The one thing that we all long for in any crisis is a silver bullet. It would be amazing if there was a simple, easy answer to the problems that face us that could just make them go away overnight, and nowhere is this more true than in the global pandemic of COVID-19 that we are facing. The novel coronavirus has brought out the tragic optimist in all of us, with people desperately hoping that everything from sex to sleep might help ensure that you are safe from the dread disease.
And the newest fad? Vitamin D. A tablet a day to make your troubles go away.
According to the news media pretty much everywhere, the key to surviving COVID-19 is to start taking Vitamin D supplements and never stop. For example, based on the Daily Mail, you’re twice as likely to die from coronavirus if you have low Vitamin D levels than normal levels. The Irish Times is reporting that Irish Vitamin D is essential to surviving the disease. Today dot com calls Vitamin D a “key nutrient” and recommends that you start taking a supplement now.
All of this sounds rosy, but as soon as you take a glance at the evidence something immediately jumps out.
You see, we actually don’t have any idea whether a Vitamin D supplement will improve COVID-19 outcomes. We don’t even really know if low levels of the vitamin make outcomes worse.
It is, as they say, a bit of a minefield.
There have been so many studies conducted looking at correlations between Vitamin D and COVID-19 already that it’s hard to summarize them all, but broadly speaking they fall into two categories — the really bad and the truly, unforgivably awful*.
Most of the recent news reports are based on three very similar studies. While there are some differences between the three, they basically did the same thing — and had the same glaring issue — so I think it’s fair to take them as a group.
What these three pieces of research did was look at Vitamin D levels from previous research in a range of countries, and then compare those countries on the death rate from coronavirus. Countries where there is previous evidence that the average Vitamin D level is lower had slightly higher levels of COVID-19 death, which the authors argued was because low Vitamin D increases your risk of respiratory infection. If low Vitamin D is associated with a higher risk of death, increasing Vitamin D might lower your risk of death, thus we should supplement immediately!
Except, of course, this quite clearly makes very little sense. These aren’t studies comparing people who have higher or lower levels of Vitamin D on COVID-19 outcomes. They aren’t even retrospective studies looking at COVID-19 outcomes and seeing if the people who died were more likely to have low Vitamin D. They are just vaguely correlating past measured levels of Vitamin D with the number of deaths a country is now seeing from coronavirus. If nothing else, it is far too premature to do this, because death rates are evolving at an incredibly quick rate — the correlation might disappear entirely over the next month even if you find it today. It’s also just terrible science — lots of things correlate well, but that doesn’t mean anything about whether one causes another.
Which brings us to the second group of studies that’s hitting headlines everywhere. Again, the sheer number makes it hard to go through everything, but broadly speaking these are studies that look at people who’ve had COVID-19, and what their Vitamin D levels were. And while this sounds good — it’s a much better way to look at the issue than useless country-level correlations — the actual studies that have so far been published seem to all be, well, terrible.
There’s this piece of research from Indonesia, where they picked a group of patients — with no information on how — conducted somewhat inappropriate statistical tests, and misdefined their methodology. Another study looking at the question got their analysis entirely the wrong way around, so instead of looking at whether Vitamin D prevented death from coronavirus they looked at whether death from coronavirus impacted Vitamin D levels. Both of these studies appear to have been done without ethical approval, which if true is immensely problematic and may very well be illegal.
But being bad is no barrier to impact — one study that has been downloaded 10,000s of times already found that low Vitamin D levels were more common people who admitted to ICU, but included just 20 patients. Meanwhile, the biggest study of this kind so far found no difference in Vitamin D levels between people who were or were not infected with COVID-19, but was extremely preliminary and has only seen a few thousand views, never mind downloads.
One thing you’ll note about all the research above is that pretty much every study is a pre-print. Pre-prints are a way of getting research out quickly to get rapid responses, without waiting for peer-review, but they have some really important limitations. It’s not surprising, perhaps, that research that hasn’t been peer-reviewed might make very basic errors in statistical design, but it does mean that you should take all of this with a HUGE pinch of salt.
This doesn’t mean pre-prints are useless or scientifically invalid, but it does make caution really important when you read them. Unfortunately, the media has taken entirely the opposite view, with hundreds of stories about the presumed benefits of Vitamin D despite these studies being extremely flawed and very unlikely to show anything of the sort.
The other issue with all of these studies is that they are very simple observational research. Most of them don’t even adjust for age, which we know influences both Vitamin D levels and your risk of dying from COVID-19. It’s hard to know whether the results seen are the result of confounding factors, study design, or might have something to do with Vitamin D itself.
It’s also always worth being wary about studies proclaiming benefits from Vitamin D based on observational research, because the last 2 decades are littered with similar hypothesized benefits that have turned out to be wrong. Based on studies like these, Vitamin D has been tested rigorously for everything from heart disease to asthma, and failed to help pretty much every time.
There may be an association between Vitamin D levels and coronavirus outcomes, and it’s certainly possible that lower Vitamin D could cause you to get sicker, but at this point we have no possible way of knowing if that’s the case or not. The research just isn’t there to support it either way.
Should You Supplement?
Ultimately, the question is about supplements, because as I said we all want that silver bullet. Vague correlations are interesting, but we want to know whether we should start popping pills or not. The problem is that, so far, we don’t have any research that even begins to look at this question.
In all of this research, there isn’t anything— not a single study I can find anywhere — actually looked at Vitamin D supplementation and COVID-19. There may be an increased risk associated with having low Vitamin D levels, but that doesn’t necessarily mean that taking a supplement will help, especially when you consider the potential harms that supplementation can bring.
We do know that Vitamin D supplementation may help reduce the risk of having a respiratory infection, based on a very rigorous 2017 systematic review, but even this has some caveats. At best, the benefit was tiny — a 3% absolute reduction in risk over the course of roughly 1 year — and not consistent across all populations. In fact, half of the studies included in this review looked at children, which is a problem given that for COVID-19 we’re really worried about adults. On top of this, the review found no benefit at all for serious outcomes like hospitalization and/or death from Vitamin D.
To put it another way, let’s take the results at face value. Say out of 10,000 people 1,000 will get COVID-19 over the next 12 months. Of these, 200 will be hospitalized, and 10 will die. If they all took Vitamin D, we might expect to see only 970 get COVID-19, but we’d still see 200 hospitalized and 10 dying, based on the results of the systematic review.
The best evidence we have for non-COVID-19 infections seems to indicate that the benefit, if any, from Vitamin D supplementation is pretty negligible. It’s hard to say if this is applicable to the coronavirus — the respiratory infections examined in this study were very diverse and may not translate — but even so it’s not great news.
So should you take a Vitamin D supplement? Here’s where it gets complicated. There are some risks associated with supplementation — every medical intervention has some level of risk — but they are generally accepted to be pretty small. The benefit is probably negligible, and certainly no silver bullet, but given that very few things seem to help with COVID-19, some people might be keen to take any benefit they can find.
Unfortunately, I can’t give you an easy answer. At this point in time, we really have no idea if supplementation is helpful or harmful in this case. If you want to shell out a few dollars on a multivitamin, it’s probably not going to hurt you, but the likelihood that it will make a huge difference in your coronavirus risk is incredibly small.
One thing I would say is that, with many of the hardest-hit areas heading into summer, there might be a better solution than supplements anyway. If you aren’t already, going outside for a brisk walk/run every day in the morning (when it’s not too sunny and/or hot) will up your Vitamin D levels and get some exercise in to boot. It’s a win/win.
Sadly, Vitamin D is probably not a silver bullet against COVID-19.
If only it were.
You can also listen to me host the fantastic Sensationalist Science podcast here:
*Note: there may be a decent study out there somewhere, but the media attention has very much focused on some of the most meaningless research.