Intended for healthcare professionals

Rapid response to:

Editor's Choice

The curious case of the Danish mask study

BMJ 2020; 371 doi: (Published 26 November 2020) Cite this as: BMJ 2020;371:m4586

Read our latest coverage of the coronavirus outbreak

Rapid Response:

Conclusions from the Danish mask study

Dear Editor

The DANMASK-19, the first trial of surgical mask use during Covid-19, was finally published. For some authors, the study was inconclusive.[1] That is an inaccurate statement.

The trial was designed to detect at least a 50% protection against a SARS-CoV-2 infection given by mask wearing, assuming that the incidence rate in people not wearing masks was 2% per month. That assumption turned out to be correct, as the authors found a 2% per month rate during the study, and therefore the sample size calculation carried out before the study was right. Since the research did not find a statistically significant result for the effect of masks on SARS-CoV-2 infection, the conclusion we must draw is that masks give a protection that is less than 50%. That is a very important message to give to the many people thinking that masks give a protection that is close to 100%. Should those people attend crowded enclosed spaces assuming that masks protect them, they would put themselves at risk.

Covid-19 is being increasingly recognised as an airborne disease, meaning that the virus can fluctuate in the air, a bit like a gas. The DANMASK-19 study proved that surgical facemasks have limited air filtering capacity with respect to SARS-CoV-2. The kinds of masks we would need to protect ourself should therefore filter the virus in a much more efficient way compared to surgical masks. The study did not evaluate whether people with masks are less likely to infect someone else. However, given that we now know that surgical masks have limited filtering capacity, we must derive that it is very unlikely that surgical masks provide a substantial protection from an infectious wearer.

Another conclusion from the study comes from the interpretation of the infection incidence rates. Thanks to the DANMASK-19 study, Denmark is the only country in the world for which we have reliable estimates around the rates from both the first and the second epidemic wave. The study was conducted just after the first epidemic peak in Denmark (April–May 2020), so we know that the rate during that period of time was about 2% per month. During the second peak (Oct–Nov 2020), the infection rate in the general population was just about 2% per month,[2] hence very similar compared to the first peak. That suggests that all mitigating measures that have been taken in Denmark to prevent a second wave had no tangible effect on the spread of Covid-19. Those measures included a universal mask mandate.

The message from the science of epidemiology is clear: isolation is the only scientifically-sound measure that we have to prevent a SARS-CoV-2 infection now. Whether or not imposing isolation to populations through lockdowns is a political decision that has to take many factors into account, but the distinction between evidence-based and non evidence-based measures is unequivocal.

[1] Abbasi K. The curious case of the Danish mask study. BMJ 2020;371:m4586.
[2] Worldometer. Covid-19 coronavirus pandemic. (accessed 25 Nov 2020).

Competing interests: No competing interests

30 November 2020
Antonio I Lazzarino
Medical Doctor and Epidemiologist
University College London
UCL Department of Primary Care and Population Health, London NW3 2PF, United Kingdom