⚠ Evidence warning: Clinical trials or other evidence do not support the claims that vitamin D blood levels of 40–60 ng/mL and practitioners' repletion doses protect against colds and flu. This page keeps it as a report of what people, websites or other media say.
Vitamin D Status and Respiratory Infection
The Martineau et al. individual-participant-data meta-analysis published in the BMJ in 2017, pooling 25 randomized controlled trials with more than 11,000 participants, found that Vitamin D supplementation lowered the odds of at least one acute respiratory infection (ARI) by approximately 12 percent overall. Among participants supplemented daily or weekly, the benefit was concentrated in the deficient: those with baseline 25-hydroxyvitamin D below 25 nmol/L (10 ng/mL) had roughly 70 percent lower odds of ARI. Monthly or quarterly bolus dosing did not produce the benefit. A 2025 update of the same analysis, pooling 40 trials and 61,589 participants, found a similar overall estimate that was no longer statistically significant, and no difference by baseline vitamin D level or dosing frequency. The mechanism is the cathelicidin LL-37 antimicrobial peptide pathway: Vitamin D's active form (1,25-dihydroxyvitamin D) binds the Vitamin D receptor in respiratory epithelial cells and macrophages, drives transcription of cathelicidin, and arms the innate immune system against enveloped and non-enveloped respiratory viruses including influenza A and rhinovirus.
Table of Contents
- Mechanism — Cathelicidin LL-37 and the Vitamin D Receptor
- The Martineau 2017 BMJ Meta-Analysis
- The Urashima Schoolchildren Trial (Influenza A)
- Why the Benefit Concentrates in the Deficient
- The Seasonal Latitude Pattern
- Daily and Weekly Dosing vs Monthly Boluses
- The 25-Hydroxyvitamin D Target Range
- What Trials Gave and What Practitioners Use
- Cofactors — Magnesium, K2, and Vitamin A
- Cautions and Upper-Limit Toxicity
- Key Research Papers
- Connections
- Featured Videos
Mechanism — Cathelicidin LL-37 and the Vitamin D Receptor
The classical model of Vitamin D (calcium absorption and bone mineralization) does not begin to explain its respiratory-infection effect. The respiratory mechanism operates through a fundamentally different pathway, characterized in detail by Philip Liu, Robert Modlin, and colleagues at UCLA in a landmark 2006 Science paper.
The pathway works as follows.
- An invading microbe is detected by an immune cell via toll-like receptor (TLR) signaling, particularly TLR2. In the 2006 Science paper the cells were human macrophages and the microbe was Mycobacterium tuberculosis, the tuberculosis bacterium living inside them; the same pathway in the cells lining the airways (respiratory epithelium) and against viruses comes from other research, not from that paper.
- TLR signaling upregulates both the Vitamin D receptor (VDR) and the CYP27B1 enzyme (which converts circulating 25-hydroxyvitamin D to its active 1,25-dihydroxyvitamin D form) in the responder cell.
- If sufficient circulating 25-hydroxyvitamin D is available as substrate, the cell produces active 1,25-D locally.
- 1,25-D binds the upregulated VDR, the VDR-RXR heterodimer translocates to the nucleus, and binds Vitamin D response elements (VDREs) in the promoters of innate-immune genes — most importantly, the cathelicidin antimicrobial peptide (CAMP) gene.
- The cell secretes cathelicidin LL-37, a 37-amino-acid antimicrobial peptide with broad activity against bacteria, enveloped viruses (including influenza A), and non-enveloped viruses (including rhinovirus). LL-37 disrupts microbial membranes, neutralizes endotoxin, and chemoattracts additional immune cells.
The clinical implication is that this entire cascade requires circulating 25-hydroxyvitamin D as substrate. A patient with serum 25-hydroxyvitamin D below 20 ng/mL has insufficient substrate to mount the LL-37 response when their respiratory epithelium is challenged. A patient with serum 25-hydroxyvitamin D in the 40–60 ng/mL range has plenty of substrate. The Martineau 2017 analysis fits this picture: among people dosed daily or weekly, the benefit was larger in those who started deficient. The Urashima trial did not measure starting vitamin D levels (its larger effect was in children who were not already taking other vitamin D supplements), and the 2021 and 2025 updates of the Martineau analysis found no difference in effect by starting vitamin D level.
The Martineau 2017 BMJ Meta-Analysis
The Martineau et al. 2017 BMJ meta-analysis is the single most influential modern paper on Vitamin D and respiratory infection. It is methodologically strong for three reasons.
- Individual-participant data (IPD) rather than aggregate trial data. The authors contacted the original investigators of 25 randomized trials and obtained the individual-patient datasets, allowing subgroup analyses that would have been impossible with summary-level data alone.
- Pre-registered analysis plan. The subgroup analyses (baseline 25-D status, dosing frequency, age) were declared in advance, not discovered post hoc.
- Sample size. 11,321 participants across 25 trials, ages 0 to 95.
Headline findings.
- Overall: Vitamin D supplementation reduced the proportion of participants experiencing at least one acute respiratory infection by 12% (adjusted odds ratio 0.88, 95% confidence interval (CI) 0.81–0.96, p=0.003).
- Among those receiving daily or weekly vitamin D, baseline 25-D below 25 nmol/L (10 ng/mL): adjusted odds ratio (OR) 0.30 (95% confidence interval 0.17–0.53). Roughly 70% lower odds.
- Among those receiving daily or weekly vitamin D, baseline 25-D at or above 25 nmol/L: adjusted odds ratio 0.75 (95% confidence interval 0.60–0.95). Roughly 25% lower odds.
- Daily or weekly dosing without bolus dose: protective. Single high-dose boluses (e.g., 100,000 international units (IU) monthly or quarterly): not protective and in some studies trending toward harm.
The dosing-frequency finding is mechanistically explainable. A single 100,000 international units bolus produces a transient 25-D spike followed by feedback suppression of the activating enzyme (CYP27B1) and induction of the inactivating enzyme (CYP24A1). The net effect can actually impair local 1,25-D synthesis in tissue for weeks. Daily 1,000–4,000 international units produces a stable serum 25-D level without triggering the regulatory backlash.
How the Evidence Changed After 2017
The same research group updated the analysis twice. The 2021 update (data from 48,488 participants in 43 trials) still found a small overall protective effect (odds ratio 0.92, 95% CI 0.86–0.99), but no significant effect within any group defined by starting vitamin D level. The 2025 update pooled 40 trials with 61,589 participants. The overall effect was similar in size (odds ratio 0.94) but no longer statistically significant (95% CI 0.88–1.00, p=0.057), and no pre-specified subgroup — age, starting vitamin D level, dosing frequency or dose size — changed the result. The authors also reported a lopsided funnel plot (Egger's test p=0.002), a statistical pattern often linked to small-study bias. Vitamin D did not change the share of people with at least one serious adverse event in either update.
The Urashima Schoolchildren Trial (Influenza A)
The 2010 Urashima et al. trial published in the American Journal of Clinical Nutrition was one of the first high-quality randomized trials specifically testing Vitamin D for influenza prevention. Japanese schoolchildren (n=334) were randomized to 1,200 international units/day Vitamin D3 or placebo during the December-March influenza season.
Results.
- Laboratory-confirmed influenza A: 10.8% in the Vitamin D group vs 18.6% in the placebo group (relative risk 0.58, p=0.04). Approximately 42% reduction in influenza A incidence.
- Influenza B: no statistically significant difference (a finding now believed to relate to differences in viral susceptibility to LL-37).
- Asthma attacks in the asthma subgroup: 2 in the Vitamin D group vs 12 in the placebo group (relative risk (RR) 0.17). A striking secondary finding.
The Urashima trial is one of several pieces of evidence (along with Camargo et al. 2012 in Mongolian children and Aloia & Li-Ng 2007 in postmenopausal African-American women) that established the influenza-specific signal for Vitamin D supplementation. Later single trials (VIDARIS in New Zealand, VITAL substudies) gave mixed results. Jolliffe's 2021 update of the Martineau meta-analysis still found a small overall benefit (odds ratio 0.92, 95% CI 0.86–0.99); its 2025 update found a similar estimate that was no longer statistically significant (odds ratio 0.94, 95% CI 0.88–1.00).
For young children, a 2026 Cochrane review of 107 randomised trials (31,521 participants) of vitamin D given in pregnancy and early childhood found low-certainty evidence of a slight drop in the share of children under five who needed a health-care visit for an acute respiratory infection (risk ratio 0.95, 95% CI 0.91–1.00; 10 studies, 2,447 participants). It probably did not lower the number of such visits per child, and higher doses probably did no better than lower doses.
Why the Benefit Concentrates in the Deficient
The Martineau IPD analysis showed a dramatic concentration of benefit in the most deficient participants — among those dosed daily or weekly, about 70% lower odds of ARI at baseline 25-D <10 ng/mL versus about 25% lower odds in those above. The 2021 and 2025 updates did not find a difference by starting level. The 2017 pattern has a clean mechanistic interpretation.
The LL-37 cathelicidin pathway is substrate-limited at low circulating 25-D. Once 25-D is in the sufficient range (cited as >30 ng/mL in the Endocrine Society's 2011 guideline, with optimal often described as 40–60 ng/mL; the society's 2024 guideline found no clear evidence defining an optimal level for disease prevention), additional substrate provides no further benefit because the rate-limiting step is elsewhere (typically TLR signaling, VDR expression, or CYP27B1 activity, not substrate availability).
On this model the implication is twofold.
- In the 2017 analysis, people starting below 10 ng/mL who took daily or weekly vitamin D had about 70% lower odds of a respiratory infection. An effect of this size is rare among preventive interventions.
- People already in the sufficient range showed a smaller benefit. On the substrate model, once the deficiency is corrected the benefit has been captured; pushing 25-D into the 80+ ng/mL range has not been shown to improve the immune outcome and starts to introduce its own risks (calcium dysregulation, kidney stones).
On this reading, whether vitamin D lowers cold and flu risk depends on a person's starting level. The 2021 and 2025 pooled analyses did not confirm a difference by starting level, and the Endocrine Society's 2024 guideline suggests against routine 25-hydroxyvitamin D testing in people without a specific medical reason for it.
The Seasonal Latitude Pattern
The most striking population-level evidence for the Vitamin D / respiratory-infection link is the seasonal pattern of influenza incidence at temperate latitudes. Influenza is a winter disease in the northern hemisphere (peaking December through March) and a winter disease in the southern hemisphere (peaking June through September). At the equator, where seasonal Vitamin D variation is minimal, influenza is endemic year-round without strong seasonal peaks.
John Cannell and colleagues advanced this observation into a coherent hypothesis in a 2006 Epidemiology and Infection paper. The hypothesis is that the winter dip in cutaneous Vitamin D synthesis (ultraviolet B-driven, requires solar zenith angle above approximately 35°) drives population-level 25-D deficiency, which in turn drives population-level susceptibility to influenza and other respiratory infections that are circulating year-round. The viral pathogen is always present; the host immune competence varies seasonally with sunlight exposure.
Multiple lines of evidence support this hypothesis: (1) clinical trial benefit concentrated in deficient patients, (2) the seasonal pattern in temperate latitudes, (3) the equatorial absence of strong seasonality, (4) the documented winter 25-D nadir in northern populations (typically 30–50% lower than summer peak), and (5) the experimental cathelicidin mechanism.
Daily and Weekly Dosing vs Monthly Boluses
The Martineau IPD meta-analysis identified one critical practical finding: daily or weekly Vitamin D dosing protected against ARI; monthly or larger bolus dosing did not (adjusted odds ratio 0.81, 95% CI 0.72–0.91 with daily or weekly dosing and no boluses, versus 0.97, 0.86–1.10 with one or more boluses; P for interaction = 0.05). The 2021 update also saw protection only in trials using daily dosing (odds ratio 0.78, 0.65–0.94) and daily dose equivalents of 400–1,000 international units (odds ratio 0.70, 0.55–0.89), although it found no statistically significant interaction with dose or dosing frequency. What is not known: the 2025 update found no difference by dosing frequency or dose size, so the daily-versus-bolus difference has not been confirmed.
The mechanism is regulatory. The body normally maintains a tight feedback loop on Vitamin D metabolism:
- Low 25-D upregulates the activating enzyme CYP27B1, increasing local production of 1,25-D in target tissues.
- High 25-D (or high 1,25-D) downregulates CYP27B1 and upregulates the inactivating enzyme CYP24A1, which catabolizes both 25-D and 1,25-D.
One proposed explanation (a hypothesis, not a measured trial result) is that a single 100,000–300,000 international units bolus produces a transient serum 25-D spike that triggers the CYP24A1 inactivation cascade, so that for weeks afterward, even when serum 25-D is still in the optimal range numerically, local 1,25-D production in respiratory tissue is suppressed and the cathelicidin response is impaired. The 2017 analysis found no protection with bolus dosing (adjusted odds ratio 0.97), not a higher infection risk than placebo.
In the 2017 and 2021 analyses, protection was seen with daily (or weekly) dosing, not monthly or quarterly boluses. The 2021 analysis linked protection with daily doses of 400–1,000 international units for up to 12 months, and the Urashima trial used 1,200 international units a day. Practitioners commonly use 2,000–5,000 international units of D3 a day, a higher range than the doses linked to protection in the 2021 analysis.
The 25-Hydroxyvitamin D Target Range
The official institutional ranges vary by organization.
- Institute of Medicine (2011): sufficient at 20 ng/mL (50 nmol/L). Focused on bone-health outcomes.
- Endocrine Society (2011 guideline): sufficient at 30 ng/mL (75 nmol/L). Considered broader outcomes.
- Endocrine Society (2024 guideline, which replaced the 2011 one): found no clear evidence defining an optimal 25-hydroxyvitamin D level for disease prevention and suggests against routine testing; it suggests against vitamin D above the Dietary Reference Intakes for healthy adults under 75. It suggests extra vitamin D for children and teenagers aged 1–18 (partly because of its potential to lower the risk of respiratory infections), adults 75 and older, pregnancy and high-risk prediabetes, and daily rather than intermittent high doses for people over 50 who take it.
- A vitamin D advocacy group and a vitamin D research nonprofit: optimal at 40–60 ng/mL based on observational data on infection, cancer, and all-cause mortality.
The Martineau IPD analysis compared only two groups (below and at or above 10 ng/mL). The bands below are a commonly used interpretation built on it, not results measured in that analysis:
- Below 10 ng/mL: profoundly deficient. Highest susceptibility. Largest supplementation benefit.
- 10–20 ng/mL: deficient. Substantial supplementation benefit.
- 20–30 ng/mL: insufficient. Modest supplementation benefit.
- 30–60 ng/mL: sufficient/optimal. Minimal additional supplementation benefit but still appropriate to maintain.
- Above 100 ng/mL: excessive. Risk of hypercalcemia begins to accumulate, particularly if sustained for months.
Vitamin D practitioners and advocacy groups commonly name 40–60 ng/mL as the target for the typical adult and describe it as reachable in most northern adults with daily 2,000–5,000 international units D3 plus seasonal sun exposure. They commonly use a 25-hydroxyvitamin D blood test before and after starting supplementation; the test is widely available from commercial labs, costs $30–$80 cash-pay, and gives a clear number. The Endocrine Society's 2024 guideline suggests against routine testing in people without a specific medical reason for it.
What Trials Gave and What Practitioners Use
What the trials gave
- Urashima 2010 (schoolchildren): 1,200 international units of D3 a day through the December–March influenza season.
- 2021 pooled analysis: protection was linked with daily doses of 400–1,000 international units for up to 12 months.
- 2017 pooled analysis: no protection with one or more monthly or quarterly boluses.
The blood test
- The test practitioners use is 25-hydroxyvitamin D — not 1,25-dihydroxyvitamin D, which is tightly regulated and not a reliable marker of status. The Endocrine Society's 2024 guideline suggests against routine testing in people without a specific medical reason for it.
- Practitioners treat a value below 30 ng/mL as room for benefit.
Repletion doses practitioners use
These schedules come from clinical practice, not from the respiratory-infection trials above. Doses above 4,000 international units a day are above the Institute of Medicine's adult upper limit (see Cautions).
- 25-D 10–20 ng/mL: 5,000 international units D3 daily for 8–12 weeks, then a retest. Maintenance is typically 2,000–3,000 international units/day after repletion.
- 25-D 20–30 ng/mL: 3,000–5,000 international units D3 daily for 8 weeks, then a retest.
- 25-D 30–40 ng/mL: 2,000 international units D3 daily for ongoing maintenance.
- 25-D >50 ng/mL: already in the range practitioners call optimal; current intake is usually continued, with a yearly retest.
Form and timing
- D3 (cholecalciferol) raises 25-D more effectively than D2 (ergocalciferol).
- Vitamin D is fat-soluble, so it is commonly taken with a meal containing fat, such as breakfast with eggs or lunch with avocado.
- Some people report that vitamin D taken in the evening disrupts their sleep.
Cofactors — Magnesium, K2, and Vitamin A
Vitamin D does not operate in isolation. Three cofactors materially affect its function.
Magnesium
Magnesium is required as a cofactor for every enzyme in the Vitamin D metabolic pathway: the hydroxylases that convert D3 to 25-D in the liver, the activating enzyme CYP27B1 that converts 25-D to 1,25-D in tissue, and the VDR itself (which requires magnesium for ligand binding). Magnesium-deficient patients can take Vitamin D supplements and never raise their serum 25-D appropriately because the activation enzymes are not functional. When serum 25-D does not rise as expected on Vitamin D, practitioners look to magnesium deficiency first. A typical adjunct they use is 200–400 mg/day magnesium glycinate or citrate, often taken a few hours apart from Vitamin D.
Vitamin K2
Vitamin D drives calcium absorption from the gut. Vitamin K2 (menaquinone-4 (MK-4) and menaquinone-7 (MK-7) forms) activates the proteins (osteocalcin, matrix Gla protein) that direct absorbed calcium into bone and away from soft tissue / arterial wall. High-dose Vitamin D supplementation without K2 may contribute to arterial calcification in some individuals over years. Practitioners commonly pair Vitamin D doses above 2,000 international units/day with menaquinone-7 (MK-7) at 90–180 mcg/day.
Vitamin A
The VDR and the RAR (retinoic acid receptor, activated by Vitamin A's active form) both heterodimerize with the same partner — RXR (retinoid X receptor). At extreme levels of either vitamin, competition for RXR becomes a factor and the two systems antagonize each other. At physiological levels of both, they cooperate. On this model, a massive imbalance — such as 30,000 international units/day Vitamin A taken alongside 50,000 international units/day Vitamin D — would keep either from working normally.
Cautions and Upper-Limit Toxicity
Vitamin D toxicity is uncommon but real. The toxic syndrome is hypercalcemia, with confusion, weakness, polyuria, kidney stones, and potentially soft-tissue calcification.
- Tolerable upper intake limit (IOM): 4,000 international units/day for adults. Most experts consider this conservative; sustained intake up to 10,000 international units/day is generally well-tolerated in healthy adults, though it is a higher bar than most people need to clear.
- Toxicity concern rises with intake above 10,000 international units/day sustained for months, particularly without serum 25-D monitoring. Case reports of toxicity at 50,000–100,000 international units/day for months are common; case reports of toxicity at 5,000 international units/day are rare.
- Sarcoidosis, primary hyperparathyroidism, lymphoma: in these conditions the body makes active vitamin D (1,25-D) outside the kidney without normal control, so vitamin D supplements can raise blood calcium (hypercalcemia), with confusion, weakness, heavy urination (polyuria) and kidney stones.
- Thiazide diuretics reduce urinary calcium excretion and can amplify the hypercalcemia risk of Vitamin D supplementation in older adults. Serum calcium is the measure used to check for this.
- Pregnancy: 4,000–5,000 international units/day during pregnancy is well-supported by the Hollis and Wagner trials.
Key Research Papers
- Martineau AR, Jolliffe DA, Hooper RL, et al. Vitamin D supplementation to prevent acute respiratory tract infections: systematic review and meta-analysis of individual participant data. BMJ 2017. — PubMed PMID: 28202713
- Jolliffe DA, Camargo CA Jr, Sluyter JD, et al. Vitamin D supplementation to prevent acute respiratory infections: a systematic review and meta-analysis of aggregate data from randomised controlled trials. Lancet Diabetes Endocrinol 2021. — PubMed PMID: 33798465
- Urashima M, Segawa T, Okazaki M, et al. Randomized trial of vitamin D supplementation to prevent seasonal influenza A in schoolchildren. Am J Clin Nutr 2010. — PubMed PMID: 20219962
- Liu PT, Stenger S, Li H, et al. Toll-like receptor triggering of a vitamin D-mediated human antimicrobial response. Science 2006. — PubMed PMID: 16497887
- Cannell JJ, Vieth R, Umhau JC, et al. Epidemic influenza and vitamin D. Epidemiol Infect 2006;134(6):1129-40. — PubMed PMID: 16959053
- Camargo CA Jr, Ganmaa D, Frazier AL, et al. Randomized trial of vitamin D supplementation and risk of acute respiratory infection in Mongolia. Pediatrics 2012. — PubMed PMID: 22908115
- Aloia JF, Li-Ng M. Re: epidemic influenza and vitamin D. Epidemiol Infect 2007;135(7):1095-6. — PubMed PMID: 17352842
- Ginde AA, Mansbach JM, Camargo CA Jr. 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. — PubMed PMID: 19237723
- Bergman P, Lindh AU, Bjorkhem-Bergman L, Lindh JD. Vitamin D and respiratory tract infections: a systematic review and meta-analysis of randomized controlled trials. PLoS One 2013;8(6):e65835. — PubMed PMID: 23840373
- Hewison M. Antibacterial effects of vitamin D. Nat Rev Endocrinol 2011;7(6):337-45. — PubMed PMID: 21263449
- Goodall EC, Granados AC, Luinstra K, et al. Vitamin D3 and gargling for the prevention of upper respiratory tract infections: a randomized controlled trial. BMC Infect Dis 2014. — PubMed PMID: 24885201
- Charan J, Goyal JP, Saxena D, Yadav P. Vitamin D for prevention of respiratory tract infections: a systematic review and meta-analysis. J Pharmacol Pharmacother 2012;3(4):300-3. — PubMed PMID: 23326099
- Jolliffe DA, Camargo CA Jr, Sluyter JD, 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. — PubMed PMID: 39993397
- Demay MB, Pittas AG, Bikle DD, et al. Vitamin D for the Prevention of Disease: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab 2024;109(8):1907-1947. — PubMed PMID: 38828931
- van Arragon M, Grant CC, Scragg RK, Jordan VM. Vitamin D for preventing acute respiratory infections in children up to five years of age. Cochrane Database Syst Rev 2026;4(4):CD015111. — PubMed PMID: 42037591
Connections
- All Remedies
- Benefits Hub
- Cold and Flu Treatments (Main Page)
- Zinc Acetate Lozenges
- Elderberry and Antiviral Activity
- Steam and Saline Rinses
- Vitamin D3 (Main Page)
- Vitamin K2 (cofactor)
- Vitamin A (RXR cofactor)
- Magnesium (D activation cofactor)
- Zinc
- Vitamin D Lab Test (25-hydroxyvitamin D)
- Immune Boosting
- Pulmonology
- Influenza — the illness the Urashima schoolchildren trial measured, where the Vitamin D effect was largest.