Recent headlines suggest that I could protect myself from Covid and also cut my risk of dying from Covid by half just by taking Vitamin D. If that is the case, why aren't we all taking it? We hand out masks; why not Vitamin D?
Let's pause here for a reality check: Lots of diseases are correlated with Vitamin D deficiency – but most are not caused by it. It’s a correlation, not a causation. Vitamin D deficiency is often a marker of overall poor health; sicker individuals are more likely to develop low Vitamin D. Treating the deficiency treats the symptom, not the cause. In trials of Vitamin D vs. placebo for preventing disease, Vitamin D often disappoints.
But maybe Covid is different. Covid is a respiratory infection, and in meta-analyses severe Vitamin D deficiency does appear to increase the risk of respiratory infections. We also have a plausible mechanism through which Vitamin D could influence Covid. The latest hypothesis asserts that a “bradykinin storm” is largely responsible for the severe symptoms of Covid. A surge of bradykinin, a peptide related to inflammation, leads to blood vessel dilation and increased permeability (“leaky vessels”). This contributes to swelling and fluid retention in the lungs and numerous other effects throughout the body, including cognitive deficits, severe hypokalemia and resulting cardiac arrythmias, and hypotension.
There are complicated pathways involved in the production and regulation of bradykinin. Covid involves the renin-angiotensin system (RAS), which is involved in blood pressure control, fluid management, and inflammation. Vitamin D modulates this system somewhat, which is how it could play a role in modulating the inflammation and bradykinin storm that can occur with Covid. The relationship between Vitamin D and bradykinin is not simple or straightforward, but it is there (find Vitamin D in the top left of the following image).
Pathways involved in a bradykinin storm
It’s not enough, though, to have a plausible explanation for Vitamin D’s role. We need human studies, and we now have them. Let’s go through some of the studies that have received the most attention.
1: Observational study of individuals tested for Covid (U.S.): Association of Vitamin D Status and Other Clinical Characteristics With COVID-19 Test Results [1]
Question asked: Do individuals with low Vitamin D have higher rates of Covid infection?
How the question was answered: Researchers reviewed records for 489 adults who had been tested for Covid and had had Vitamin D testing within the last year, at a single center in the U.S.
Findings: Those who were deficient in Vitamin D (less than 20 ng/ml) had higher rates of Covid than those with sufficient Vitamin D: 19% vs. 12%. The risk of having a positive test persisted even when correcting for other medical conditions.
Conclusion: There is likely an association between Vitamin D deficiency and developing Covid. This does not necessarily imply that the low Vitamin D caused a susceptibility to Covid. People with low Vitamin D could be more susceptible to infection because they have other health issues. For example, nursing home residents are more likely to be deficient in Vitamin D – and they are also at high risk for Covid spread. The study did attempt to correct for some of this and still found an association between Vitamin D and Covid. Overall, this study provides a solid basis for future research, and reinforces the recommendation to correct Vitamin D deficiencies.
2. Observational study of individuals tested for Covid (U.S.): SARS-CoV-2 positivity rates associated with circulating 25-hydroxyvitamin D levels [2]
Question asked: Is low Vitamin D associated with rate of Covid infection?
How the question was answered: Researchers sifted through Quest lab results to find patients who had had Covid testing and had also had Vitamin D levels checked within the last year. They ended up with over 190,000 individuals from the United States. All had zip code information available (which was used to estimate race). Vitamin D levels were grouped and then the rate of Covid positivity was compared between groups.
Finding: Low Vitamin D levels associated with higher rates of Covid
Here is one graph from the study, showing that as Vitamin D levels rise, Covid positivity rates fall:
For Vitamin D levels under 30 ng/ml, Covid rates appear clearly higher than for levels above 30. After getting above 40 ng/ml, the benefit is less clear. [2]
Conclusion: This is a convincing association, but not evidence for causation. People with Vitamin D levels under 30 ng/ml seem to have higher rates of Covid infection. But those with low Vitamin D likely had other medical conditions, which were not addressed. This study was not designed to address this – it only looked at lab results, not medical history.
Having said that, the data include a large number of individuals and the effect of Vitamin D level is striking. This provides a solid basis for further research.
3. Observational study of Covid patients (Iran): Vitamin D sufficiency, a serum 25-hydroxyvitamin D at least 30 ng/mL reduced risk for adverse clinical outcomes in patients with COVID-19 infection [3]
Question asked: Is low Vitamin D associated with poor outcomes in patients diagnosed with Covid?
How the question was answered: Researchers reviewed the records of 235 patients diagnosed with symptomatic Covid in a single emergency department in Iran. They compared patients with Vitamin D levels greater than or equal to 30 ng/ml to those with Vitamin D levels below 30. The also compared the mortality of patients over age 40.
Findings:
ICU admission and intubation: The two groups had similar rates.
Severity of disease: For patients who were sufficient in Vitamin D (30 ng/ml or higher), 63% had “severe” disease, compared to 77% of those who were deficient. Severe disease defined as: reporting shortness of breath, breathing fast, oxygen saturation below or equal to 93%, or lung infiltrates on CT scan covering more than half of the lungs.
Mortality: When looking at adults over age 40, the mortality rate was higher for deficient adults: 20% vs. 9.7%.
Scatterplot from the study showing mortality risk.
Patients who had sufficient levels of Vitamin D had lower death rates than those with deficient Vitamin D. Note that low Vitamin D was not a death sentence - most people with low Vitamin D survived. [3]
Interesting to note: When comparing mortality, they excluded everyone under the age of 40. You can see on the graph that there were lots of young people with low Vitamin D levels who did not die. (No one under 40 died.) This was not described in the methods, which leads me to believe that it was done after the fact in order to find a statistically significant difference in mortality.
Conflict of interest alert: The senior author on the last two studies listed was Michael F Holick, MD. Holick has been a very vocal advocate of Vitamin D testing and supplementation. He also has extensive financial ties to the lab testing industry (Quest, as well as two companies that produce Vitamin D tests) and the pharmaceutical industry (multiple corporations that sell supplements and medications frequently prescribed with Vitamin D). He has even received money from the tanning industry. Liz Szabo wrote an excellent article on Holick that was published in the New York Times in 2018.
Holick's level of association with these industries is concerning. It does not mean that the studies are wrong. But it definitely adds a possibility of bias in the data collection and interpretation.
Conclusion: Ignoring the possible influence that Holick's conflicts of interest had on the study: this is a convincing association between Vitamin D deficiency and disease severity. It was not designed to show causation. It does provide a basis for further studies.
4. Randomized controlled trial of Covid patients (Spain): Effect of calcifediol treatment and best available therapy versus best available therapy on intensive care unit admission and mortality among patients hospitalized for COVID-19: A pilot randomized clinical study. [4]
Question asked: Will giving Vitamin D to hospitalized Covid patients improve outcomes?
How the question was answered: 76 Covid patients admitted to a single hospital in Spain were randomized to receive calcifediol with standard therapy vs. standard therapy alone. Calcifediol is Vitamin D 25-OH; it is converted to the active form of Vitamin D. It was given orally on hospitalization days 1, 3 and 7, then weekly. The outcomes reviewed were ICU admission and mortality.
Findings
ICU admission: Just one of the 50 patients (2%) receiving Vitamin D needed to be admitted to the ICU, while 13 of the 26 (50%!) of the patients not receiving Vitamin D were admitted to the ICU.
Mortality: None of the 50 patients treated with Vitamin D died, while 2 of the 26 patients in the control group died.
Interesting to note: The patients were randomized, but the groups were small, and they ended up with several notable differences between the groups. Several comorbidities (other diseases that predispose someone to have more severe disease) were more common in the control group than in the Vitamin D treatment group: 19% vs. 6% for diabetes, and 58% vs 24% for hypertension. The control group also had a higher percentage of men. All of these factors could make the control group more likely to develop severe disease (whether Vitamin D had any effect or not). In addition, ICU admission was based partly on comorbidities, so the untreated group was already more likely to go to the ICU.
Another note: The authors did not look at BMI or the rate of obesity, which is now a well-known risk factor for Covid severity. Vitamin D levels were not measured, so we don’t know how many of the patients were Vitamin D deficient at the time of hospitalization, but the authors note that in their population the rate of deficiency is high.
Final note on the "standard care": The standard care for severely ill Covid patients has changed since the start of this study. Patients were given hydroxychloroquine and azithromycin, a regimen that has since fallen out of favor, but not routinely given dexamethasone, something now standard of care. We don't know if we would see the same benefit to Vitamin D if dexamethasone had been given
Conclusion: This is fairly good evidence that Vitamin D may decrease the risk of developing severe disease and death in hospitalized Covid patients. The number of patients was small, and the groups weren’t entirely similar, but even so, Vitamin D did appear to have an effect. This provides a good basis for further research. It does not mean that we could save most people from dying of Covid by just giving Vitamin D. Also, this is just one small trial. I don’t get that excited about any individual trial – you can read more about that here.
Altogether, Vitamin D does look promising.
Studies have not supported many of the sensational headlines associated with them, but they have shown a likely benefit from having sufficient Vitamin D levels. There is a reasonable chance that supplementation could decrease disease severity. There is also a reasonable chance that correcting a severe deficiency may provide some protection from getting Covid.
This does not mean that everyone should start loading up on Vitamin D. I fear that the drive to “protect” themselves will lead many people to take massive doses of Vitamin D – which can do more harm than good. Recall that Vitamin D is a hormone. Too much of it is dangerous, and you don’t need to take a lot to have a big effect. Taking 1000 units daily is safe for most people. But if you are not deficient, it likely isn't doing anything for you. A public health message telling everyone to take Vitamin D could backfire if people begin taking massive doses - doses that are available over the counter.
Isn't there at least enough evidence at this point to recommend it for everyone with Covid?
No, not really. There is suggestive evidence from a small pilot study. There is likely an effect, though we can't yet say how large that effect is. A few doses of high-dose Vitamin D are probably not going to hurt, but there are harms to massive doses of Vitamin D - and if we give it to everyone indiscriminately we will see more of those. Also remember that younger patients were excluded from the assessment of mortality risk in the third study - because no one under 40 died in their study, despite the fact that most of them were deficient in Vitamin D.
I do understand the desire to do anything possible to minimize the damage from Covid. Ideally, we would give Vitamin D in the setting of a clinical trial so we can study the effects. If that is not available, then checking Vitamin D levels and supplementing deficient individuals is a reasonable course of action. It won't eliminate the possibility of dying from Covid, but it could help.
If you are concerned about getting Covid, a moderate dose of Vitamin D may not help, but it is unlikely to do much harm. And if you are deficient, it's probably a good idea to supplement, though it may not prevent Covid. The best ways to prevent Covid are the common-sense measures: physical distancing, wearing masks, and washing hands.
Recommended reading on Michael Holick: Vitamin D, the Sunshine Supplement, Has Shadowy Money Behind It, by Liz Szabo for the New York Times. Published August 2018.
You can also read my 3-part series on Vitamin D
Part 1: The basic science of what Vitamin D does on a molecular level, reviewing studies in humans, mice, and cell cultures.
Part 2: The harms of oversupplementation with Vitamin D.
Part 3: The results of Vitamin D supplementation in large human trials.
Studies cited in article:
[1] Meltzer, D. O., Best, T. J., Zhang, H., Vokes, T., Arora, V., & Solway, J. (2020). Association of Vitamin D Status and Other Clinical Characteristics With COVID-19 Test Results. JAMA network open, 3(9), e2019722.
[4] Entrenas Castillo, M., Entrenas Costa, L. M., Vaquero Barrios, J. M., Alcalá Díaz, J. F., López Miranda, J., Bouillon, R., & Quesada Gomez, J. M. (2020). "Effect of calcifediol treatment and best available therapy versus best available therapy on intensive care unit admission and mortality among patients hospitalized for COVID-19: A pilot randomized clinical study". The Journal of steroid biochemistry and molecular biology, 203, 105751.
Top image by Maonakub at bigstockphoto.com.