Vitamin D and the immune system: what is proven and where the evidence gets thin
Many immune cells carry a receptor for vitamin D, and some even activate it themselves. That is well-studied cell biology. Whether extra vitamin D makes infections or autoimmune diseases less common is far more open than it often sounds.
Vitamin D is not a shield and not a booster. It is a building block that many of your immune cells respond to. If it is missing, that can matter. If there is enough, more of it brings no additional protection according to current data.
When you should act instead of reading on
- Seek medical help immediately if you have shortness of breath, chest pain, a high fever lasting several days or confusion, and in young children also if they drink poorly or are unusually drowsy. In acute danger: 112 (the German emergency number).
- Very frequent or unusually severe infections need a medical examination, including a check for antibody deficiency.
- Think of too much calcium in the blood if you take high doses of vitamin D and develop confusion, repeated vomiting, abdominal pain, severe thirst or pass a lot of urine (Marcinowska-Suchowierska 2018). In that case, have it checked by a doctor right away.
Prescribed medication such as cortisone, immunosuppressants, MS immunotherapies, thyroid hormones or asthma inhalers is never changed on your own because of vitamin D. Any change belongs in medical hands.
The third infection since October. The cough has barely gone and your throat is scratchy again. And in the evening you type into the search bar: vitamin D immune system. Many people know this pattern, and the question is a fair one.
A team around Ginde analysed a US population survey with 18,883 participants aged 12 and over. A recent respiratory infection was reported by 24 percent of those with levels below 10 ng/ml, 20 percent of those with 10 to under 30 ng/ml and 17 percent of those with 30 ng/ml or more. For you this means: low levels and infections occur together, but a snapshot cannot tell whether deficiency favours infections or whether sick people simply spend less time outdoors.
Ginde et al. 2009 · DOI: 10.1001/archinternmed.2008.560 [Cross-sectional, n=18,883]This is the thread running through this article: a convincing cell biology and studies whose protective effect became smaller with every larger analysis. The two fit together if you look closely. Why the level drops in winter is explained in Vitamin D deficiency in winter, and the balance between sun and skin protection in How much sun does the body need.
What to expect here
- What the vitamin D receptor does in immune cells
- How protection against infection shrank from 2017 to 2025
- What VITAL, D-Health and CORONAVIT tested
- Autoimmunity, MS, Hashimoto's, gut and asthma
- Why cut-off values differ
- Hypercalcaemia, falls and why more is not better
What the vitamin D receptor does in your immune cells
Imagine your immune system as a network of guardhouses. In each one sits a crew that listens for signals. Vitamin D is one of them.
The receiver for it is the vitamin D receptor inside the cell. When active vitamin D binds, it can switch genes on or off. A review around Baeke lists the cells that listen in: monocytes, macrophages, dendritic cells, and T and B lymphocytes. And many of these cells have their own enzymes to activate vitamin D on the spot.
Baeke et al. 2010 · DOI: 10.1016/j.coph.2010.04.001 [Mechanism Review]
The guardhouse's own workbench
In the blood, vitamin D circulates as a storage form, 25-OH vitamin D. It is switched to its active form by the enzyme 1-hydroxylase, classically in the kidney. Macrophages, the scavenger cells of the immune defence, can do this themselves.
A team around Liu activated human macrophages via sensors for bacterial components, so-called Toll-like receptors. The cells built up vitamin D receptor and 1-hydroxylase, produced the antimicrobial peptide cathelicidin and killed tuberculosis bacteria inside the cell, while serum with low 25-OH-D supported cathelicidin production poorly. For you this means: the macrophage takes the raw material from the blood and activates it itself, but this is a laboratory finding and no proof that tablets prevent tuberculosis.
Liu et al. 2006 · DOI: 10.1126/science.1123933 [In vitro]Cathelicidin is something like a body's own antibiotic. A team around Wang found a direct docking site for the activated vitamin D receptor in its gene. A team around Gombart showed that this site is conserved in primates and missing in mice, rats and dogs, so mouse experiments are of limited use here.
Wang et al. 2004 · DOI: 10.4049/jimmunol.173.5.2909 [In vitro] · Gombart et al. 2005 · DOI: 10.1096/fj.04-3284com [In vitro]
The brake in the T cells
A team around Jeffery stimulated isolated human T helper cells in the presence of active vitamin D. The cells produced less IFN-gamma, IL-17 and IL-21 but divided hardly any differently, and they produced CTLA-4 and, together with IL-2, also FoxP3, features of regulatory T cells. For you this means: the same substance that can sharpen the defence in laboratory experiments with macrophages took its foot off the accelerator in these T cells.
Jeffery et al. 2009 · DOI: 10.4049/jimmunol.0803217 [In vitro]Two cells, two tasks
Findings from human cells in the laboratory. The concentrations used there do not match what reaches the blood after a tablet.
Everything in this section comes from cell culture and gene analyses. It makes a role for vitamin D in the immune system plausible. It does not answer whether usual intake amounts flip the same switches in a living person, or how large this role is in everyday life.
Through the lens of psychoneuroimmunology this fits the picture: a regulator of calcium metabolism can also help steer immune cells. Metabolism and the immune system are closely interwoven, and here that is an interpretation, not a study finding.
Going by the laboratory findings, vitamin D does not so much turn the immune system up as it may help it keep time. More defence against pathogens and less excessive inflammation would then be two sides of the same regulation. More of a conductor than an amplifier, meant as an image and not as proof.
And now you know why the receptor made researchers curious, and why cell biology alone tells you nothing yet about your winter.
Vitamin D and infections: the story of a shrinking effect
With biology this coherent, you would expect clear protection in studies. The story went differently, and that is exactly what makes it instructive.
The 2017 analysis used individual participant data: not the grade averages of 25 classes, but every single exercise book. An odds ratio below 1.00 means a lower chance of an infection.
A team around Martineau pooled the raw data of 10,933 participants from 25 placebo-controlled trials. For at least one acute respiratory infection, the odds ratio was 0.88 overall. With daily or weekly dosing without additional boluses it was 0.81, with one or more large single doses, so-called boluses, 0.97 without detectable protection. Within daily or weekly dosing it was 0.30 with baseline levels below 25 nmol/L and 0.75 at 25 nmol/L or above. For you this means: the overall effect was small, and the 0.30 comes from a subgroup within a subgroup, not a promise of protection for everyone.
Martineau et al. 2017 · DOI: 10.1136/bmj.i6583 [Meta-analysis of RCTs, k=25, n=10,933]A team around Jolliffe from the same group updated the analysis with 46 included trials and 75,541 participants. In the main analysis of 37 trials, 61.3 percent on vitamin D and 62.3 percent on placebo had at least one respiratory infection, odds ratio 0.92, and no subgroup by baseline level showed a significant effect. In trials with daily dosing the odds ratio was 0.78, but there was no significant interaction with dosing frequency. For you this means: deficiency as an amplifier of the effect could no longer be confirmed here.
Jolliffe et al. 2021 · DOI: 10.1016/S2213-8587(21)00051-6 [Meta-analysis of RCTs, k=46, n=75,541]In 2025 the team around Jolliffe presented a second update with 40 trials and 61,589 participants. The odds ratio was 0.94 with a confidence interval of 0.88 to 1.00 and p = 0.057, with no influence of age, baseline level, dosing frequency or dose size, and the funnel plot was asymmetric. For you this means: the group that described the protection in 2017 now states itself that it is no longer statistically confirmed.
Jolliffe et al. 2025 · DOI: 10.1016/S2213-8587(24)00348-6 [Meta-analysis of RCTs, k=40, n=61,589]Our article on winter deficiency cites 2017 and 2021 with a small protective effect, and 2025 continues this line: a similar direction, but no longer statistically confirmed.
A team around Autier split the 2021 infection trials by size. Trials with 25 to under 248 participants came to an odds ratio of 0.69, trials with 248 to 16,000 participants to 0.98. For you this means: the protection showed up mainly in small trials, which, however, may also more often study people with a more severe deficiency.
Autier et al. 2025 · DOI: 10.1371/journal.pone.0303316 [Meta-analysis of RCTs]The shrinking protection curve
Important for reading this: 2017 is an adjusted odds ratio from individual participant data, 2021 and 2025 are pooled odds ratios from trial-level data. The same endpoint, at least one acute respiratory infection, but different methods. The points show a direction, not an exact comparison.
A Cochrane team around van Arragon analysed 107 trials of vitamin D given in pregnancy or early childhood. The proportion of children with a doctor visit for a respiratory infection was slightly lower, relative risk 0.95, with low certainty of evidence, and a higher dose brought no recognisable advantage. For you this means: the possible effect is small, and vitamin D in infants and children should be discussed with the paediatrician.
van Arragon et al. 2026 · DOI: 10.1002/14651858.CD015111.pub2 [Systematic Review]A team around Bergman gave 140 people with antibody deficiency or more than four bacterial respiratory infections per year either 4000 IU vitamin D3 daily or placebo for one year (study data). The infection score was 202 versus 249 points, and the authors themselves name a single centre, a small sample and a selected group as limitations. For you this means: a small signal in people who are constantly ill, but frequent infections need a medical work-up first.
Bergman et al. 2012 · DOI: 10.1136/bmjopen-2012-001663 [RCT, n=140]The 2024 Endocrine Society guideline suggests vitamin D for 1 to 18 year olds, partly because of its possible influence on respiratory infections. For healthy adults under 75, it advises against supplementing above the current intake reference values in order to prevent disease. Both are conditional recommendations (grade 2), with low certainty of evidence for children and very low to moderate certainty for adults. Both explicitly apply only to people without an established indication for vitamin D treatment or for measuring the level.
Demay et al. 2024 · DOI: 10.1210/clinem/dgae290 [Guideline]
Vitamin D as a high-dose burst at the first hint of a sore throat is widespread. Studies on exactly this use are missing, and the data on large single doses tend to speak against it: bolus doses showed no detectable protection in 2017.
An effect that shrinks is not a scandal. This is how science works: small trials often see large effects, large trials bring them back down to earth. A research group that re-evaluates its own number shows exactly the care you can wish for.
And now you know why the same question was answered differently in 2017 than in 2025, and why both answers are honest.
VITAL, D-Health, CORONAVIT: disproven or narrowed down?
Headlines like “Vitamin D does nothing” usually refer to these three trials. What matters is who they studied.
Supported by large RCTs and meta-analysesA team around Manson gave 25,871 men aged 50 and over and women aged 55 and over in the USA either 2000 IU vitamin D3 daily or placebo for a median of 5.3 years (study data). Cancer occurred in 793 versus 824 people, hazard ratio 0.96, major cardiovascular events did not become less frequent, and there was no excess of hypercalcaemia. For you this means: according to the full text of the autoimmune analysis, the baseline level in participants with a blood sample averaged 30.7 ng/ml, and only 12.9 percent were below 20 ng/ml, so these were mostly well-supplied older adults.
Manson et al. 2019 · DOI: 10.1056/NEJMoa1809944 [RCT, n=25,871]A team around Pham analysed the Australian D-Health trial, in which 21,315 people aged 60 to 79, volunteers up to 84, received 60,000 IU vitamin D or placebo monthly for up to five years (study data). The infection risk did not fall, odds ratio 0.98, and the placebo group was at 77.5 nmol/L during the trial. For you this means: the authors describe their participants as largely sufficiently supplied, so what was tested was a monthly high dose in well-supplied people.
Pham et al. 2021 · DOI: 10.1016/S2213-8587(20)30380-6 [RCT, n=21,315]A team around Jolliffe offered people in the United Kingdom a finger-prick blood test, and those below 75 nmol/L received 800 or 3200 IU daily for six months (study data), compared with a group without an offer. In the offer groups, 86.3 percent were below this threshold. At least one respiratory infection occurred in 4.6 percent without an offer and in 5.7 and 5.0 percent with the lower and higher dose, and Covid-19 did not become less frequent either. For you this means: CORONAVIT tested a strategy for many people in an open-label design, not supplementation of well-supplied people and not the targeted treatment of a severe deficiency.
Jolliffe et al. 2022 · DOI: 10.1136/bmj-2022-071230 [RCT, n=6,200]Three reasons why these null trials narrow the hypothesis down
- People who already have enough. VITAL and D-Health mainly studied well-supplied people. A full tank can hardly be made fuller by topping up, at least that is the obvious interpretation.
- Bursts instead of rhythm. D-Health gave monthly high doses, and bolus doses had already shown no protection in 2017. This still remains open, because in 2025 dosing frequency made no difference.
- A strategy instead of targeted treatment. CORONAVIT tested an offer for many people, not the treatment of a severe deficiency diagnosed by a doctor.
The opposite direction belongs here too: the analyses from 2021 and 2025 did not confirm that people with a severe deficiency in particular benefit. Plausible, but open.
A null trial does not say: vitamin D does not matter. It says: in these people, with this form of dosing and for this question, no benefit could be shown. That narrows down hope and may protect against risks without additional benefit.
And now you know why the same trials prove failure for some people and draw a useful boundary for others.
Autoimmune diseases: what VITAL showed and what came after
In an autoimmune disease, the immune defence turns against your own body. If vitamin D dampens inflammatory T cells in the laboratory, do such diseases then occur less often?
Signal from an RCT, narrow and not lastingWithin VITAL, a team around Hahn confirmed all new autoimmune diseases from medical records, at a mean age of 67.1 years and a median follow-up of 5.3 years. A confirmed autoimmune disease occurred in 123 people on vitamin D and 155 on placebo, hazard ratio 0.78, interval 0.61 to 0.99, P = 0.05. For you this means: a large placebo-controlled trial with fewer new autoimmune diseases on vitamin D, just at the threshold of significance and in older people.
Hahn et al. 2022 · DOI: 10.1136/bmj-2021-066452 [RCT]The full text contains a preplanned time-window analysis: excluding the first two years, the hazard ratio was 0.61, and in the last three years the vitamin D group had 39 percent fewer people affected. That is not the main result.
A team around Costenbader followed 21,592 participants for two more years after they stopped taking the supplements. After seven years, the hazard ratio for vitamin D was 0.98. With 65 cases from the intake period confirmed later, the result for the trial period also shifted to 0.85 with an interval of 0.70 to 1.04, so no longer significant. Omega-3 was at 0.83 after seven years with an interval of 0.70 to 0.99. For you this means: the advantage of vitamin D did not last beyond stopping.
Costenbader et al. 2024 · DOI: 10.1002/art.42811 [RCT, n=21,592]How omega-3 can be measured in the blood is explained in Measuring the omega-3 index.
These data do not mean that vitamin D prevents autoimmune diseases. They mean that a connection is plausible and needs further testing. The signal was narrow, disappeared after stopping, comes from people aged 50 and over and showed no protection for the thyroid. The authors themselves note that it may not apply to diseases that usually begin in younger people.
My interpretation, not a study finding: an effect that disappears after stopping points more towards ongoing support than towards permanent reprogramming. Vitamin D would then be part of daily supply rather than a one-off impulse.
And now you know why the 2022 headline and the 2024 follow-up only make sense together.
MS, Hashimoto's, gut and asthma: four fields, four levels of evidence
“Vitamin D lowers antibodies.” “Vitamin D does nothing.” If you live with a chronic disease, you read both. Often there is a true core, and a boundary is missing.
| Condition | Observation | Intervention studies | Status |
|---|---|---|---|
| Multiple sclerosis | Human Low levels before diagnosis, genetically supported | Clinical Less MRI activity, relapses not significant | DGN: check level, make up for deficiency |
| Hashimoto's | Human Lower levels | Clinical Antibodies fall, function unchanged | Open |
| Crohn's, colitis | Human Deficiency more common | Clinical CRP somewhat lower, activity unchanged | Open |
| Asthma | Human Closer link with infections | Clinical Positive in 2017, Cochrane 2023 without effect | Open in severe deficiency |
A reading aid, not a treatment overview.
Multiple sclerosis: strong signals, a cautious guideline
A team around Munger compared stored blood samples from 257 US military personnel who later developed MS with controls. Among white participants, the odds ratio per 50 nmol/L higher 25-OH-D was 0.59, while among Black and Hispanic participants no significant association was found. For you this means: the level was measured before symptoms, so reverse causation is less likely here.
Munger et al. 2006 · DOI: 10.1001/jama.296.23.2832 [Cohort, nested case-control]A team around Mokry used four gene variants that lower the level slightly for life as a natural experiment. Per genetically determined decrease of one standard deviation, the odds of MS were 2.0-fold higher, and 1.7-fold after excluding possible side effects of the variants. For you this means: a stronger indication of a causal role, but no proof that tablets in adulthood prevent MS.
Mokry et al. 2015 · DOI: 10.1371/journal.pmed.1001866 [Mendelian Randomisation]In established MS, a Cochrane review around Jagannath found no benefit for relapses, disability or contrast-enhancing lesions across 12 trials with 933 people, with very low quality of evidence.
Jagannath et al. 2018 · DOI: 10.1002/14651858.CD008422.pub3 [Systematic Review]
A team around Thouvenot gave 316 untreated people with early clinically isolated syndrome at 36 French centres either 100,000 IU cholecalciferol or placebo every two weeks for 24 months (study data). Disease activity occurred in 60.3 versus 74.1 percent, hazard ratio 0.66, also with fewer MRI lesions, but all ten clinical secondary endpoints showed no significant difference, including relapses. For you this means: less activity on imaging, no detectable difference in relapses.
Thouvenot et al. 2025 · DOI: 10.1001/jama.2025.1604 [RCT, n=316]The 2026 S2k guideline on multiple sclerosis recommends, with strong consensus, checking the level and making up for a deficiency, ideally with daily or weekly dosing. Even with normal levels, supplementation up to the high-normal range can be considered, with the explanation that a positive effect has not been proven. According to the guideline, the D-Lay MS data do not allow a clear conclusion. Ultra-high-dose therapies should not be given, and for the Coimbra protocol, for which efficacy studies are lacking, the guideline states that risks cannot be ruled out.
In MS, vitamin D does not replace immunotherapy. Any supplementation should be agreed with the treating neurologist.
Hemmer et al. 2026, DGN S2k guideline MS, no DOI [Guideline]
Hashimoto's: antibodies are not the same as function
A meta-analysis around Wang of 20 case-control studies found lower 25-OH-D levels and more frequent deficiency in autoimmune thyroid disease, odds ratio 2.99. Whether this is cause, consequence or a shared background remains open.
Wang et al. 2015 · DOI: 10.3390/nu7042485 [Meta-analysis of observational studies, k=20]
A meta-analysis around Jiang pooled six clinical trials with 258 people with Hashimoto's thyroiditis. TPO antibodies fell on vitamin D, with wide variation, while TSH, free T3 and free T4 did not change significantly. For you this means: a falling antibody level is not the same as better thyroid function.
Jiang et al. 2022 · DOI: 10.1111/jcpt.13605 [Meta-analysis of RCTs, k=6, n=258]VITAL fits with this: according to the full text, autoimmune thyroid disease occurred in 21 people on vitamin D and 11 on placebo, hazard ratio 1.63, not significant. No protection for the thyroid can be read from this. The dose of a thyroid hormone is never changed on your own because of vitamin D or an antibody level. More on the condition in Understanding Hashimoto's thyroiditis, and on the antibodies in Lowering Hashimoto's antibodies.
Crohn's disease and ulcerative colitis: the level rises, the course stays the same
In 72,719 women, a team around Ananthakrishnan estimated vitamin D status using a prediction score: the highest compared with the lowest quarter was associated with a hazard ratio of 0.54 for Crohn's disease, and without significance for ulcerative colitis. A meta-analysis around Del Pinto found an odds ratio of 1.64 for deficiency in IBD, not adjusted. Inflammation, impaired absorption and cortisone may partly explain this.
Ananthakrishnan et al. 2012 · DOI: 10.1053/j.gastro.2011.11.040 [Cohort, n=72,719] · Del Pinto et al. 2015 · DOI: 10.1097/MIB.0000000000000546 [Meta-analysis of observational studies]
A meta-analysis around Guo pooled 17 studies with 1127 patients. The level rose and CRP fell somewhat, while erythrocyte sedimentation rate, activity index and relapse rate did not change significantly. For you this means: the lab value moves, the disease activity in these data does not detectably.
Guo et al. 2021 · DOI: 10.1039/d1fo00613d [Meta-analysis, k=17, n=1127]An integrative perspective, complementary to gastroenterological treatment, can be found in Crohn's disease and ulcerative colitis, integrative.
Asthma: a finding that disappeared with more studies
In 2017, a team around Jolliffe had calculated from individual data of 955 people fewer asthma attacks requiring cortisone as tablets or intravenously, adjusted incidence rate ratio 0.74. A Cochrane team around Williamson, with Martineau and Jolliffe as co-authors, then pooled 20 studies in 2023 and found, for such attacks, 226 per 1000 on vitamin D versus 219 per 1000 on placebo, odds ratio 1.04, with high certainty of evidence, although severe deficiency below 25 nmol/L was rare. For you this means: the asthma effect disappeared with more data, and only severe deficiency remains open.
Jolliffe et al. 2017 · DOI: 10.1016/S2213-2600(17)30306-5 [Meta-analysis of RCTs, n=955] · Williamson et al. 2023 · DOI: 10.1002/14651858.CD011511.pub3 [Systematic Review]Asthma inhalers and other prescribed asthma medication are never reduced because of vitamin D. If you are considering diet as a further pillar in autoimmunity, read Autoimmune protocol and paleo, there too as a complement, not a replacement.
A value that moves is not yet a course that improves. In Hashimoto's and IBD, lab values moved, in MS the imaging did, while thyroid function, bowel disease activity and relapses were not detectably different.
And now you know why there is not one answer on vitamin D and autoimmunity, but four.
The lab value: why 20, 30 or no cut-off at all, and why inflammation can lower it
You are holding your lab report: 25-OH vitamin D, a number, a reference range. Online you read a completely different cut-off. Which one is right?
What is measured is the storage form in the blood, not the active form. Unit, measurement method and reference range depend on the laboratory. That is why there is no table to read off here, but rather the question of why professional societies draw different lines.
Bone-related
20 ng/ml, 50 nmol/lInstitute of Medicine 2011: for autoimmune diseases and other endpoints outside the bone, the evidence was insufficient. The German Nutrition Society (DGE) names 50 nmol/l or more as an indicator of optimal supply.
Beyond the bone
30 ng/ml, 75 nmol/lEndocrine Society 2011: deficiency below 20 ng/ml, insufficiency 21 to 29 ng/ml, target above 30 ng/ml. Pludowski assigns this value to guidelines that look beyond the bone.
No target for healthy people
no cut-offEndocrine Society 2024: no clear evidence for an optimal target value, no routine testing. Explicitly not where there is an established indication for treatment or measurement.
Disease-specific
50 to 125 nmol/lTarget range of the DGN guideline for people with MS, not a cut-off for everyone.
Ross et al. 2011 · DOI: 10.1210/jc.2010-2704 [Reference value] · DGE 2012 · DOI: 10.1159/000337547 [Reference value] · Holick et al. 2011 · DOI: 10.1210/jc.2011-0385 [Guideline] · Pludowski et al. 2018 · DOI: 10.1016/j.jsbmb.2017.01.021 [Review]
The different cut-offs are not a measurement error but answers to different questions: What does the bone need? What might matter beyond that? What is proven for healthy people?
The fuel gauge on a hill
When a car drives steeply uphill, the fuel gauge drops even though there is as much in the tank as in the valley. For the vitamin D value there are indications of something similar.
A team around Waldron measured values in 30 people before and 48 hours after a knee or hip replacement. CRP rose on average from 5.0 to 116.0 mg/L, and 25-OH-D fell from 56.2 to 46.0 nmol/L. For you this means: the value dropped without any change in sun exposure or intake, and the authors consider a low status in chronic inflammation more likely to be a possible consequence than a cause.
Waldron et al. 2013 · DOI: 10.1136/jclinpath-2012-201301 [Prospective human study, n=30]A review around Silva found a drop in 25-OH-D after an inflammatory event in 6 of 8 longitudinal studies. A review around Autier set 290 cohorts against 172 randomised trials: low levels went along with inflammation, infections and MS, but in the randomised trials vitamin D did not change how often these diseases occurred. The authors therefore interpret low 25-OH-D as a possible marker of disease, an explanatory model and not a rule. In MS, the genetic study around Mokry points more towards a causal role, without proving it.
Silva and Furlanetto 2015 · DOI: 10.1016/j.nutres.2014.12.008 [Systematic Review] · Autier et al. 2014 · DOI: 10.1016/S2213-8587(13)70165-7 [Systematic Review]
Please do not interpret a vitamin D value yourself and do not derive a treatment from it. If it was measured in the middle of an infection, after surgery or during an inflammatory flare, discuss the timing of the measurement with your doctor.
Clinically, I observe two patterns: many people with recurrent infections have never had their level measured, and in others it was measured during the week of fever. This is experience, not a study finding. What a broader measurement can look like is described in Micronutrient analysis in whole blood, and a second example of inflammation and lab values in Iron and inflammation.
And now you know why the same number can mean something different depending on the guideline, the laboratory and the timing of the measurement.
Why more is not better: hypercalcaemia, falls and upper limits
If a little is good, a lot must be better? With vitamin D, that can lead into a dead end. This is covered in detail in Vitamin D overdose and high-dose therapy, and here are only the key figures.
A team around Sanders gave 2256 community-dwelling women aged 70 and over with increased fracture risk a single dose of 500,000 IU cholecalciferol or placebo every autumn or winter (study data). On vitamin D there were 83.4 versus 72.7 falls per 100 person-years, rate ratio 1.15, fractures were also more frequent, ratio 1.26, and in a post hoc analysis the falls ratio in the first three months after the dose was 1.31. For you this means: the very high single dose was associated with more falls and fractures, most clearly shortly after the dose.
Sanders et al. 2010 · DOI: 10.1001/jama.2010.594 [RCT, n=2256]In Zurich, a team around Bischoff-Ferrari gave 200 people aged 70 and over, after a fall, monthly doses of 24,000 IU, 60,000 IU or 24,000 IU plus calcifediol for one year (study data). Leg function did not differ between the groups, and 66.9 percent fell on 60,000 IU and 66.1 percent on 24,000 IU plus calcifediol, compared with 47.9 percent on 24,000 IU. For you this means: the groups that were more likely to reach higher lab values fell more often.
Bischoff-Ferrari et al. 2016 · DOI: 10.1001/jamainternmed.2015.7148 [RCT, n=200]In Calgary, 311 healthy adults aged 55 to 70 received 400, 4000 or 10,000 IU daily for three years (study data). Bone density at the radius decreased in this order by 1.2, 2.4 and 3.5 percent, and bone strength did not differ. The authors around Burt see no benefit, and whether it causes harm remains open.
Burt et al. 2019 · DOI: 10.1001/jama.2019.11889 [RCT, n=311]
Upper limit and signs of overdose
In 2023 the European Food Safety Authority set a tolerable upper intake level of 100 µg vitamin D per day for adults and adolescents aged 11 and over, and 50 µg for children aged 1 to 10. The critical endpoint was persistently increased calcium excretion in the urine. This upper limit is a safety limit, not a target.
EFSA 2023 · DOI: 10.2903/j.efsa.2023.8145 [Regulatory document]
A clinical review around Marcinowska-Suchowierska names confusion, apathy, repeated vomiting, abdominal pain, passing a lot of urine, severe thirst and dehydration as the most common signs of toxicity, in overdose typically with 25-OH-D above 150 ng/ml. In granulomatous diseases and some lymphomas, the body can react with hypersensitivity.
Marcinowska-Suchowierska et al. 2018 · DOI: 10.3389/fendo.2018.00550 [Review]
Vitamin D only after consulting a doctor, even in usual amounts
- If you have kidney disease. The kidney activates vitamin D and regulates calcium excretion. If it is diseased, any supplementation belongs in the hands of the treating doctor.
- If you have sarcoidosis or another granulomatous disease. There, immune cells can activate vitamin D themselves, similar to the macrophage from the first section, so that even usual amounts can raise calcium.
- If your blood calcium is already raised. Here, the cause is clarified first.
The MS guideline also warns: doses that lead to hypercalcaemia can have pro-inflammatory and thus negative effects on autoimmune diseases. Too much vitamin D can therefore cause harm precisely where many people hope for the most from it. How vitamin D interacts with magnesium and K2 is explained in Taking vitamin D correctly, and where calcium ends up in Vitamin K2 and D3.
The body turns vitamin D into a hormone. With hormones, “more is better” is rarely a good compass. The goal is not the highest possible value, but a sufficient one, measured at the right time and interpreted by a doctor.
And now you know why good supply and a high dose are two different things.
What you can take away: three levers
So what do you do with all this? First, the map of the evidence.
Supported by large RCTs and meta-analyses
No confirmed protection against infection in 2025, no benefit of high single doses in well-supplied people or of test-and-treat, more falls after very high annual doses, no asthma effect.
Signals from RCTs, narrow or not lasting
Fewer autoimmune diseases in VITAL, gone after stopping. Less MRI activity in early MS.
Mechanistically plausible, human studies thin
Receptor, activation in the macrophage, cathelicidin, dampened T cell messengers.
Association, direction open
Low levels in infections, Hashimoto's and IBD, genetically supported in MS.
Clinical tradition without a strong study base
High-dose burst at the start of a cold.
Clinically, I observe
Levels never measured in recurrent infections, and values from the week of fever.
- Have frequent or severe infections checked by a doctor instead of taking high doses on your own. The vitamin D level can be one building block, alongside a blood count, inflammation markers and, depending on the situation, immunoglobulins.
- Make up for a diagnosed deficiency with regular doses rather than large bursts, with a dose set and monitored by a doctor. Arguments for this are the missing bolus protection in 2017, D-Health, Sanders and the Endocrine Society, which in people over 50 with an indication prefers daily doses to infrequent high doses.
- Take the other pillars of immune defence seriously. Sleep, exercise, daylight and diet can support the immune defence. What cold exposure can offer is explained in Does cold strengthen the immune system?, and when supplementation makes sense in principle in When dietary supplements make sense.
For me, vitamin D is not an immune booster but a building block whose absence can throw the immune defence out of rhythm. That is why, with frequent infections or autoimmune diseases, I look at the level, read it in context and make up for a deficiency.
I take seriously what the large trials teach: in well-supplied people, more brings no additional protection, very high burst doses can cause harm, and vitamin D never replaces immunotherapy or specialist treatment.
Immune defence does not mean sitting out the winter. Immune defence means living the winter: being outdoors, meeting people, sleeping, moving your body, and having things looked at closely when needed. If you would like to have your situation assessed: below this article you will find the option to book an appointment.
And now you know why the most important step is not the next pack of supplements, but the right question.
Frequently asked questions about vitamin D and the immune system
Does vitamin D strengthen the immune system?
Strengthening is the wrong picture. In the laboratory, active vitamin D sharpens the defence against pathogens and dampens inflammatory T cell responses, so it tends to regulate rather than amplify. In large studies, extra vitamin D brought no confirmed additional benefit to people who were already well supplied.
What exactly does vitamin D do in the immune system?
Many immune cells carry a vitamin D receptor. In cell studies, macrophages activate vitamin D themselves and produce the antimicrobial peptide cathelicidin, and in T helper cells IFN-gamma, IL-17 and IL-21 fell. Whether usual intake amounts do the same in the body has not been shown.
Does vitamin D protect against colds and infections?
Reliable protection has not been demonstrated. The odds ratio for at least one respiratory infection was 0.88 in 2017, 0.92 in 2021 and 0.94 in 2025 with an interval of 0.88 to 1.00, so without statistically significant protection.
Can vitamin D deficiency lead to frequent infections?
People with low levels report infections more often. That is an association and not proof, because inflammation can itself lower the level. Frequent infections need a medical work-up, including a check for antibody deficiency.
Should I take high-dose vitamin D when I have a cold?
Robust studies to support this are missing. Bolus doses showed no protection in 2017, monthly high doses in D-Health led to no fewer infections, and after a very high annual dose Sanders 2010 found more falls. Whether vitamin D makes sense, and how much, depends on the measured level and is for a doctor to decide.
What do the large studies on vitamin D and the immune system show?
VITAL, D-Health and CORONAVIT found no clear benefit in mostly well-supplied people, with monthly high doses or with a test-and-treat strategy. They do not disprove that a deficiency can matter, but they narrow down the benefit of routine supplementation.
Can vitamin D lower the risk of autoimmune disease?
That is open. In VITAL, 123 people on vitamin D and 155 on placebo developed an autoimmune disease, hazard ratio 0.78 at P = 0.05, and two years after stopping it was 0.98. This does not mean that vitamin D prevents autoimmune disease.
What does vitamin D do in Hashimoto's?
In a small meta-analysis, TPO antibodies fell, while TSH, free T3 and free T4 did not change significantly. In VITAL, autoimmune thyroid disease occurred numerically more often on vitamin D. The dose of a thyroid hormone is never changed on your own because of vitamin D.
What role does vitamin D play in multiple sclerosis?
Low and genetically lower levels were associated with a higher MS risk. D-Lay MS showed less activity on MRI, but no significant difference in relapses. The DGN guideline recommends checking the level and making up for a deficiency. Vitamin D does not replace immunotherapy.
What does vitamin D do in Crohn's disease, ulcerative colitis or asthma?
In IBD, CRP fell slightly, while activity and relapses did not change significantly. In asthma, a Cochrane review in 2023 found no difference in severe attacks, with high certainty of evidence. Prescribed medication is never reduced because of vitamin D.
Can vitamin D dampen inflammation in the body?
In the laboratory, active vitamin D lowered inflammatory T cell messengers, and in IBD CRP fell slightly. The direction also runs the other way: after surgery, 25-OH-D fell within 48 hours from 56.2 to 46.0 nmol/L while CRP rose sharply.
Which vitamin D level is right for the immune system?
There is no proven immune target value. Bone-related references name 20 ng/ml, the Endocrine Society named more than 30 ng/ml in 2011 and in 2024 found no clear evidence for an optimal target value. Reference ranges depend on the laboratory, which is why a value belongs in medical interpretation.
Do children need vitamin D for their immune system?
In 2024 the Endocrine Society suggests vitamin D for 1 to 18 year olds, and a Cochrane review found a slight reduction in doctor visits in children up to five years of age, with low certainty of evidence. The EFSA upper limit for ages 1 to 10 is 50 µg per day. Vitamin D in children should be discussed with the paediatrician.
Can you take too much vitamin D, and how would you notice?
Yes. Signs are confusion, repeated vomiting, abdominal pain, passing a lot of urine and severe thirst, and then please seek medical assessment right away. The EFSA upper limit for adults is 100 µg per day. With kidney disease, sarcoidosis or raised calcium, any intake should be discussed with a doctor beforehand.
Where to go from here
Twelve paths from here.
Vitamin D deficiency: why the sun is not enough in winter
Why the sun in Berlin is too low in the sky from October to March.
If you want to weigh sun against skin protectionHow much sun does the body need?
Between fear of skin cancer and deficiency, weighed by skin type.
If it is about taking it with its partnersTaking vitamin D correctly
Magnesium and vitamin K2 as partners.
If you want to know where too much beginsVitamin D overdose and high-dose therapy
Where meeting requirements ends and high dosing begins.
If the path of calcium interests youVitamin K2 and D3: where calcium ends up
Bones, blood vessels and the path of calcium.
If your thyroid is your topicUnderstanding Hashimoto's thyroiditis
A dysregulation of the immune defence, and why L-thyroxine remains important.
If your TPO antibodies are raisedLowering Hashimoto's antibodies
What selenium, gluten, vitamin D and myo-inositol do to the TPO titre.
If your gut is chronically inflamedCrohn's disease and ulcerative colitis, integrative
Diet, LDN and microbiome, assessed as complements.
If you are thinking about diet in autoimmunityAutoimmune protocol and paleo
What studies show and why AIP stays time-limited.
If you want to train your defences without a supplementDoes cold strengthen the immune system?
Ice bathing assessed, including limits and risks.
If the second VITAL finding interests youMeasuring the omega-3 index
How omega-3 can be measured in the blood.
If you want to know when supplementation makes senseWhen dietary supplements make sense
When a supplement closes a gap and when it gets in the way.
Scientific sources
- Demay MB et al. Vitamin D for the Prevention of Disease: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2024;109(8):1907-1947. PMID: 38828931 · DOI: 10.1210/clinem/dgae290 [Guideline]
- Holick MF et al. Evaluation, treatment, and prevention of vitamin D deficiency: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2011;96(7):1911-30. PMID: 21646368 · DOI: 10.1210/jc.2011-0385 [Guideline]
- Ross AC et al. The 2011 report on dietary reference intakes for calcium and vitamin D from the Institute of Medicine: what clinicians need to know. J Clin Endocrinol Metab. 2011;96(1):53-8. PMID: 21118827 · DOI: 10.1210/jc.2010-2704 [Reference value, expert panel]
- German Nutrition Society. New reference values for vitamin D. Ann Nutr Metab. 2012;60(4):241-6. PMID: 22677925 · DOI: 10.1159/000337547 [Reference value, professional society]
- EFSA NDA Panel et al. Scientific opinion on the tolerable upper intake level for vitamin D, including the derivation of a conversion factor for calcidiol monohydrate. EFSA J. 2023;21(8):e08145. PMID: 37560437 · DOI: 10.2903/j.efsa.2023.8145 [Regulatory document]
- Pludowski P et al. Vitamin D supplementation guidelines. J Steroid Biochem Mol Biol. 2018;175:125-135. PMID: 28216084 · DOI: 10.1016/j.jsbmb.2017.01.021 [Review]
- Hemmer B, Gehring K, et al. Diagnose und Therapie der Multiplen Sklerose, NMOSD und MOGAD, S2k-Leitlinie, 2026. Deutsche Gesellschaft für Neurologie (German Society of Neurology), Living Guideline 9.0, AWMF register number 030/050. No DOI, AWMF register [Guideline]
- Baeke F et al. Vitamin D: modulator of the immune system. Curr Opin Pharmacol. 2010;10(4):482-96. PMID: 20427238 · DOI: 10.1016/j.coph.2010.04.001 [Mechanism Review]
- Liu PT et al. Toll-like receptor triggering of a vitamin D-mediated human antimicrobial response. Science. 2006;311(5768):1770-3. PMID: 16497887 · DOI: 10.1126/science.1123933 [In vitro]
- Wang TT et al. Cutting edge: 1,25-dihydroxyvitamin D3 is a direct inducer of antimicrobial peptide gene expression. J Immunol. 2004;173(5):2909-12. PMID: 15322146 · DOI: 10.4049/jimmunol.173.5.2909 [In vitro]
- Gombart AF, Borregaard N, Koeffler HP. Human cathelicidin antimicrobial peptide (CAMP) gene is a direct target of the vitamin D receptor and is strongly up-regulated in myeloid cells by 1,25-dihydroxyvitamin D3. FASEB J. 2005;19(9):1067-77. PMID: 15985530 · DOI: 10.1096/fj.04-3284com [In vitro]
- Jeffery LE et al. 1,25-Dihydroxyvitamin D3 and IL-2 combine to inhibit T cell production of inflammatory cytokines and promote development of regulatory T cells expressing CTLA-4 and FoxP3. J Immunol. 2009;183(9):5458-67. PMID: 19843932 · DOI: 10.4049/jimmunol.0803217 [In vitro]
- Ginde AA, Mansbach JM, Camargo CA. Association between serum 25-hydroxyvitamin D level and upper respiratory tract infection in the Third National Health and Nutrition Examination Survey. Arch Intern Med. 2009;169(4):384-90. PMID: 19237723 · DOI: 10.1001/archinternmed.2008.560 [Cross-sectional, n=18,883]
- Martineau AR et al. Vitamin D supplementation to prevent acute respiratory tract infections: systematic review and meta-analysis of individual participant data. BMJ. 2017;356:i6583. PMID: 28202713 · DOI: 10.1136/bmj.i6583 [Meta-analysis of RCTs, k=25, n=10,933]
- Jolliffe DA 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 Endocrinol. 2021;9(5):276-292. PMID: 33798465 · DOI: 10.1016/S2213-8587(21)00051-6 [Meta-analysis of RCTs, k=46, n=75,541]
- Jolliffe DA et al. Vitamin D supplementation to prevent acute respiratory infections: systematic review and meta-analysis of stratified aggregate data. Lancet Diabetes Endocrinol. 2025;13(4):307-320. PMID: 39993397 · DOI: 10.1016/S2213-8587(24)00348-6 [Meta-analysis of RCTs, k=40, n=61,589]
- Autier P et al. Vitamin D, acute respiratory infections, and Covid-19: The curse of small-size randomised trials. A critical review with meta-analysis of randomised trials. PLoS One. 2025;20(1):e0303316. PMID: 39808630 · DOI: 10.1371/journal.pone.0303316 [Meta-analysis of RCTs]
- van Arragon M et al. Vitamin D for preventing acute respiratory infections in children up to five years of age. Cochrane Database Syst Rev. 2026;4(4):CD015111. PMID: 42037591 · DOI: 10.1002/14651858.CD015111.pub2 [Systematic Review]
- Bergman P et al. Vitamin D3 supplementation in patients with frequent respiratory tract infections: a randomised and double-blind intervention study. BMJ Open. 2012;2(6). PMID: 23242238 · DOI: 10.1136/bmjopen-2012-001663 [RCT, n=140]
- Manson JE et al. Vitamin D Supplements and Prevention of Cancer and Cardiovascular Disease. N Engl J Med. 2019;380(1):33-44. PMID: 30415629 · DOI: 10.1056/NEJMoa1809944 [RCT, n=25,871]
- Pham H et al. 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. 2021;9(2):69-81. PMID: 33444565 · DOI: 10.1016/S2213-8587(20)30380-6 [RCT, n=21,315]
- Jolliffe DA et al. Effect of a test-and-treat approach to vitamin D supplementation on risk of all cause acute respiratory tract infection and covid-19: phase 3 randomised controlled trial (CORONAVIT). BMJ. 2022;378:e071230. PMID: 36215226 · DOI: 10.1136/bmj-2022-071230 [RCT, n=6,200]
- Hahn J et al. Vitamin D and marine omega 3 fatty acid supplementation and incident autoimmune disease: VITAL randomized controlled trial. BMJ. 2022;376:e066452. PMID: 35082139 · DOI: 10.1136/bmj-2021-066452 [RCT]
- Costenbader KH et al. Vitamin D and Marine n-3 Fatty Acids for Autoimmune Disease Prevention: Outcomes Two Years After Completion of a Double-Blind, Placebo-Controlled Trial. Arthritis Rheumatol. 2024;76(6):973-983. PMID: 38272846 · DOI: 10.1002/art.42811 [RCT, n=21,592]
- Munger KL et al. Serum 25-hydroxyvitamin D levels and risk of multiple sclerosis. JAMA. 2006;296(23):2832-8. PMID: 17179460 · DOI: 10.1001/jama.296.23.2832 [Cohort, nested case-control]
- Mokry LE et al. Vitamin D and Risk of Multiple Sclerosis: A Mendelian Randomization Study. PLoS Med. 2015;12(8):e1001866. PMID: 26305103 · DOI: 10.1371/journal.pmed.1001866 [Mendelian Randomisation]
- Jagannath VA et al. Vitamin D for the management of multiple sclerosis. Cochrane Database Syst Rev. 2018;9(9):CD008422. PMID: 30246874 · DOI: 10.1002/14651858.CD008422.pub3 [Systematic Review]
- Thouvenot E et al. High-Dose Vitamin D in Clinically Isolated Syndrome Typical of Multiple Sclerosis: The D-Lay MS Randomized Clinical Trial. JAMA. 2025;333(16):1413-1422. PMID: 40063041 · DOI: 10.1001/jama.2025.1604 [RCT, n=316]
- Wang J et al. Meta-analysis of the association between vitamin D and autoimmune thyroid disease. Nutrients. 2015;7(4):2485-98. PMID: 25854833 · DOI: 10.3390/nu7042485 [Meta-analysis of observational studies, k=20]
- Jiang H et al. Effects of vitamin D treatment on thyroid function and autoimmunity markers in patients with Hashimoto's thyroiditis-A meta-analysis of randomized controlled trials. J Clin Pharm Ther. 2022;47(6):767-775. PMID: 34981556 · DOI: 10.1111/jcpt.13605 [Meta-analysis of RCTs, k=6, n=258]
- Ananthakrishnan AN et al. Higher predicted vitamin D status is associated with reduced risk of Crohn's disease. Gastroenterology. 2012;142(3):482-9. PMID: 22155183 · DOI: 10.1053/j.gastro.2011.11.040 [Cohort, n=72,719]
- Del Pinto R et al. Association Between Inflammatory Bowel Disease and Vitamin D Deficiency: A Systematic Review and Meta-analysis. Inflamm Bowel Dis. 2015;21(11):2708-17. PMID: 26348447 · DOI: 10.1097/MIB.0000000000000546 [Meta-analysis of observational studies]
- Guo Y et al. Effects of oral vitamin D supplementation on inflammatory bowel disease: a systematic review and meta-analysis. Food Funct. 2021;12(17):7588-7606. PMID: 34231596 · DOI: 10.1039/d1fo00613d [Meta-analysis, k=17, n=1127]
- Jolliffe DA et al. Vitamin D supplementation to prevent asthma exacerbations: a systematic review and meta-analysis of individual participant data. Lancet Respir Med. 2017;5(11):881-890. PMID: 28986128 · DOI: 10.1016/S2213-2600(17)30306-5 [Meta-analysis of RCTs, n=955]
- Williamson A et al. Vitamin D for the management of asthma. Cochrane Database Syst Rev. 2023;2(2):CD011511. PMID: 36744416 · DOI: 10.1002/14651858.CD011511.pub3 [Systematic Review]
- Autier P et al. Vitamin D status and ill health: a systematic review. Lancet Diabetes Endocrinol. 2014;2(1):76-89. PMID: 24622671 · DOI: 10.1016/S2213-8587(13)70165-7 [Systematic Review]
- Silva MC, Furlanetto TW. Does serum 25-hydroxyvitamin D decrease during acute-phase response? A systematic review. Nutr Res. 2015;35(2):91-6. PMID: 25631715 · DOI: 10.1016/j.nutres.2014.12.008 [Systematic Review]
- Waldron JL et al. Vitamin D: a negative acute phase reactant. J Clin Pathol. 2013;66(7):620-2. PMID: 23454726 · DOI: 10.1136/jclinpath-2012-201301 [Prospective human study, n=30]
- Sanders KM et al. Annual high-dose oral vitamin D and falls and fractures in older women: a randomized controlled trial. JAMA. 2010;303(18):1815-22. PMID: 20460620 · DOI: 10.1001/jama.2010.594 [RCT, n=2256]
- Bischoff-Ferrari HA et al. Monthly High-Dose Vitamin D Treatment for the Prevention of Functional Decline: A Randomized Clinical Trial. JAMA Intern Med. 2016;176(2):175-83. PMID: 26747333 · DOI: 10.1001/jamainternmed.2015.7148 [RCT, n=200]
- Burt LA et al. Effect of High-Dose Vitamin D Supplementation on Volumetric Bone Density and Bone Strength: A Randomized Clinical Trial. JAMA. 2019;322(8):736-745. PMID: 31454046 · DOI: 10.1001/jama.2019.11889 [RCT, n=311]
- Marcinowska-Suchowierska E et al. Vitamin D Toxicity-A Clinical Perspective. Front Endocrinol (Lausanne). 2018;9:550. PMID: 30294301 · DOI: 10.3389/fendo.2018.00550 [Review]
- The immune biology comes from the laboratory, and whether it transfers to usual intake amounts has not been shown.
- The odds ratio of 0.30 is a subgroup within a subgroup and was not confirmed in 2021 and 2025. Autier has for years taken a critical reading of the vitamin D trials, and this counter-reading is also unproven.
- The VITAL baseline level and the 39 percent come from the full text of Hahn 2022, the latter from a time-window analysis.
- The caution box is based on physiology and a clinical review.
- The interpretation that an effect disappearing after stopping points to ongoing support is my own interpretation.
- What is deliberately not included here: no dose recommendation, no intake schedule, no source of supply, no universally valid reference range. All doses are study data. No paragraph implies that cortisone, immunosuppressants, MS immunotherapies, thyroid hormones or asthma medication should be changed, or that a medical assessment can be postponed.