Danish study on mask efficacy only tells us half the story

An article in the Spectator by Professor Carl Heneghan and Tom Jefferson from the Oxford Centre for Evidence-Based Medicine claimed that a new study from Denmark shows that wearing face masks “does not significantly reduce the rates of [Covid-19] infection”.

The headline on the article was originally “Landmark Danish study shows face masks have no significant effect”; this was subsequently changed following criticism to “Landmark Danish study finds no significant effect for facemask wearers” (our emphasis). This distinction is important.

The study, which compared infection rates between people who were asked to wear masks and those who weren’t found that the group who wore masks had a slightly lower chance of infection, but not so different as to be statistically significant.

That means the researchers couldn’t be sure the lower chance of infection was real, or due to other factors, including chance, in the study. 

In fact the study was only designed to detect a reduction in infection rate of 50% in the group which wore masks (specifically, a reduction from 2% of people getting infected to 1%). The effect which was found fell below this threshold, and had a wide range of uncertainty.

But the original framing of the Spectator article missed out a key piece of context, one which the authors of the study themselves acknowledged. Because the study only looked at rates of infection among mask-wearers, it only looked at half of the issue around transmission.

As Professor Paul Hunter of the University of East Anglia said in a comment given to the Science Media Centre, “The results of the DANMASK-19 randomised controlled trial on face mask use is a good study of the potential value of wearing a face mask to protect the wearer. It was not designed to show whether potentially infected individuals wearing facemasks would reduce the overall transmission of the infection.”

This potential effect of mask-wearing (stopping the wearer from spreading an infection) is known as “source control”. 

The US Centers for Disease Control and Prevention says: “Source control refers to the use of masks to cover a person’s mouth and nose and to help reduce the spread of large respiratory droplets to others when the person talks, sneezes, or coughs. 

“This can help reduce the spread of SARS-CoV-2, the virus that causes COVID-19, by someone who is infected but does not know it.

It is the major reason given for mask-wearing as a way to reduce the spread of Covid-19—for example, the UK government advice on face coverings says that “face coverings are mainly intended to protect others, not the wearer”.

The Danish study’s authors make it clear that their study did not look at this question, saying: “The findings,[...] should not be used to conclude that a recommendation for everyone to wear masks in the community would not be effective in reducing SARS-CoV-2 infections, because the trial did not test the role of masks in source control of SARS-CoV-2 infection”

Despite the headline change, the Spectator article does not directly address this point.

It claims that there is little evidence from trials that face masks protect wearers, especially in the community (as opposed to in healthcare settings). 

It adds: “The only studies which have shown masks to be effective at stopping airborne diseases have been ‘observational’ – which observe the people who ordinarily use masks, rather than attempting to create a randomised control group.”

It then goes on to note problems with observational studies, such as the difficulty in getting people to accurately recall whether they were or were not doing certain behaviours. 

At no point does it discuss other evidence that suggests the possible benefits of mask-wearing as a form of source control, as opposed to benefitting the wearer. 

For example, a paper published in Nature Medicine back in April found that the shedding of respiratory viruses (including coronavirus, rhinovirus and the flu virus) were significantly lower from infected people who wore masks than those who did not.

As noted, the study’s results had a wide range of uncertainty. The slightly lower level of infection seen among the mask-wearing group could indicate anything from a reduction in infection that fell just below the 50% threshold, to no effect at all, or even a small possible increase in the risk of infection.

Professor James Naismith of the Rosalind Franklin Institute and University of Oxford, also told the Science Media Centre: ““The credible interval of their data ranged from an increase of 23% to a reduction of 46% of infection to the mask wearer... It concludes wearing a mask does not cut the infection rate of mask wearers by 50%. I would have been stunned if it had.”

So while the study is good evidence that mask-wearing does not cut your chance of catching Covid-19 in half, it doesn’t tell us whether they may have a smaller protective effect on the wearer, and says nothing about whether they might prevent transmission to others.