Revisiting March 2020: What Were Epidemiologists Thinking about Masks?

Lacking clear information on the benefits of community-level face mask use,epidemiologists in early 2020 engaged in internal discussion about the potential harms and benefits of this intervention, considering aspects such as the limited existing research, the limited supply and interrupted supply chains of masks, what was known at the time about the epidemiology of SARS-CoV-2 transmission, and concerns around the potential for “risk compensation” if people who were wearing masks then engaged in fewer other preventive measures (Bamber and Christmas 2020; Brosseau 2020; Brosseau and Sietsema 2020; Cheng 2020; Javid, Weekes, and Matheson 2020; King 2020). Based on these discussions, many applied epidemiologists, including those at the World Health Organization and Centers for Disease Control, initially advised against the use of face masks by the general public. Instead, they stressed the importance of hygiene and distancing-based interventions, such as hand-washing, social distancing, and quarantine.

Over time, however, new information emerged. First, it became clear that at least some subset of Americans would be amenable to wearing masks. Second, we learned that SARS-CoV-2 could be transmitted by individuals who were not (yet) symptomatic (Gandhi, Yokoe, and Havlir 2020). Finally, as the availability and use of both fabric and surgical masks increased, it became clear that even when individuals wearing masks did increase their risk behaviors (by, for example, joining protests), the evidence did not suggest that transmission in these settings was any higher than if attendees had been unmasked (Dave et al. 2020). Together, these observations have shifted most applied epidemiologists and public health officials towards encouraging the use of face masks by all individuals (Greenhalgh et al. 2020; Roderick et al. 2020).

However, this recommendation does not mean that the academic epidemiology of face mask usage by the general public during respiratory outbreaks has necessarily advanced much beyond what we knew in January 2020, and many academic epidemiologists remain agnostic about the value of face masks. In fact, if anything, it may be fair to say that academic epidemiologists have fewer answers about the science of face masks than we did 10 months ago—simply because we now have more questions. 

Previous research on face mask usage in respiratory outbreaks focused chiefly on evaluating either N95 masks or surgical masks, both of which are subject to regulatory standards. In contrast, many of the face masks used by the general public during the COVID-19 pandemic are made from fabric, both commercially and homemade, and the filtration efficacy of these masks is both unknown and potentially highly variable (Aydin et al. 2020; Davies et al. 2013; Tcharkhtchi et al. 2020). In addition, previous studies of face mask usage typically assumed individuals had been provided with training and guidance on how to appropriately don, doff, and wear face masks to maximize their benefits. In reality, adherence both in terms of frequency and correctness of face mask use is extremely variable among the general public. Despite this, existing attempts to model the population impacts of community-level face mask use have typically assumed perfect adherence and correct usage (Ferguson et al. 2020). Academic epidemiologists likely will be investigating and debating these topics for many years to come, both to fully characterize the causal effect of community level mask-wearing strategies and to explore the actual risks and benefits that result from these (Bundgaard et al. 2020; Doung-ngern et al. 2020).

In short, the current recommendation of experts is to wear masks--even cloth masks--in many settings, not so much to protect yourself as to protect others. Lynn Peeples offers a nice readable overview of the existing evidence base in favor of wearing masks in Nature (October 6, "Face masks: what the data say.
The science supports that face coverings are saving lives during the coronavirus pandemic, and yet the debate trundles on. How much evidence is enough?" But Peeples is a fair-minded presenter of the evidence, and so she also includes statements like: 

I wear a mask in all the recommended settings, because wearing a mask seems very likely to be better than not wearing one. But I'm also aware that countries of the European Union have recently seen a huge growth in COVID-19 cases, from well below the US level to well above it, despite widespread mask-wearing in many countries. As the WHO kept saying as it kept revising its guidance about masks, masks aren't enough---or at least certainly not the fabric or gauze masks that most of the public wear. They need to be combined with social distancing, hand-washing, self-quarantining when possibly or actually infected, and similar steps. If wearing a mask gives people a sense that they are incapable of transmitting the disease themselves or invulnerable to the disease when transmitted by others, then when the epidemiologists finish their studies of masks five or ten years down the road, they may find that the potential benefits of mask-wearing were offset when people reduced their other efforts to minimize the spread of COVID-19. 

I understand that opinions evolve. But at least in theory, public health experts have been holding conferences and publishing learned reports about pandemics for years. Wearing masks is not a high-tech intervention. It seems like the kind of issue where one might expect that public health experts have worked out a recommendation in advance, rather than making a parade of their indecisiveness--and then criticizing those who are skeptical about playing follow-the-leader as they switch recommendations.