Do mask mandates work?

This week there was an update to a Cochrane review, which studies the way physical interventions can interrupt or reduce the spread of respiratory viruses. The review, which Tom Jefferson is the lead author of, looks at evidence from 78 randomised trials with over 610,000 participants. In other words, this review is exactly the sort of higher-quality evidence you want when making healthcare decisions.  

The review’s fifth update looked at handwashing, antiseptic use, social distancing and barriers such as masks, gloves, gowns and visors.

Given past controversies, it’s worth looking at what the review says about the effects medical or surgical masks have on the way respiratory diseases spread.  

Interestingly, 12 trials in the review, ten in the community and two among healthcare workers, found that wearing masks in the community probably makes little or no difference to influenza-like or Covid-19-like illness transmission. Equally, the review found that masks had no effect on laboratory-confirmed influenza or SARS-CoV-2 outcomes. Five other trials showed no difference between one type of mask over another.

This is the second update of the review since the start of the pandemic. The first update was delayed by seven months due to unexplained editorial decisions. It was too late when it came out in November 2020 to make a difference to national Covid policy; by then, activism, low-quality observational evidence and government policy had set the agenda for mask mandates, and the damage had been done.  

Often these government policies relied on observational studies on mask usage and the spread of Covid. But there are lots of flaws in observational evidence. For example, in the absence of a study protocol setting out methods before the study is done, it is possible to shift the dates of an observational analysis to suit the rise and fall in infections. So if you time your analysis near the peak of infections, the results will favour mask interventions as the infection rate quickly decreases.  

But when we pointed out in November 2020 the troubling lack of robust evidence on face masks and the problems with observational studies, we were shouted down, removed from Facebook and put on the government’s secret watchlist.   

What many also failed to notice at the time was that studies that look at individuals – as opposed to populations – can lead to erroneous policy decisions. Studies which involve individuals frequently track people who have specifically chosen to wear a mask. But policies on mandatory masks are very different – they involve lots of people who don’t like wearing masks every day, and many people who won’t wear one at all. A study which only looks at keen mask-wearers will not reflect how people comply on a population level. 

The Cochrane review findings report relatively low adherence to mask-wearing, which is similar to what happens in the real world. With better adherence and higher quality masks (and if you are careful when you step out the door), you might reduce your risk in specific settings by a small amount. However, when you scale up any potential small benefits to those who step out the door regularly, the effect doesn’t stack up as a population-based intervention.  

Mandates that affected the whole population never made sense. Moreover, even in high adherence populations such as Japan, they have not stemmed an inevitable rise in infections. Part of the problem may be that during the pandemic the government had to be seen to be doing something. Interventions like handwashing and vaccines are invisible, but masks acted as a visible sign of compliance.  

What we have witnessed in this pandemic are strong beliefs about what works and what doesn’t. At times, it’s been more like a football match, with cheerleaders on either side goading the opposition. Several policies such as mask mandates, restrictions, and unproven interventions now seem absurd in hindsight. And as the culture of fear has lifted, the population has become all too aware of their detrimental effects.  

We failed to follow an evidence-based approach during the pandemic. We are now left with the human, social and economic aftermath of evidence-free policies.