A growing body of evidence suggests that vitamin D plays a critical role in preventing and treating COVID-19. Observational data demonstrate that people who take vitamin D regularly are 34 percent less likely to develop COVID-19. Findings from a recent study show that high-dose vitamin D booster therapy reduces the risk of death among COVID-19 patients.

The retrospective, cross-sectional observational study involved 986 COVID-19 patients admitted to two hospitals in the United Kingdom (Royal Preston Hospital and University Hospitals of Leicester) over a period of approximately seven months in 2020. The authors of the study categorized the patients according to their vitamin D status: replete (> 50 nmol/L), insufficient (25 to 50 nmol/L), or deficient (< 25 nmol/L), and whether they received high-dose vitamin D booster therapy or not.

The average age of the participants at Royal Preston Hospital was 76 years, and their average vitamin D levels were 45 nmol/L (insufficient). The average age of the participants at University Hospitals of Leicester was 70 years, and their average vitamin D levels were 43 nmol/L (insufficient). A total of 151 of the patients received vitamin D booster therapy; 66 percent of these received 40,000 IU weekly, and 32 percent received 20,000 IU weekly.

The data from this observational study indicated that those who received vitamin D therapy were 87 percent less likely to die from COVID-19, regardless of their baseline vitamin D levels. Clinical trials using varying vitamin D doses and frequencies are critical to determining whether vitamin D is indeed causal in the fight against COVID-19. Some randomized controlled trials have shown that weekly and daily vitamin D doses are protective against respiratory infections but single monthly doses are not. Furthermore, inter-individual variation in response to vitamin D supplementation is fairly common. In a sample of 35 healthy young people given 8,000 IU of vitamin D, 14 were high responders, 11 were mid responders, and 10 were low responders

The time and circumstances surrounding administration are important, too. For example, one recent study involving just 208 COVID-19 patients found no benefit when seriously ill patients received vitamin D roughly 10 days after symptom onset in conjunction with other treatments, including dexamethasone. The FMF team has identified several flaws in this study’s design and implementation.

First, the patients who received vitamin D had more pre-existing conditions (obesity, high blood pressure, and diabetes) than the patients who received the placebo at baseline, so one would expect these patients to have worse outcomes, regardless of supplementation. In addition, there is considerable discordance between the primary endpoints: More of the patients who received vitamin D died, but fewer were admitted to the ICU or were ventilated, an anomaly that suggests the caregivers were not blinded and were biased to undertreat the vitamin D group.

Perhaps the greatest flaw is how underpowered the study is. With a sample size of only 208 participants, the authors of the study would have 80 percent power to detect a 50 percent difference in hospital length-of-stay. Setting the bar such that hospitalization stays need to be cut in half for vitamin D to be deemed successful is unrealistic. Even if vitamin D supplementation did cut hospitalization in half, there is still a 20 percent chance the study would not show it. (For comparison, millions of people are prescribed low dose “baby” aspirin to prevent a major coronary event despite the fact that 265 people would have to be treated in order for just one person to actually benefit from it.)

Finally, title of the article is incorrect. As explained in this article in Nature, studies never prove the null hypothesis, they only reject or fail to reject it. In this study, the null hypothesis is that there is no difference in outcome between vitamin D and a placebo. Because the findings in this study did not reach statistical significance, it failed to prove there is a difference but did not prove that there was no difference.

A randomized controlled trial to investigate the prophylactic use of vitamin D for reducing COVID-19-associated hospitalization and mortality (VIVID Trial) is currently underway and is still recruiting participants. This type of trial may be the best hope for proving the clinical benefit of vitamin D in COVID-19 since people would take vitamin D before illness occurs. Hospital-based treatment trials are challenging because patients are already very sick, and it may be too late for vitamin D to be protective.

• Link to full study.