Study at a glance
What question did the studies ask?
Does giving vitamin D during serious illness improve recovery or restore immune function? Studies included critically ill adults, hospitalised COVID-19 patients and a small trial in sepsis, a serious illness involving a harmful response to infection. Another study used existing patient records to examine persistent symptoms months after infection.
What did they find?
The larger hospital trials did not establish their main recovery or survival benefits, even when blood vitamin D rose. The small sepsis study found selected biological changes. Some comparisons using existing patient records suggested fewer persistent symptoms, but could not establish causation.
What does this mean?
A changed blood test or immune marker does not by itself show that a patient recovers faster or that the whole immune system works better.
What are the limits?
Large doses given during illness differ from vitamin D status before illness. Small groups, uncertain secondary findings and differences between patients limit conclusions. Research treatments are not personal dosing instructions. The evidence does not establish a general immune-recovery benefit.
Read the full article below ↓Vitamin D trials in serious illness distinguish a rise in blood vitamin D from a demonstrated improvement in recovery. Large critical-care and hospital trials did not establish the main clinical benefits being tested, while a small sepsis trial found selected biological changes without proving restored immune competence. Longer-term observational findings remain vulnerable to confounding. This review asks what the evidence establishes about recovery and resistance to immune dysregulation. Read the shorter account, Vitamin D, Serious-Illness Recovery and the Limits of Immune Resilience Claims: Study Summary.
Clinical recovery and immune recovery are not interchangeable
Recovery after serious illness can involve survival, leaving hospital, regaining organ function, returning to ordinary activities and resolving persistent symptoms. These outcomes matter to patients, but they do not all measure the same process. A treatment might change a blood marker without shortening illness, or affect a clinical outcome without establishing which biological pathway caused the change.
The wider topic in Vitamin D and Systemic Resilience concerns coordination across body systems. A hospital outcome therefore cannot automatically be assigned to an immune mechanism. Similarly, Vitamin D and Immune Modulation describes adjustment of immune activity, but a reduction in one inflammatory signal does not establish restoration of a well-regulated system. A strong claim of immune recovery needs appropriate functional measurements and evidence connecting them with the clinical result.
The intervention also needs to match the question. Giving a large vitamin D3 dose during acute illness is not the same as maintaining adequate vitamin D signalling before illness begins. Circulating 25-hydroxyvitamin D, used to classify status, is not a complete measure of signalling within every tissue. The trials below test specified treatment strategies; they do not identify a universal concentration that guarantees resilience.
VITdAL-ICU: an initial critical-care trial
Amrein and colleagues’ VITdAL-ICU trial analysed 475 critically ill adults with vitamin D deficiency. It compared an initial 540,000 international units (IU) vitamin D3 dose followed by 90,000 IU monthly for five months with placebo. The primary outcome was hospital length of stay. Median stay was 20.1 days in the vitamin D group and 19.3 days in the placebo group, with P=0.98. The intervention therefore did not establish the intended shortening of hospital care.
Overall hospital and six-month mortality comparisons also did not establish a clear treatment benefit. A favourable mortality finding in a subgroup with 25-hydroxyvitamin D at or below 12 ng/mL (30 nmol/L) generated an important hypothesis, but it was not the primary trial result. The subgroup plan was developed while treatment allocation remained blinded. That is relevant to interpretation, yet it does not turn the subgroup into an independently randomised second trial.
The six-month interaction analysis did not establish a convincing difference in treatment effect by deficiency subgroup. A favourable estimate within one subgroup is not enough to prove that the effect differs from the rest of the trial. The result should therefore be reported alongside the primary null outcome and the overall uncertainty, rather than replacing them.
The current corrected principal report and relevant supplementary material were assessed for these claims. Its formal correction is identified in the sources. The trial remains important because it directly tested a clinical recovery strategy in deficient critically ill patients. It does not demonstrate that vitamin D restored immune regulation, nor establish a personal treatment regimen outside specialist care.
VIOLET: a larger test in deficient critically ill patients
VIOLET was a larger multicentre trial of an early, single enteral vitamin D3 dose of 540,000 international units (IU) versus placebo. It randomised 1,360 patients, with 1,078 whose deficiency was confirmed by laboratory testing forming the primary analysis population. Deficiency was defined as 25-hydroxyvitamin D below 20 ng/mL, equivalent to 50 nmol/L. This selection directly addresses a limitation often raised about trials in vitamin-D-replete volunteers.
At 90 days, death occurred in 125 of 531 participants with available follow-up in the vitamin D group, or 23.5%, and 109 of 528, or 20.6%, in the placebo group. The absolute difference was 2.9 percentage points, with a 95% confidence interval from 2.1 points lower to 7.9 points higher. The trial did not demonstrate a mortality benefit, and the uncertainty should not be rewritten as equivalence.
The trial stopped early for futility, meaning that its monitoring process judged the intended benefit unlikely to be demonstrated by continuing as planned. This reduced the achieved sample below the original target, but the result still supplies substantial contrary evidence to the proposed clinical benefit. Other recovery and organ-related outcomes did not establish a matching advantage despite increased vitamin D concentrations.
Some infection-related subgroup analyses raised concern about higher mortality, but these were exploratory comparisons with additional uncertainty. They should not be promoted into a proven harm in every infected patient. Equally, the observed outcomes do not support portraying this high-dose strategy as harmless or reliably beneficial. The primary finding constrains a treatment claim without resolving every possible regimen or pre-illness exposure.
VIOLET and VITdAL-ICU are separate trials. Their relationship is useful because the later study tested a related acute strategy in a larger confirmed-deficient population. They do not justify selecting only the earlier favourable subgroup while ignoring the later main result. The combined interpretation remains tied to acute treatment, not an assertion that ordinary vitamin D physiology is irrelevant.
Hospitalised COVID-19: higher blood levels without earlier discharge
Murai and colleagues randomised 240 adults hospitalised with moderate-to-severe COVID-19 in Brazil; 237 were included in the principal analysis. A single 200,000 IU oral vitamin D3 dose was compared with placebo. Treatment began after illness was already established, around ten days after symptom onset on average. Participants were not all vitamin-D-deficient, which matters when interpreting the intervention.
Median hospital stay was seven days in both groups. The discharge hazard ratio was 1.07, with a 95% confidence interval of 0.82 to 1.39. Mortality, intensive-care admission and mechanical ventilation did not show significant treatment benefits. These secondary outcomes were imprecise, so the result should not be converted into proof that every clinically relevant effect was excluded.
Blood vitamin D rose, but this biochemical response did not establish improved recovery. Timing, the selected hospital population and the influence of severe outcomes on length-of-stay analysis constrain generalisation. Planned cytokine analyses were not performed because of funding limitations. A claim that the treatment restored inflammatory balance therefore cannot be supplied by those missing measurements.
The trial concerns treatment after admission. It does not directly answer whether a different prior exposure changes the chance of becoming ill. Nor does it show that a larger dose, a different vitamin D form or earlier administration would work. Those are distinct hypotheses that require their own evidence, not explanations that can be presumed when a trial is null.
Sepsis: a small mechanistic trial
Quraishi and colleagues randomised 30 adults with severe sepsis or septic shock to placebo, 200,000 IU cholecalciferol or 400,000 IU cholecalciferol, with ten participants per group. Cholecalciferol is vitamin D3. The trial followed biological measurements over seven days. Cathelicidin is an antimicrobial peptide; the measured LL-37 component is relevant to innate defence, introduced in Vitamin D and Innate Immunity.
The intervention increased vitamin D measurements, and the higher-dose group showed a greater rise in LL-37 by day five than the lower-dose and placebo groups, suggesting a biological response. This is useful mechanistic evidence, but not proof of improved clinical recovery. A measured peptide concentration does not establish that a patient’s immune system has returned to normal function or that infections will be cleared more effectively.
The groups were very small, several outcomes were examined and follow-up for the key biological questions was short. Estimated bioavailable vitamin D depended on a calculation whose assumptions were not validated for critical illness. A reported readmission significance value could not be reconciled confidently with the small event counts and is not used here to claim benefit. The trial’s reliable contribution is bounded to its assessed biological findings.
This distinction is also relevant to Vitamin D and Adaptive Immunity. Altering an innate defence marker does not demonstrate restored antigen-specific memory or coordinated adaptive immunity. A system-level recovery claim is broader than a single peptide result. The study helps formulate a mechanism to test; it does not complete the clinical evidence chain.
Persistent symptoms after infection: an observational comparison
Di Filippo and colleagues examined existing records and later symptom information from 132 endocrine outpatients infected before vaccination. Fifty had longstanding cholecalciferol supplementation and 82 had not. The study assessed persistent symptoms at six months. Unmatched comparisons did not establish a significant difference, while matched and adjusted analyses suggested lower Long COVID risk among supplemented participants.
This was not a randomised treatment trial. People who used supplements could also differ in health behaviour, clinical care and other characteristics. Matching and statistical adjustment can address measured differences but cannot remove all unmeasured confounding. Small event counts, multiple analyses, incomplete pre-infection vitamin D measurements and telephone-based symptom reconstruction further restrict the inference.
The study excluded intensive-care patients and certain major comorbidities, so it does not replicate the critical-care trials in a less acute form. It also did not directly measure immune function or demonstrate that an immune pathway mediated the association. Its longer follow-up adds a relevant clinical question, but does not establish that supplementation causes recovery from immune dysregulation.
The investigators’ breakthrough-infection report belongs to the same registered programme. Different eligibility periods do not prove that the participants are independent. These reports should not be counted as separate confirmations of one causal protective effect without participant-level information. Likewise, exploratory dose-threshold analyses cannot establish an appropriate treatment target.
What each endpoint contributes
| Endpoint | What it can tell us | What requires additional evidence |
|---|---|---|
| Blood vitamin D rises | The intervention changed the measured exposure | Effective tissue signalling and clinical benefit |
| An antimicrobial peptide changes | A selected biological response occurred | Improved pathogen control or immune coordination |
| Hospital stay changes | A clinical course differed under the analysis used | The responsible biological mechanism |
| Mortality differs | Survival differed within the studied comparison | A particular immune pathway caused the difference |
| Persistent symptoms differ | Longer-term clinical experiences differed | Causation and demonstrated restoration of immune function |
The broader idea of Vitamin D and Adaptive Capacity concerns adjustment to changing demands. None of these trials measured every dimension of that capacity. A useful synthesis therefore asks how directly each endpoint answers the proposed recovery question, rather than treating all outcomes as interchangeable indicators of resilience.
There is also a distinction from structural healing. Vitamin D, Injury Recovery and Inflammation Resolution: Full Evidence Review examines burns, fractures and tissue injury. Those studies address tissue repair rather than the complete clinical course of sepsis or critical illness. Similarly, the inflammatory patterns introduced in Vitamin D and Chronic Inflammation may be relevant to persistent dysregulation, but the present studies do not establish that lowering chronic inflammation is the route to improved recovery.
Conclusion
The four core randomised programmes (VITdAL-ICU, VIOLET, Murai’s hospital trial and Quraishi’s sepsis trial) contribute different levels of evidence. The larger clinical trials did not establish their main proposed recovery or survival benefits. The small sepsis study supplies a biological signal, and the Long COVID cohort supplies an uncertain longer-term association. Neither overrides the clinical null results or proves restored immune competence.
These findings do not establish that additional vitamin D produces system-level immune recovery or resistance to dysregulation. They also do not show that every possible form, timing or baseline nutritional state has been tested. The defensible conclusion is specific: the assessed acute strategies and observational comparisons do not support a general promise of faster or more complete immune recovery.
Sources
Related reading
Vitamin D, Serious-Illness Recovery and the Limits of Immune Resilience Claims: Study Summary
A plain-English account of vitamin D in hospital and sepsis studies, including null recovery results, small biomarker findings and uncertainty.
Read more →Evaluation of serum vitamin D levels in COVID-19 positive critically ill adults
Mean ± SD serum 25(OH)D levels amongst the study participants were 48.5 ± 27.7 nmol/L.
Original paper ↗Read more →Effective food hygiene principles and dietary intakes to reinforce the immune system for prevention of COVID-19: a systematic review
This study aimed at reviewing effective food hygiene principles and dietary intakes to reinforce the immune system for prevention of corona virus disease 2019 (COVID-19).
Original paper ↗Read more →Impact of the COVID-19 pandemic on growth, nutrition and immune function in infants and toddlers
Pre-pandemic (2017 to 2018) and pandemic (2021 to 2023) cohorts were compared.
Original paper ↗Read more →Suboptimal Consumption of Relevant Immune System Micronutrients Is Associated with a Worse Impact of COVID-19 in Spanish Populations
Coronavirus disease 2019 (COVID-19) has caused a global health crisis and the factors behind its differential impact on COVID-19 among populations are still being known.
Original paper ↗Read more →The Role of Vitamin C, Vitamin D, and Selenium in Immune System against COVID-19
Low levels of micronutrients have been associated with adverse clinical outcomes during viral infections.
Original paper ↗Read more →Vitamin D regulation of immune function during covid-19
Whether such will be the case for SARS-CoV2 remains to be seen, but is currently being proposed and investigated.
Original paper ↗Read more →Effect of test-and-treat vitamin D supplementation on SARS-CoV-2 vaccine efficacy and immunogenicity (CORONAVIT; Jolliffe et al., 2022): Full Evidence Review
Detailed analysis of a randomized controlled trial examining vitamin D supplementation and its effects on SARS-CoV-2 vaccine immunogenicity and efficacy in adults with sub-optimal vitamin D status.
Read more →Analysis of vitamin D level among asymptomatic and critically ill COVID-19 patients and its correlation with inflammatory markers
The objective of current study is to analyze the vitamin D level in COVID-19 patients and its impact on the disease severity.
Original paper ↗Read more →Determinants of severity among hospitalised COVID-19 patients: Hospital-based case-control study, India, 2020
Risk factors for the development of severe COVID-19 disease and death have been widely reported across several studies.
Original paper ↗Read more →Epidemiology, outcomes, and utilization of intensive care unit resources for critically ill COVID-19 patients in Libya: A prospective multi-center cohort study
This study aimed to determine the mortality and morbidity of COVID-19 patients in intensive care units (ICU) following 60 days after ICU admission, and explore the factors that influence in‐ICU mortality rate.
Original paper ↗Read more →Investigating the Relationship between Vitamin D and Persistent Symptoms Following SARS-CoV-2 Infection
The emergence of persistent symptoms following SARS-CoV-2 infection, known as long COVID, is providing a new challenge to healthcare systems.
Original paper ↗Read more →Micronutrient Status of Critically Ill Patients with COVID-19 Pneumonia
Micronutrient deficiencies can develop in critically ill patients, arising from factors such as decreased intake, increased losses, drug interactions, and hypermetabolism.
Original paper ↗Read more →The Immunomodulatory Activity of High Doses of Vitamin D in Critical Care Patients with Severe SARS-CoV-2 Pneumonia: A Randomized Controlled Trial
Methods: A nonblinded, randomized controlled trial was conducted.
Original paper ↗Read more →Immune Response to Vaccination against COVID-19 at Different Second-Dose Intervals and Their Associations with Metabolic Parameters
Thus, we aimed to assess whether the metabolic status of individuals affects their humoral immune responses to the vaccination.
Original paper ↗Read more →Increased blood immune regulatory cells in severe COVID-19 with autoantibodies to type I interferons
Regarding Bregs, compared to the milder, non-hospitalized cases, the proportion of IL-35+ and IL-10+ Bregs was elevated in the blood of severe COVID-19 patients, and to a higher extent in those with auto-Abs to type I IFNs.
Original paper ↗Read more →Increased complications of proximal femur fractures during the COVID-19 pandemic: a nationwide medical claims database study in Japan
The proximal femur fractures undergoing surgery during the COVID-19 pandemic or countermeasure period were compared with the pre-pandemic in a Japanese national inpatient data.
Original paper ↗Read more →The impact of vitamin D, vitamin C, and zinc supplements on immune status among Jordanian adults during COVID-19: cross-sectional study findings
A total of 615 adults Jordanian participants were enrolled in this study, an online: based cross sectional survey was used as a tool for this study.
Original paper ↗Read more →Vitamin D Modulates Humoral Responses to SARS-CoV-2 Vaccination in Autoimmune Thyroiditis: An Endocrine-Immune Perspective Supported by Network Pharmacology, Molecular Docking, and Molecular Dynamics Simulations
Autoimmune thyroiditis (AIT) is characterized by dysregulated endocrine: immune interactions, and vitamin D has been proposed as a potential immunomodulatory factor influencing vaccine-induced immune responses.
Original paper ↗Read more →Vitamin D Status and Immune Response in Hospitalized Patients with Moderate and Severe COVID-19
The frequency of CD3+CD4+ T helper (Th) cells was decreased in patients with 25(OH)D level ≤ 11.4 ng/mL, compared to healthy controls (HCs).
Original paper ↗Read more →Original sources
Early High-Dose Vitamin D3 for Critically Ill, Vitamin D–Deficient Patients
Ginde et al., for the VIOLET investigators (2019).
Publication details
Ginde et al., for the VIOLET investigators (2019). Early High-Dose Vitamin D3 for Critically Ill, Vitamin D: Deficient Patients. The New England journal of medicine. 381(26):2529 to 2540. DOI: 10.1056/NEJMoa1911124.
Effect of a Single High Dose of Vitamin D3 on Hospital Length of Stay in Patients With Moderate to Severe COVID-19: A Randomized Clinical Trial
Published in JAMA. 2021;325(11):1053 to 1060. DOI: 10.1001/jama (2020).
Publication details
Murai et al. (2021). Effect of a Single High Dose of Vitamin D3 on Hospital Length of Stay in Patients With Moderate to Severe COVID-19: A Randomized Clinical Trial. JAMA. 2021;325(11):1053 to 1060. DOI: 10.1001/jama.2020.26848.
Effect of cholecalciferol supplementation on vitamin D status and cathelicidin levels in sepsis: A randomized, placebo-controlled trial
Quraishi et al. (2015).
Publication details
Quraishi et al. (2015). Effect of cholecalciferol supplementation on vitamin D status and cathelicidin levels in sepsis: A randomized, placebo-controlled trial. Critical care medicine. 43(9):1928 to 1937. DOI: 10.1097/CCM.0000000000001148.
Effect of High-Dose Vitamin D3 on Hospital Length of Stay in Critically Ill Patients With Vitamin D Deficiency: The VITdAL-ICU Randomized Clinical Trial
Published in JAMA. 312(15):1520-1530. DOI: 10.1001/jama (2014).
Publication details
Amrein et al. (2014). Effect of High-Dose Vitamin D3 on Hospital Length of Stay in Critically Ill Patients With Vitamin D Deficiency: The VITdAL-ICU Randomized Clinical Trial. JAMA. 312(15):1520-1530. DOI: 10.1001/jama.2014.13204.
Typographical Errors
Published in JAMA. 312(18):1932. Correction. DOI: 10.1001/jama (2014).
Publication details
JAMA (2014). Typographical Errors. JAMA. 312(18):1932. Correction. DOI: 10.1001/jama.2014.14740.
Chronic cholecalciferol supplementation and adequate vitamin D status are associated with reduced risk of Long COVID
Di Filippo et al. (2026).
Publication details
Di Filippo et al. (2026). Chronic cholecalciferol supplementation and adequate vitamin D status are associated with reduced risk of Long COVID. Journal of endocrinological investigation. 49(8):1955-1969. DOI: 10.1007/s40618-026-02858-8.