select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to ‘Evidence Summary: Coronavirus (COVID-19) and the use of face coverings in education settings’

A report on the use of face coverings in education settings has been published by the Department for Education.


Dr Julian Tang, Honorary Associate Professor/Clinical Virologist, Respiratory Sciences, University of Leicester, said:

“This report on face coverings in schools is a fair summary of the epidemiological evidence (despite omitting some key US studies) on the real-world effectiveness of face masks in schools – including the various caveats about some of the practical difficulties – but some of these can be overcome with time.  People in the UK are still not yet really used to wearing masks – but again, this will come over time.

“Masking and other social distancing interventions are more likely to show a more significant impact against a more transmissible variant like omicron.  However, I disagree with the statement that RCTs are the best way to assess face mask effectiveness in any setting – especially using school absenteeism to measure the impact of masks.

“RCTs are not good for mask studies for various reasons – they are usually used for drug trials, which are quite different:

  • the trial drugs are not otherwise widely available – whereas you can get COVID-19 from anywhere – even outside of work – so the exposure outside the study area cannot be controlled well – people don’t wear masks at home typically, so if they are in the ‘mask’ group, they might get infected from home when not wearing a mask – and be deemed as a ‘mask’ failure that way.
  • unlike a pill that you swallow once 2-3 times a day, you have to wear your mask all the time to get the benefits – with only very brief periods where you can take them off to eat/drink – and this is very difficult for participants, to maintain this degree of proper mask wearing for so long.

“Nevertheless, despite this, a large-scale RCT in Bangladesh did show a benefit of community mask wearing:

“Various modelling and epidemiological studies (including those from Southeast Asia) have shown that universal masking reduces the spread of SARS-CoV-2

“There are also many mechanistic laboratory studies showing the effectiveness of masks in both containing the virus if worn by those infected, as well as reducing the exposure to incoming virus aerosols for those who are susceptible.  Of course, these are conducted in ideal circumstances, rather than with children fiddling with the masks and moving between classes, etc.  The key is that if enough children wear masks, then the benefits will go beyond those individuals to the wider class – as masks will capture and retain outgoing virus particles and reduce the overall airborne virus concentration that can be inhaled by others.

“Such studies use either human volunteers or manikins that are exposed to appropriately sized particles to act as aerosols to test their penetration through various forms of masks: sometimes we can also use more benign viruses like seasonal influenza and some other bacteriophage (viruses that only infected bacteria) markers:

(bidirectional protection – including home-made masks)

(incoming aerosol protection – surgical and higher respirators against particle and influenza aerosols)

(outgoing aerosol containment – surgical masks with influenza and other seasonal respiratory viruses)

“Also – most people don’t really understand how masks work – they do not just act as a sieve for particles of different size – but function by sedimentation (for larger particles), impaction and interception (for smaller and larger particles), as well as diffusion and electrostatic attraction (for the smallest particles).  Thus, once captured on the mask material, it will be quite difficult to dislodge infectious particles onto hands in sufficient numbers to then lead to infection through self-inoculation:

“So overall, this is a welcome report.  No intervention by itself is perfect – and all interventions work best in combination with others – including vaccination and ventilation in schools and other indoor spaces.”


Prof Jonathan Ball, Professor of Molecular Virology, University of Nottingham, said:

“The evidence-base quoted in the report in support of mask-wearing in schools does seem weak to me.  Without doubt the best way to understand the impact of this kind of intervention would be to have a large, randomised controlled trial, where infection rates in a range of secondary schools where masks are adopted is compared with similar schools where masks are not mandated.  We do this for new treatments and vaccines, but don’t take the same robust approach with non-pharmaceutical interventions.  Some argue that withholding use of an intervention that might decrease risk is unethical – but such a stance assumes in advance that the intervention works – yet in the case of mask-wearing in an education setting this seems far from certain to me.”


Prof Sarah Lewis, Professor of Molecular Epidemiology, University of Bristol, said:

“The report outlines the body of evidence that the decision to recommend face masks in secondary schools rests on.  It is made clear in the report these are largely observational studies which are prone to bias and the findings from these are mixed.

“The report points out that: ‘The most reliable way to measure the impact of the use face coverings would be to run a randomised-controlled trial (RCT) comparing the change in the COVID-19 absence rate between schools randomly assigned to be in the treatment group (i.e., using face coverings) versus schools assigned to the control group (i.e., not using face coverings).’  This is true although the justification for not conducting these, because we are in the middle of a pandemic, is weak given that RCTs of daily testing versus isolation have been carried out in schools and many RCTs of vaccines and potential medications for covid have also been carried out.

“The new study presented in this report is a comparison of covid absence rates 2-3 weeks later in 123 schools which introduced masks on the 1st October 2020 with covid absence rates in 1192 schools which did not have a policy of mask wearing in school.  There were several differences between the two sets of schools included in this study including the covid absence rates at the start of the study (the schools which introduced masks had much higher rates).  The researchers tried to adjust for these factors in their analysis.

“The researchers found in their adjusted analysis covid absence declined from the 1st October to 15-21st October in both the schools which introduced mask wearing and in the control schools, more so in the mask schools but there wasn’t strong evidence of a difference between the two.

“This study shows us that mask wearing was introduced into schools when covid absence was high (probably due to a local outbreak) and 2-3 weeks later covid absence had declined in those schools to have similar covid absence rates to control schools.  However, we do not know whether this was due to mask wearing, because outbreaks tend to subside over time due to changes in behaviour such as increased testing, reducing social contacts, isolation of cases etc.

“The study and research outlined in this report does not provide strong justification for introducing this policy in schools.  The negative impacts of mask wearing on communication and learning in schools are outlined in the report and mask wearing has an especially detrimental effect on those with hearing impairments who are excluded from class discussion.  Where there is insufficient evidence of a benefit of a policy and evidence of harms the default should be not to intervene.”



All our previous output on this subject can be seen at this weblink:



Declared interests

Prof Jonathan Ball: “Involved in an early-stage feasibility study to assess impact of mask wearing in schools to reduce respiratory virus infections.”

None others received.

in this section

filter RoundUps by year

search by tag