Evidence and Rumor
Making Sense of Treatments Claimed to Prevent or Treat COVID-19
Claims abound throughout traditional and social media about supposed treatments that allegedly can prevent or treat COVID-19. Common sense dictates that not all of these claims are true, but how does one sift through all the drugs, vitamins, minerals, and devices that are said to thwart the virus that causes COVID-19, SARS-CoV-2?
Usually, we await the results of at least one large randomized controlled trial (RCT) (ore on what this is later) before accepting that a new treatment is likely to be safe and effective. For any drug or device that is explicitly aimed at treating a specific disease(s), this is generally what the US Food and Drug Administration (FDA) requires for approval. There are many supplements, medical foods, and “alternative” treatments that do not advertise a benefit for a specific disease but rather claim to improve some aspect of health, like “boosting” the immune system or improving memory. These drugs are not regulated by the FDA, but healthcare professionals and scientists still usually demand that rigorously conducted RCTs be completed before they will recommend them to their patients.
These are, however, not ordinary times. The coronavirus pandemic is a health emergency and it may not be advisable to wait the years it can take for the most definitive studies to be completed for a putative new treatment. As the virus spreads, leaving in its wake death, disability, emotional turmoil, and a ruined economy, we are looking for quick solutions whenever possible. Anything that just might help is going to get attention in the form of press releases from medical schools and pharmaceutical companies, stories in newspapers and broadcast media, and posts on websites and social media. These can sometimes exaggerate the potential benefits of a new treatment or, worse, obscure its potential harms. On the other hand, these communications alert us that something is out there that might work.
Vitamin D is a good case in point. Vitamin D has been touted as potentially both lowering the risk to be infected with the COVID-19 virus and reducing mortality among those infected. This has led a lot of people to take vitamin D supplements, hoping to prevent being infected with the novel coronavirus. Yet to date there are noD completed, published RCTs demonstrating that vitamin D either lowers the risk of getting infected or improves outcomes if one gets infected. So what is the basis for vitamin D and COVID-19 and how does one decide whether to take extra vitamin D during the pandemic? Let’s review three things: first, what exactly is vitamin D; second, what is the difference between an observational study and a randomized controlled trial; and third, how strong is the evidence that vitamin D plays a role in COVID-19?
Vitamin D Biology
Vitamin D is a fat-soluble vitamin (meaning that unlike water-soluble vitamins, the body can store vitamin D) that was discovered in 1922 by Sir Edward Mellanby in Great Britain. Severe vitamin D deficiency leads to a serious weakening of bones and impaired walking, a condition called rickets. The main source of vitamin D is the sun, although it is also found at fairly low levels in some foods like fortified milk. As ultraviolet radiation from the sun hits the skin it causes a reaction that produces vitamin D. But this molecule must be converted in two steps to be active, first in the liver and then by the kidney. Its most obvious role is to increase the absorption of calcium from the gastrointestinal tract into the bloodstream, thus making calcium available for bone strength and health. Without sufficient vitamin D, there is insufficient calcium and bones get weaker.
Our main source of vitamin D is from the sun. It is also found in fairly levels in some foods and fortified milk and of course in supplements (source: Shutterstock).
In addition to its importance for bone health, vitamin D is also known to modulate immune function and reduce inflammation, and this is the basis for believing it might be involved in COVID-19. Vitamin D has been implicated to play a role in a variety of other conditions, including respiratory diseases, cancer, and diabetes.
It has been remarkably difficult to set upper and lower blood level limits for vitamin D or its active metabolite, in part because there is so much variability between different tests and in part because it has been very hard to figure out exactly what blood level is associated with good health. We know the level below which rickets can occur, but this condition is rare in medium- and high-income countries like the U.S.: most Americans get at least enough vitamin D to prevent rickets. We also have pretty good information about blood levels above which serious adverse outcomes occur, most notably too much calcium in the blood which can clog arteries and cause heart and kidney damage. How much is good for the heart, lungs, immune system, and other organs, however, is in dispute. This means that there are reasonable differences of opinion about how much vitamin D people should get every day, either from sun, foods, supplements or some combination of these. The current recommendation is 600 International Units (IU) for people aged one to 70 and 800 IU for people over 71, but some experts think more is needed for optimal health. The safe upper daily limit for adults is 4,000 IU. That leaves a lot of room between what is generally recommended and what people can tolerate, opening up the question of how much we really need.
The Difference Between Observational and Randomized Controlled Studies
We mentioned above that the gold standard for evaluating the benefits and risks of any new drug or device is the randomized controlled trial (RCT), but this is not the only kind of informative research study. Many important findings emerge from observational studies, which sometimes can be done more quickly than RCTs. In a case like the vitamin D story, an observational study was done by researchers at the University of Chicago in which they looked at medical records of more than 4,300 people who had been tested for COVID-19 and found that of 499 of them who had vitamin D levels recorded before the pandemic, people who were “likely” vitamin D deficient were 77% more likely to get infected with SARS-CoV-2.
This study is called observational because the experimenters did not control any aspect of what happened, they observed how many people got infected and their vitamin D levels and found an association between them. This is very important information, but there are two important things to remember about observational studies. First, because the experimenters don’t control anything, other variables that were unmeasured in the study could also be causing the apparent link between vitamin D and coronavirus infection. For example, perhaps poor people have less access to healthcare and live in more crowded conditions and also have less access to fortified milk. Anyone of those three things (plus many others) could be the reason for the group to have higher rates of infection. The study did not report on income, healthcare access, or living conditions and if any of these were the real reason for the higher risk of getting infected, then giving someone vitamin D would probably not help them.
And second, because of this issue of unmeasured variables, observational studies only indicate associations between two or more things but cannot establish cause and effect. There is an association between height and weight: taller people generally weigh more than shorter people, but one does not cause the other (that is, separate processes determine how tall a person weighs and how tall they are). In other words, the Chicago study does not tell us whether low vitamin D levels actually cause a person to be at higher risk to get COVID, but it does suggest it might be the case.
The only way to be sure that vitamin D actually causes the problem would be to randomly assign people in a study to receive either vitamin D or a placebo and then see how many in each group either get infected or, if already infected, survive. If the group that gets vitamin D does better than the placebo group, we have evidence of a causal relationship. Because in such a randomized study the only difference between groups is who gets vitamin D, we can rule out all other variables as being the cause of risk of infection.
But RCTs have their problems too. First, in order to get people in each group who are the same on everything except whether they get vitamin D, researchers often have to exclude a lot of people who have situations or conditions that might bias the results. Therefore, RCTs are restricted to people who meet study requirements and may not give an accurate portrayal of who actually would benefit from vitamin D, assuming it beats placebo. Because observational studies do not usually exclude anybody from the study, they are more generalizable. Second, RCTs are very expensive and take a long time to complete. In a situation like the pandemic, we really don’t want to wait for a definitive answer if solid observational data suggest something might save lives.
Randomized clinical trials are considered the gold standard for determining if a new treatment is safe and effective, but in order to keep the intervention and non-intervention groups very similar, many people are disqualified from participating at the eligibility stage, making the trial results limited in generalizability (source: Schulz KF, Altman DG, ((Moher D; for the CONSORT Group)) (2010). "CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials". Br Med J. 340: c332).
The Vitamin D Story
There are several parts to the story of a possible role for vitamin D in COVID in addition to the University of Chicago study described above. First, it is known that people with dark skin are less able to convert sunlight into vitamin D and it is clear that people of color have higher rates of infection and higher mortality rates when infected than do white people. However, there are of course a host of other reasons why this may be the case, including crowded living conditions, lack of access to healthcare, and higher rates of pre-existing medical conditions.
Second, vitamin D is known, as mentioned earlier, to have a variety of effects on the immune system. One of these is an anti-inflammatory effect, which could conceivably be important in reducing the hyperactive immune response (sometimes referred to as “cytokine storm”) that appears to increase symptom severity in some patients with COVID-19. Thus, there is a plausible biological basis for thinking that vitamin D deficiency may be involved in COVID-19 and that vitamin D treatment might be helpful.
Third, there are other observational studies that link “deficient” levels of vitamin D to risk and mortality associated with COVID-19, although some of them are “pre-prints”—articles that are posted online before they have been peer reviewed for publication in a journal (for example this study).
In the face of this, however, are at least two observational studies that did not find any association between vitamin D deficiency and COVID-19.
If all of this sounds confusing, you can perhaps take some solace in knowing that it is confusing experts in the fields of endocrinology, infectious diseases, and immunology as well. The UK National Institute for Health and Care Excellence (NICE) opined at the end of June that there is insufficient evidence to recommend vitamin D supplementation for any respiratory infection. On the other hand, six medical societies, while again insisting that “The current data do not provide any evidence that vitamin D supplementation will help prevent or treat COVID-19 infection,” still released a statement advocating that people get at least 15 to 30 minutes of sun every day (while taking steps to avoid sunburn) or, if that is not possible because of quarantine, stay-at-home, or other factors, taking vitamin D supplements.
All of these statements make the point that we don’t have definitive clinical trial evidence that vitamin D supplementation will help prevent or treat COVID-19 and we have mixed observational evidence that vitamin D deficiency is involved in increasing risk or severity of illness. We would rather not waitfor big RCTs to be done on the one hand while on the other we don’t want to take things that don’t work and may even be harmful. What can we say about vitamin D, then?
First, almost everyone agrees that people who are really vitamin D deficient should take a supplement, so people at increased risk for COVID-19 or who are already infected might do well at least to have a blood level determined.
Second, there doesn’t seem to be any harm in following the recommendation to get 15 to 30 minutes of sun exposure every day (but be careful not to get sunburned and increase your risk for skin cancer). Remember, however, that this is just our read of the conflicting literature and recommendations; we are not in a position to offer anyone medical advice.
The hardest thing of all is to know what to do about taking a supplement even if you don’t have a clear vitamin D deficiency. If you can’t get outside every day, perhaps the six medical society statement “that most adults 19 years and older obtain between 400-1000 International Units (IUs) of vitamin D daily from food and/or with supplements (ideal intake depends on age and sex)” is a reasonable course. Bear in mind, however, that there are significant adverse consequences of taking too much vitamin D. If you can still get vitamin D supplements, don’t gulp them down indiscriminately.
The vitamin D situation illustrates a problem we all have during a health emergency. Balancing the need to use potentially life-saving interventions as quickly as possible with ensuring that those interventions actually work and cause no harm is not an easy task under ordinary circumstances; it is even harder during a pandemic. The press and social media will pick up on every even remote possibility and often give the impression that things are settled science. Upon further review by scientists, what seems like the latest breakthrough often loses some of its shine. Our job at Critica is to try to help you through these complexities so you at least have a broad enough range of information to be able to make your own, informed decisions.