This article was originally posted in the Montreal Gazette.
During the pandemic, vitamin D has been touted by many as a cheap, easy way to prevent or cure COVID during the pandemic. But there’s a simple reason I don’t recommend it for these purposes: It doesn’t work.
My issue with the sunshine vitamin is that the accumulated evidence from the past few years has been fairly clear on one point. Once seen as the cure for everything, vitamin D has had a rough time lately. Initially, there was great hope and hype that it would prevent everything from heart attacks to cancer. But then came the randomized trials and the reality is that vitamin D does none of those things.
The problem is that many of the earlier, more optimistic trials, were based on observational evidence. Simply put, researchers looked at people with low vitamin D levels in their blood and found that they had more heart disease, more cancer and more disease over all. Not unreasonably, they supposed that low vitamin D levels were making people sick. What many failed to consider though is that they got the relationship backwards. It wasn’t that vitamin D deficiency made people sick, but rather that sick people tend to become vitamin D deficient because the don’t eat properly and don’t go outside as much. This pitfall, called reverse causation, can be overcome if you do a randomized trial. Take a bunch of people, give half vitamin D and the others a placebo, and you can see if fewer people in the vitamin D group get sick. Sadly, they don’t. Vitamin D doesn’t prevent heart disease or cancer and routine supplementation for these purposes has no benefit.
You would think that we humans would learn from our mistakes, but amazingly, we don’t. When COVID struck, people made the same mistakes. Early on, researchers looked at people who got COVID and measured the vitamin D levels in their blood and found that the ones with low vitamin D levels did worse. Ignored was what should by now be the obvious problem of reverse causation. People with low vitamin D levels are often older and sicker than the general population and will obviously have worse outcomes if they catch COVID-19. You can adjust for these problems statistically to a certain extent, but you can never erase them completely. In reality, much better genetic evidence suggested that low vitamin D levels didn’t increase the risk of catching or being hospitalized with COVID-19 and even some early randomized data in ICU patients showed no benefit.
Now, two randomized trials that were just published will, one hopes, drive the point home. In the CORONAVIT study, researchers tried a test-and-treat strategy. They randomized people to get either usual care or to get a vitamin D test. Those who had vitamin D deficiency got either a low dose (800 units/day) or a high dose supplement (3200 units/day). Neither dose reduced the risk of catching COVID-19. Vitamin D proponents often claim that the negative studies failed because researchers used too low a dose or didn’t identify people with deficiency before hand. Both critiques were addressed in this study, and still vitamin D showed no benefit. What’s more, a second study published in the same issue of the British Medical Journal used cod liver oil rather than a vitamin supplement, but came out equally negative.
Unfortunately, some people don’t let the facts get in the way of a good story, and many truly believe that vitamin D should work based on some preliminary lab data. But when tested in a randomized fashion it neither prevents not treats COVID. Whether you take a low dose or high, test for deficiency before, or use cod liver oil rather than a supplement, the results are the same. They are unfortunately and reliably negative.