and have varying sensitivity and specificity, which can complicate determinations, but the IOM established 20 to 40 ng/mL as vitamin D sufficient, and levels below 20 ng/mL are suboptimal for skeletal health (ranges for ideal extraskeletal health are not established, though as above, based on data for supplementation for a number of conditions, it seems there isn’t additional benefit outside the IOM’s therapeutic range). Some have argued that for assessment of vitamin D status is not sufficient to examine just calcidiol levels because at optimal vitamin D levels, parathyroid hormone levels are also suppressed, and because vitamin D levels represent just one aspect of calcium homeostasis, for proper assessment serum calcium, PTH, vitamin D binding protein, and vitamin D should all be assessed at the same time to give an accurate representation of vitamin D status and calcium homeostasis.

Vitamin D deficiency, as above, may occur in states of chronic low grade inflammation and critical illness. However, the classic manifestation of vitamin D deficiency is rickets (in children) and osteomalacia (in adults). There are several forms of these conditions whose etiologies may not depend on vitamin D and so for the sake of simplicity, I will limit the discussion to those in which vitamin D deficiency is the cause. The incidence of rickets declined dramatically after the introduction of cod liver oil and the discovery that UV irradiation was effective prevention. Breastfed infants were particularly at risk (even today vitamin D supplementation is recommended by the American Academy of Pediatrics for breastfed infants; you may note above that the AAP bases its recommendations on the ranges set by the IOM). Rickets due to vitamin D deficiency results from mineralization defects in the bone. In the setting of inadequate vitamin D intake, calcium absorption does not occur at rates necessary to ensure adequate bone mineralization, producing rickets. Additionally, the growth plates (epiphyseal plates) do not close until after puberty, and the state of vitamin D deficiency promotes hypertrophy of the chondrocytes at the mineralization front. This produces a short stature, widened long bone ends, rachitic rosary, and skeletal deformations. Children also frequently become “bow-legged” or get “knocked knees.” Beyond the skeletal abnormalities, children with rickets can develop hypocalcemic seizures, and cardiomyopathy. For the development of rickets, serum levels of 12 ng/mL of calcidiol or lower are generally required. Calcidiol levels may be nearly normal if calcium levels are significantly reduced.

Osteomalacia is essentially the adult form of rickets, occurring after the growth plates have closed. Unlike rickets, there aren’t really dramatic bony abnormalities that can be seen with osteomalacia, but osteomalacia is an important cause of osteopenia, and therefore an increased risk of skeletal fractures. Osteomalacia can also cause bone pains. Because of the role of vitamin D in calcium homeostasis, osteomalacia may also be accompanied by muscle weakness, muscle spasms, gait abnormalities, and even hypocalcemic seizures and tetany. Osteomalacia generally occurs at calcidiol levels < 10 ng/mL.

The pervasive and largely unsupported claim about the prevalence of deficiency has driven many individuals to take to supplementation as a fix. The IOM set a tolerable upper intake of 4000 IU/day in adults, and per Rooney et al., 3% of adults between 1999–2014 had exceeded this. As has hopefully been established, supplementation of vitamin D in a patient who is already replete is unlikely to yield benefits. What, then, are the risks? Firstly, it’s important to point out here that vitamin D toxicity is relatively rare (particularly with the elderly but it is a significant concern for infants and children as in the elderly even very high doses of vitamin D often do not produce signs of toxicity), but when it does occur it can be devastating. At the lower extreme, persistently elevated vitamin D levels can result in increased formation of calcium-based kidney stones, polyuria, thirst, as well as muscle weakness, falls, and increased risk of fractures. However, at very high doses, vitamin D can also cause dangerous hypercalcemia, which can result in coma or death, as well as heart arrhythmias. Dehydration can also result which can produce kidney injury, and patients may even require dialysis as treatment. In general, toxicity is very difficult to achieve by either sun exposure or diet (if not impossible altogether) but supplementation with pharmacologic quantities of vitamin D can cause it. Importantly, some patient cohorts face a disproportionate risk of vitamin D toxicity, particularly those with granulomatous conditions such as sarcoidosis or tuberculosis. In both of these conditions, macrophages may express 1α-hydroxylase to induce rapid conversion of calcidiol to calcitriol. Hypercalcemia results from the combination of increased osteoclast activity and increased calcium absorption from the intestines. Hypercalcemia has also been reported for other mycobacterial infections e.g. leprosy, M. avium complex, and even the BCG vaccine, as well as a variety of fungal infections, cat-scratch disease, and Pneumocystis pneumonia. Noninfectious conditions in which hypercalcemia has been documented also include Wegener’s granulomatosis, Crohn’s disease, infantile subcutaneous fat necrosis, giant cell polymyositis, berylliosis, silicone-induced granuloma, paraffin-associated granulomas, and talc granuloma. For all of these conditions, the cause is inappropriate extrarenal production of calcitriol. Hypercalcemia due to vitamin D can also occur in the setting of lymphoma, but hypercalcemia can also be due to inappropriate expression of PTH-related peptide or other cytokines. For a detailed discussion, see Tebben et al.

Vitamin D in COVID-19

It is often difficult to tell whether vitamin D is an important risk factor for a given disease state because it confounds with so many factors associated with poor overall health. Observations have been made that vitamin D deficient patients are at greater risk for COVID-19, as well as severe outcomes, but because of the retrospective design of such studies, a causal link cannot be established. Most studies are similar to this one (though I found the methods here to be more rigorous than many other retrospective studies), which cites the Martineau meta-analysis discussed above as evidence supporting vitamin D’s role in preventing COVID-19 (though as above, I am skeptical that Martineau’s conclusions can be extended in that manner). I do appreciate this study’s transparency however about its limitations (emphasis mine):

First, vitamin D deficiency may be a consequence associated with a range of chronic health conditions or behavioral factors that plausibly increase COVID-19 risk…

If the observed association were due to confounding by behavioral or other health factors, such associations might have been expected, although our limited sample size might be inadequate to identify such effects…

Another limitation is that only a few individuals received higher doses of vitamin D3 or had relative high vitamin D levels, limiting power to assess whether vitamin D dose or levels are associated with the likelihood of COVID-19

Still, it wouldn’t be correct to suggest that retrospective studies uniformly show effects of vitamin D in COVID-19, as for example Hastie et al showed that after adjustment for confounders there was no difference in the mortality or hospitalization risk from COVID-19 on the basis of patient vitamin D levels. Interestingly, this study even showed a harm to vitamin D supplementation in COVID-19.

In short, there’s not enough evidence to make a conclusion on the basis of these retrospective designs. In situations like these, Mendelian randomization studies can be very instructive, and as it happens, a preprint has done just this with vitamin D levels and COVID-19 risk by examining genes relating to vitamin D function (vitamin D binding protein, 25-hydroxylase (CYP2R1), 7-dehydrocholesterol reductase (DHCR7), and 24-hydroxylase (CYP24A1)) and comparing them to negative controls. The results? No effect was observed. The authors are careful to note that an effect may be present but could be too small to be detectable. Additionally, unfortunately the genetic data being examined vastly overrepresented people of European ancestry, which is a key limitation. Notably they did use Endocrine Society ranges to define deficiency and insufficiency and there was no difference between the deficient and insufficient groups. Mendelian randomization isn’t perfect because a significant portion of the population supplements their vitamin D, and thus this could bias results towards a negative effect (if people who have genetic causes for vitamin D deficiency supplement such that they are no longer deficient, no difference will result in how they do compared with vitamin D replete people assuming that vitamin D were important for COVID-19 outcomes). Still, this study design is superior to most other observational studies because it isn’t subject to many of the confounders that vitamin D levels can change with. Autier et al’s meta-analysis I think demonstrates it quite clearly: low vitamin D levels are a surrogate for chronic illness and ill health generally, and thus it would be expected that these individuals do more poorly with COVID-19.

This area is something that really needs prospective randomized controlled trials to give a definitive answer. I was able to track down this study that attempted to do this, but it is so flawed that I don’t think it’s particularly helpful in answering the questions, also noted here. Another study garnered a lot of attention a while ago but there are issues here too. For one thing, the study calls itself double-blinded and open label. These terms are mutually exclusive. There are significant information gaps in the study regarding what actually happened with the patients and the small sample size makes it difficult to make generalizable conclusions. An extremely thorough analysis of the Cordoba study can be found here.

Paul Sax and I are of the same opinion: this would represent a really nice, easy fix, but I’m inclined to think that once the dust settles, it’s going to show no value for vitamin D in COVID-19. If I thought for a moment that there were good evidence for vitamin D being helpful in COVID-19, I would say as much, but at this point I strongly doubt it. Vitamin D’s role should only be complementary rather than as a therapy itself outside of the clinical trial setting, and certainly not as a replacement to vaccination or any therapies that have proven efficacy e.g. corticosteroids.

Now: please never make me ever have to write about vitamin D again.


  1. Abrams SA, Coss-Bu JA, Tiosano D. 2013. Vitamin D: effects on childhood health and disease. Nat Rev Endocrinol. 9(3):162–170.
  2. Adams JS, Hewison M. 2008. Unexpected actions of vitamin D: new perspectives on the regulation of innate and adaptive immunity. Nat Clin Pract Endocrinol Metab. 4(2):80–90.
  3. Ahmad N, Mohamed Sobaihi M, Al-Jabri M, Al-Esaei NA, Al Zaydi AM. 2018. Acute respiratory failure and generalized hypotonia secondary to vitamin D dependent rickets type 1A. Int J Pediatr Adolesc Med. 5(2):78–81.
  4. Armstrong D. 2018 Sep 26. The Child-Abuse Contrarian. New Yorker. [accessed 2021 Feb 10].
  5. Autier P, Mullie P, Macacu A, Dragomir M, Boniol Magali, Coppens K, Pizot C, Boniol Mathieu. 2017. Effect of vitamin D supplementation on non-skeletal disorders: a systematic review of meta-analyses and randomised trials. Lancet Diabetes Endocrinol. 5(12):986–1004.
  6. Baughman RP, Lower EE. 2014. Goldilocks, vitamin D and sarcoidosis. Arthritis Res Ther. 16(3):111.
  7. Bergman P, Norlin A-C, Hansen S, Rekha RS, Agerberth B, Björkhem-Bergman L, Ekström L, Lindh JD, Andersson J. 2012. Vitamin D3 supplementation in patients with frequent respiratory tract infections: a randomised and double-blind intervention study. BMJ Open. 2(6):e001663.
  8. Bikle D, Christakos S. 2020. New aspects of vitamin D metabolism and action — addressing the skin as source and target. Nat Rev Endocrinol. 16(4):234–252.
  9. Bikle DD. 2014. Vitamin D metabolism, mechanism of action, and clinical applications. Chem Biol. 21(3):319–329.
  10. Bischoff-Ferrari HA, Dawson-Hughes B, Orav EJ, Staehelin HB, Meyer OW, Theiler R, Dick W, Willett WC, Egli A. 2016. Monthly high-dose vitamin D treatment for the prevention of functional decline: A randomized clinical trial. JAMA Intern Med. 176(2):175.
  11. Borel P, Caillaud D, Cano NJ. 2015. Vitamin D bioavailability: state of the art. Crit Rev Food Sci Nutr. 55(9):1193–1205.
  12. Boron WF, Boulpaep EL. 2016. Medical Physiology. 3rd ed. Philadelphia, PA: Elsevier — Health Sciences Division.
  13. Bouillon R. 2017. Comparative analysis of nutritional guidelines for vitamin D. Nature Reviews Endocrinology. 13(8):466–479.
  14. Brown LL, Cohen B, Tabor D, Zappalà G, Maruvada P, Coates PM. 2018. The vitamin D paradox in Black Americans: a systems-based approach to investigating clinical practice, research, and public health — expert panel meeting report. BMC Proc. 12(Suppl 6):6.
  15. Brunton L, Knollman B, Hilal-Dandan R. 2017. Goodman and gilman’s the pharmacological basis of therapeutics, 13th edition. 13th ed. McGraw-Hill Education/Medical.
  16. Camargo CA, Ganmaa D, Frazier AL, Kirchberg FF, Stuart JJ, Kleinman K, Sumberzul N, Rich-Edwards JW. 2012. Randomized trial of vitamin D supplementation and risk of acute respiratory infection in Mongolia. Pediatrics. 130(3):e561-e567.
  17. Cantorna MT, Snyder L, Arora J. 2019. Vitamin A and vitamin D regulate the microbial complexity, barrier function, and the mucosal immune responses to ensure intestinal homeostasis. Crit Rev Biochem Mol Biol. 54(2):184–192.
  18. Carpenter TO, Shaw NJ, Portale AA, Ward LM, Abrams SA, Pettifor JM. 2017. Rickets. Nat Rev Dis Primers. 3(1):17101.
  19. Carroll A. 2016. Why take vitamin D supplements if they don’t improve health? JAMA Health Forum. A5(1). doi:10.1001/jamahealthforum.2016.0013. [accessed 2021 Jan 23].
  20. Cereda E, Bogliolo L, Lobascio F, Barichella M, Zecchinelli AL, Pezzoli G, Caccialanza R. 2021. Vitamin D supplementation and outcomes in coronavirus disease 2019 (COVID-19) patients from the outbreak area of Lombardy, Italy. Nutrition. 82(111055):111055.
  21. cew. 2014 Jun 2. Child Abuse Pediatrics Certification. [accessed 2021 Feb 10].
  22. Christakos S, Dhawan P, Verstuyf A, Verlinden L, Carmeliet G. 2016. Vitamin D: Metabolism, molecular mechanism of action, and pleiotropic effects. Physiol Rev. 96(1):365–408.
  23. Clinical lipidology: A companion to braunwald’s heart disease. 2014. 2nd ed. Philadelphia, PA: Saunders.
  24. Dallas SL, Prideaux M, Bonewald LF. 2013. The osteocyte: an endocrine cell … and more. Endocr Rev. 34(5):658–690.
  25. Deeks ED. 2018. Denosumab: A review in postmenopausal osteoporosis. Drugs Aging. 35(2):163–173.
  26. Degirolamo C, Sabbà C, Moschetta A. 2016. Therapeutic potential of the endocrine fibroblast growth factors FGF19, FGF21 and FGF23. Nat Rev Drug Discov. 15(1):51–69.
  27. Denburg MR, Hoofnagle AN, Sayed S, Gupta J, de Boer IH, Appel LJ, Durazo-Arvizu R, Whitehead K, Feldman HI, Leonard MB, et al. 2016. Comparison of two ELISA methods and mass spectrometry for measurement of vitamin D-binding protein: Implications for the assessment of bioavailable vitamin D concentrations across genotypes: Comparison of Elisa and LC-ms/ms methods for measurement of dbp. J Bone Miner Res. 31(6):1128–1136.
  28. Franceschi C, Garagnani P, Parini P, Giuliani C, Santoro A. 2018. Inflammaging: a new immune-metabolic viewpoint for age-related diseases. Nat Rev Endocrinol. 14(10):576–590.
  29. Ganmaa D, Uyanga B, Zhou X, Gantsetseg G, Delgerekh B, Enkhmaa D, Khulan D, Ariunzaya S, Sumiya E, Bolortuya B, et al. 2020. Vitamin D supplements for prevention of tuberculosis infection and disease. N Engl J Med. 383(4):359–368.
  30. Gittoes NJL. 2016. Vitamin D — what is normal according to latest research and how should we deal with it? Clin Med. 16(2):171–174.
  31. Goldring SR. 2015. The osteocyte: key player in regulating bone turnover. RMD Open. 1(Suppl 1):e000049.
  32. Griffin G, Hewison M, Hopkin J, Kenny R, Quinton R, Rhodes J, Subramanian S, Thickett D. 2020. Vitamin D and COVID-19: evidence and recommendations for supplementation. R Soc Open Sci. 7(12):201912.
  33. Hansdottir S, Monick MM, Lovan N, Powers L, Gerke A, Hunninghake GW. 2010. Vitamin D decreases respiratory syncytial virus induction of NF-kappaB-linked chemokines and cytokines in airway epithelium while maintaining the antiviral state. J Immunol. 184(2):965–974.
  34. Hansen KE, Johnson MG. 2016. An update on vitamin D for clinicians. Curr Opin Endocrinol Diabetes Obes. 23(6):440–444.
  35. Hastie CE, Pell JP, Sattar N. 2021. Vitamin D and COVID-19 infection and mortality in UK Biobank. Eur J Nutr. 60(1):545–548.
  36. Heaney RP, Dowell MS, Hale CA, Bendich A. 2003. Calcium absorption varies within the reference range for serum 25-hydroxyvitamin D. J Am Coll Nutr. 22(2):142–146.
  37. Henderson CM, Fink SL, Bassyouni H, Argiropoulos B, Brown L, Laha TJ, Jackson KJ, Lewkonia R, Ferreira P, Hoofnagle AN, et al. 2019. Vitamin D-binding protein deficiency and homozygous deletion of the GC gene. N Engl J Med. 380(12):1150–1157.
  38. Hewison M. 2011. Antibacterial effects of vitamin D. Nat Rev Endocrinol. 7(6):337–345.
  39. Jameson JL, De Groot LJ. 2015. Endocrinology: Adult and pediatric, 2-volume set. 7th ed. Philadelphia, PA: Saunders.
  40. Jameson JL, Fauci AS, Kasper DL, Hauser SL, Longo DL, Loscalzo J. 2018. Harrison’s Principles of Internal Medicine 20th Edition. 20th ed. Columbus, OH: McGraw-Hill Education.
  41. Jesus JE, Landry A. 2012. Chvostek’s and Trousseau’s Signs. N Engl J Med. 367(11):e15.
  42. Kahwati LC, Weber RP, Pan H, Gourlay M, LeBlanc E, Coker-Schwimmer M, Viswanathan M. 2018. Vitamin D, calcium, or combined supplementation for the primary prevention of fractures in community-dwelling adults: Evidence report and systematic review for the US Preventive Services Task Force. JAMA. 319(15):1600–1612.
  43. Ketha H, Wadams H, Lteif A, Singh RJ. 2015. Iatrogenic vitamin D toxicity in an infant — a case report and review of literature. J Steroid Biochem Mol Biol. 148:14–18.
  44. Kim D, Nguyen QT, Lee J, Lee SH, Janocha A, Kim S, Le HT, Dvorina N, Weiss K, Cameron MJ, et al. 2020. Anti-inflammatory roles of glucocorticoids are mediated by Foxp3+ regulatory T cells via a miR-342-dependent mechanism. Immunity. 53(3):581–596.e5.
  45. Laaksi I, Ruohola J-P, Mattila V, Auvinen A, Ylikomi T, Pihlajamäki H. 2010. Vitamin D supplementation for the prevention of acute respiratory tract infection: a randomized, double-blinded trial among young Finnish men. J Infect Dis. 202(5):809–814.
  46. Lee AH, Dixit VD. 2020. Dietary regulation of immunity. Immunity. 53(3):510–523.
  47. Li H-B, Tai X-H, Sang Y-H, Jia J-P, Xu Z-M, Cui X-F, Dai S. 2016. Association between vitamin D and development of otitis media: A PRISMA-compliant meta-analysis and systematic review. Medicine (Baltimore). 95(40):e4739.
  48. Li YC. 2011. Vitamin D and the Renin-Angiotensin System. In: Vitamin D. Elsevier. p. 707–723.
  49. Maestro MA, Molnár F, Carlberg C. 2019. Vitamin D and its synthetic analogs. J Med Chem. 62(15):6854–6875.
  50. Majak P, Olszowiec-Chlebna M, Smejda K, Stelmach I. 2011. Vitamin D supplementation in children may prevent asthma exacerbation triggered by acute respiratory infection. J Allergy Clin Immunol. 127(5):1294–1296.
  51. Manaseki-Holland S, Qader G, Isaq Masher M, Bruce J, Zulf Mughal M, Chandramohan D, Walraven G. 2010. Effects of vitamin D supplementation to children diagnosed with pneumonia in Kabul: a randomised controlled trial: Vitamin D supplement during childhood pneumonia. Trop Med Int Health. 15(10):1148–1155.
  52. Marchisio P, Consonni D, Baggi E, Zampiero A, Bianchini S, Terranova L, Tirelli S, Esposito S, Principi N. 2013. Vitamin D supplementation reduces the risk of acute otitis media in otitis-prone children. Pediatr Infect Dis J. 32(10):1055–1060.
  53. Mark A. Sperling, MD, Joseph A. Majzoub, MD Ram K. Menon, MD, FRCP Constantine A. Stratakis, MD, D(MED)Sc, PhD(hc). 2021. Sperling Pediatric Endocrinology 5th Edition. Philadelphia: Elsevier.
  54. Martineau AR, Jolliffe DA, Hooper RL, Greenberg L, Aloia JF, Bergman P, Dubnov-Raz G, Esposito S, Ganmaa D, Ginde AA, et al. 2017. Vitamin D supplementation to prevent acute respiratory tract infections: systematic review and meta-analysis of individual participant data. BMJ.:i6583.
  55. Mason RS, Rybchyn MS, Abboud M, Brennan-Speranza TC, Fraser DR. 2019. The role of skeletal muscle in maintaining vitamin D status in winter. Curr Dev Nutr. 3(10):nzz087.
  56. Mazidi M, Rezaie P, Vatanparast H, Kengne AP. 2017. Effect of statins on serum vitamin D concentrations: a systematic review and meta-analysis. Eur J Clin Invest. 47(1):93–101.
  57. Meaning of elevated procalcitonin unclear in COVID-19. 2020 Apr 20. [accessed 2021 Feb 15].
  58. Medicines evidence commentary Commentary on important new evidence from medicines awareness weekly Vitamin D supplementation for preventing intensive care admissions in people with COVID-19 associated pneumonia. 2020 Sep. [accessed 2021 Feb 15].
  59. Meltzer DO, Best TJ, Zhang H, Vokes T, Arora V, Solway J. 2020. Association of vitamin D status and other clinical characteristics with COVID-19 test results. JAMA Netw Open. 3(9):e2019722.
  60. Mendel CM. 1989. The free hormone hypothesis: a physiologically based mathematical model. Endocr Rev. 10(3):232–274.
  61. Moriyama M, Hugentobler WJ, Iwasaki A. 2020. Seasonality of respiratory viral infections. Annu Rev Virol. 7(1):83–101.
  62. &na; 2007. Dangers of indoor tanning. Nursing. 37(11):66.
  63. Naot D, Musson DS, Cornish J. 2019. The activity of peptides of the calcitonin family in bone. Physiol Rev. 99(1):781–805.
  64. National Heart, Lung, and Blood Institute PETAL Clinical Trials Network, Ginde AA, Brower RG, Caterino JM, Finck L, Banner-Goodspeed VM, Grissom CK, Hayden D, Hough CL, Hyzy RC, et al. 2019. Early high-dose vitamin D3 for critically ill, vitamin D-deficient patients. N Engl J Med. 381(26):2529–2540.
  65. Newman CB, Preiss D, Tobert JA, Jacobson TA, Page RL 2nd, Goldstein LB, Chin C, Tannock LR, Miller M, Raghuveer G, et al. 2019. Statin safety and associated adverse events: A scientific statement from the American heart association. Arterioscler Thromb Vasc Biol. 39(2):e38-e81.
  66. Nielsen R, Christensen EI, Birn H. 2016. Megalin and cubilin in proximal tubule protein reabsorption: from experimental models to human disease. Kidney Int. 89(1):58–67.
  67. Norman PE, Powell JT. 2014. Vitamin D and cardiovascular disease. Circ Res. 114(2):379–393.
  68. de Oliveira C, Biddulph JP, Hirani V, Schneider IJC. 2017. Vitamin D and inflammatory markers: cross-sectional analyses using data from the English Longitudinal Study of Ageing (ELSA). J Nutr Sci. 6(e1):e1.
  69. Orimo H. 2010. The mechanism of mineralization and the role of alkaline phosphatase in health and disease. J Nippon Med Sch. 77(1):4–12.
  70. Pachter L. 2020 Nov 17. Mathematical analysis of “mathematical analysis” of a vitamin D COVID-19 trial. [accessed 2021 Feb 17].
  71. Passeron T, Bouillon R, Callender V, Cestari T, Diepgen TL, Green AC, van der Pols JC, Bernard BA, Ly F, Bernerd F, et al. 2019. Sunscreen photoprotection and vitamin D status. Br J Dermatol. 181(5):916–931.
  72. ‘Patchen BK, Clark AG, Hancock DB, Gaddis N, Cassano PA. 2021. Genetically predicted serum vitamin D and COVID-19: a Mendelian randomization study. bioRxiv. doi:10.1101/2021.01.29.21250759.
  73. Pham H, Waterhouse M, Baxter C, Duarte Romero B, McLeod DSA, Armstrong BK, Ebeling PR, English DR, Hartel G, Kimlin MG, et al. 2021. The effect of vitamin D supplementation on acute respiratory tract infection in older Australian adults: an analysis of data from the D-Health Trial. Lancet Diabetes Endocrinol. 9(2):69–81.
  74. Pike JW, Christakos S. 2017. Biology and mechanisms of action of the vitamin D hormone. Endocrinol Metab Clin North Am. 46(4):815–843.
  75. Pilz S, Verheyen N, Grübler MR, Tomaschitz A, März W. 2016. Vitamin D and cardiovascular disease prevention. Nat Rev Cardiol. 13(7):404–417.
  76. Rak K, Bronkowska M. 2018. Immunomodulatory effect of vitamin D and its potential role in the prevention and treatment of type 1 diabetes mellitus-A narrative review. Molecules. 24(1):53.
  77. Rooney MR, Harnack L, Michos ED, Ogilvie RP, Sempos CT, Lutsey PL. 2017. Trends in use of high-dose vitamin D supplements exceeding 1000 or 4000 international units daily, 1999–2014. JAMA. 317(23):2448–2450.
  78. Rosen CJ, Taylor CL. 2013. Common misconceptions about vitamin D-implications for clinicians. Nat Rev Endocrinol. 9(7):434–438.
  79. Ross AC, Taylor CL, Yaktine AL, Del Valle HB, editors. 2011. Dietary reference intakes for calcium and vitamin D. Washington, D.C., DC: National Academies Press.
  80. Rubin R. 2021. Sorting out whether vitamin D deficiency raises COVID-19 risk. JAMA. 325(4):329–330.
  81. Santos RAS, Sampaio WO, Alzamora AC, Motta-Santos D, Alenina N, Bader M, Campagnole-Santos MJ. 2018. The ACE2/angiotensin-(1–7)/MAS axis of the renin-angiotensin system: Focus on angiotensin-(1–7). Physiol Rev. 98(1):505–553.
  82. Shah A, Aeddula NR. 2020. Renal Osteodystrophy. In: StatPearls. Treasure Island (FL): StatPearls Publishing.
  83. Shoback D, Rosen CJ, Black DM, Cheung AM, Murad MH, Eastell R. 2020. Pharmacological Management of osteoporosis in Postmenopausal Women: An endocrine society guideline update. J Clin Endocrinol Metab. 105(3):587–594.
  84. Silva MC, Furlanetto TW. 2018. Intestinal absorption of vitamin D: a systematic review. Nutr Rev. 76(1):60–76.
  85. Sparks MA, Crowley SD, Gurley SB, Mirotsou M, Coffman TM. 2014. Classical Renin-Angiotensin system in kidney physiology. Compr Physiol. 4(3):1201–1228.
  86. Stipanuk MH, Caudill MA. 2012. Biochemical, physiological, and molecular aspects of human nutrition. 3rd ed. London, England: W B Saunders.
  87. Szabo L. 2018 Aug 20. The man who sold America on vitamin D — and profited in the process. [accessed 2021 Feb 10].
  88. Tangpricha V. Vitamin D Deficiency and Related Disorders. [accessed 2021 Feb 17].
  89. Taylor CL, Thomas PR, Aloia JF, Millard PS, Rosen CJ. 2015. Questions about vitamin D for Primary Care practice: Input from an NIH conference. Am J Med. 128(11):1167–1170.
  90. Tebben PJ, Singh RJ, Kumar R. 2016. Vitamin D-mediated hypercalcemia: Mechanisms, diagnosis, and treatment. Endocr Rev. 37(5):521–547.
  91. Tripkovic L, Lambert H, Hart K, Smith CP, Bucca G, Penson S, Chope G, Hyppönen E, Berry J, Vieth R, et al. 2012. Comparison of vitamin D2 and vitamin D3 supplementation in raising serum 25-hydroxyvitamin D status: a systematic review and meta-analysis. Am J Clin Nutr. 95(6):1357–1364.
  92. Vaeth M, Feske S. 2018. Ion channelopathies of the immune system. Curr Opin Immunol. 52:39–50.
  93. Vitamin D. [accessed 2021a Feb 10].
  94. Vitamin D & iron supplements for babies: AAP recommendations. [accessed 2021b Feb 17].
  95. Voet D, Voet JG. 2010. Biochemistry 4th Edition. Chichester, England: John Wiley & Sons.
  96. Vogiatzi MG, Jacobson-Dickman E, DeBoer MD, Drugs, and Therapeutics Committee of The Pediatric Endocrine Society. 2014. Vitamin D supplementation and risk of toxicity in pediatrics: a review of current literature. J Clin Endocrinol Metab. 99(4):1132–1141.
  97. Wade KH, Hall LJ. 2020. Improving causality in microbiome research: can human genetic epidemiology help? Wellcome Open Res. 4(199):199.
  98. Waldron JL, Ashby HL, Cornes MP, Bechervaise J, Razavi C, Thomas OL, Chugh S, Deshpande S, Ford C, Gama R. 2013. Vitamin D: a negative acute phase reactant. J Clin Pathol. 66(7):620–622.
  99. Wein MN, Kronenberg HM. 2018. Regulation of bone remodeling by parathyroid hormone. Cold Spring Harb Perspect Med. 8(8). doi:10.1101/cshperspect.a031237.
  100. Yavuz B, Ertugrul DT. 2012. Statins and vitamin D: A hot topic that will be discussed for a long time: A hot topic that will be discussed for a long time. Dermatoendocrinol. 4(1):8–9.
  101. Zimmerman L, McKeon B. 2020. Osteomalacia. In: StatPearls. Treasure Island (FL): StatPearls Publishing.
  102. Neale RE, Khan SR, Lucas RM, Waterhouse M, Whiteman DC, Olsen CM. 2019. The effect of sunscreen on vitamin D: a review. Br J Dermatol. 181(5):907–915
  103. Vitamin D. [accessed 2021 Apr 24].

Originally published at on February 17, 2021.



Edward Nirenberg

I write about vaccines here. You can find me on Twitter @enirenberg and at (where I publish the same content without a paywall)