If there’s one theme that I’ve written about over and over again in this newsletter, it’s the importance of high quality evidence to guide medical advice and decision-making.
The gold standard of evidence is a randomized controlled trial. Although this type of trial certainly has limitations, it remains the highest quality of evidence for a medical intervention that we have.
So when you hear a discussion about why masks are important to wear during a pandemic, it’s certainly reasonable to ask about a randomized controlled trial to evaluate their effectiveness. Up until now, we didn’t have one.
Now we do.
The DANMASK-19 trial aimed to answer the question about effectiveness of facemasks to prevent COVID-19.
This trial, obviously, made a lot of news.
The headlines are fascinating.
New York Times: “A New Study Questions Whether Masks Protect Wearers. You Need to Wear Them Anyway”
Washington Post: “Health experts dispute conservatives’ claim that new study finds masks are ineffective”
Los Angeles Times: “Face mask trial didn’t stop coronavirus spread, but it shows why more mask-wearing is needed”
We’ll get to the media coverage of this study a bit later on, but first, let’s look at the trial itself.
If you want to read the full trial, here it is.
What did this study test?
This is a cool trial.
The study was done in Denmark in the spring of 2020. This is a time when very few people were wearing masks outside of hospitals and Danish authorities did not recommend mask wearing.
The “mask group” in the study received 50 surgical masks and were instructed to wear them when outside the home. The “control group” in the study didn’t receive masks and weren’t instructed to wear masks.
3030 people were randomized to the “mask group” and 2994 were randomized to the “control group.”
Both groups of patients were followed for a month and tested for COVID if they had any symptoms and at the end of the study.
They had great adherence to mask wearing in the study - 93% of participants wore the mask completely as recommended or predominantly as recommended.
It was a sample of middle aged people, both men and women, with a variety of different jobs:
About 15% of patients didn’t complete the study for various reasons.
What did this Danish Study Find?
1.8% of patients in the mask group and 2.1% of patients in the control group got COVID:
The difference between the two groups wasn’t statistically significant, which means there was no difference in COVID infections between the groups.
So, in this study, mask wearing outside the home didn’t prevent COVID infection.
Wait, did you just say that masks didn’t prevent COVID infection?
Well, that’s the top line result. But it only tells a tiny part of the story.
Masks didn’t prevent COVID infection in this population of Danes during the spring of 2020.
But you need to be careful before extrapolating this result to draw any broad conclusions.
The biggest and most important caveat is that COVID numbers were low in Denmark during the study period. This means that risk overall was low, regardless of what precautions were taken.
We don’t know about how these results would be different if the baseline COVID rates were higher during the study period.
That’s a big limitation!
Since a lot of COVID is transmitted in the home, a place where participants weren’t wearing masks during the study, they weren’t always “protected” by a mask when there was transmission risk.
Because this study randomized individuals and didn’t randomize families or geographic clusters of people, you can’t necessarily extrapolate this to a situation where everyone is wearing masks.
In a pandemic, we don’t care about any one individual’s risk of COVID (from a public health perspective), we care about disease spread
The most important big picture issue with this study is that it doesn’t tell us about the impact of masks at reducing spread in high prevalence areas when overall risk of COVID is high.
That makes it basically impossible to extrapolate to draw conclusions public policy.
I’ve written before about the multiplicative impact of wearing masks. This means that even if a mask reduces spread by a really small amount, when you multiply that impact across huge numbers of people, you get a meaningful reduction in ongoing spread of a deadly virus.
This study wasn’t designed in a way to evaluate that.
The right way to test the question of whether masks are beneficial in a pandemic is a type of trial called “cluster randomization.” This means that instead of randomizing participants in trial, you randomize groups of people organized by geography.
Since this wasn’t a cluster randomized trial, I think that drawing policy conclusions from this study is a fool’s errand.
That doesn’t mean a trial like this is useless! On the contrary, in fact.
We are in desperate need of high quality population evidence on ways to reduce spread of the pandemic. This is a tremendous step forward.
Wait, so what’s the deal with those headlines?
It certainly sounds like the headline writers have a clear agenda.
And perhaps they should. But you can see how a mask-skeptic can read those headlines and feel like they’re being manipulated.
The headlines don’t accurately describe what the research showed. It didn’t really tell us much about whether the population spread of COVID is impacted by mask wearing.
It’s disappointing to see how a study like this is represented in the press. On one hand, it makes sense. You don’t want to discourage mask wearing when it really does seem to be a valuable way to reduce viral spread across the population.
So I don’t think that this study found that masks are “ineffective.”
Communicating about science is hard. Especially when research has limitations and we’re trying to make important personal decisions with that evidence.
The DANMASK-19 trial is a valuable contribution to the scientific literature, but it isn’t the be-all and end-all research on mask wearing.
The results of this trial were surprising! It’s ok to acknowledge that and still recognize that the bulk of the research supports masking during a pandemic.
So bottom line: I’m still advocating for wearing a mask. This trial doesn’t change much for me. But I’m glad that it was done and I will also advocate for higher quality data on population based non-pharmacologic interventions to fight this pandemic.
Thank you for reading! If you’re enjoying my newsletter, please consider sharing with your friends and family and encouraging them to subscribe!
I always appreciate any feedback or thoughts you might have. You can reply directly to this email to reach me directly.