The idea that a cheap, safe, and readily available supplement can help shield people against infection with SARS-CoV-2 and severe disease is appealing.

However, to date, the evidence that vitamin D really does protect against COVID-19 has been mixed.

The vitamin is essential for maintaining healthy bones and muscles, but there is also good evidenceTrusted Source that it provides some protection against contracting infections of the respiratory tract.

So, in theory, it could protect against COVID-19, principally a respiratory infection.

Another line of evidenceTrusted Source in its favor is the observation that people susceptible to vitamin D deficiency — including those with obesity, older adults, and individuals of Black or Asian ethnic origin — are also vulnerable to COVID-19.

An unpublished study found a link between mean levels of vitamin D in the populations of 20 European countries and the number of COVID-19 cases and mortality rates in the countries.

Several studies have found an association between vitamin D status and subsequent COVID-19 infection.

For example, one studyTrusted Source discovered that people found to be deficient in the vitamin were more likely to test positive for the infection up to a year later.

However, the scientists behind this research emphasized that only randomized clinical trials could provide definitive evidence that vitamin D supplementation can prevent COVID-19 and therefore save lives.

This is because observational studies like theirs are vulnerable to two statistical problems, known as “confounding” and “reverse causation.”

Confounding occurs when another variable that the researchers have not fully accounted for influences the outcome. For example, older age and chronic disease affect not only a person’s levels of vitamin D but also their risk of COVID-19.

Reverse causation occurs when the outcome itself affects the variable under investigation. For example, severe COVID-19 could reduce an individual’s vitamin D levels.

In a clinical trial, researchers avoid these problems by randomly assigning participants to receive either the treatment or a placebo, then following them to see what happens.

But clinical trials are costly and take a long time to yield any results.

Random Genetic Variation

Researchers at McGill University in Montreal, Canada, and collaborators from the University of Siena, Italy, used a technique called Mendelian randomization, which uses genetic variation between people to simulate a randomized controlled trial, to investigate the effects of vitamin D.

They first identified genetic variants known to influence a person’s vitamin D level.

In effect, these variants are randomly assigned to people at their conception by a combination of their parents’ gametes. This avoids any biases to the outcome from confounding variables — as the variants may break the associations with other confounders or from reverse causation — because the variants are determined before infection.

The researchers then used the presence of these variants as a proxy for vitamin D levels in 14,134 individuals from up to 11 countries testing positive for COVID-19 and over 1.2 million individuals without the disease.

They found no significant associations between the predicted vitamin D levels — by each individual’s genetic makeup — and COVID-19 infection, hospitalization, or severity of illness.