Revisiting the evidence for physical distancing, face masks, and eye protection

Derek Chu and colleagues¹ concluded, based on an analysis of a subgroup of observational studies, that health-care workers might afford greater protection against SARS-CoV-2 infection from N95 respirators than from surgical masks. They acknowledge substantial limitations and rated certainty of effect as low. We would argue it is lower still, as several studies seem to have been misclassified with regard to mask type.

Yet in the linked Comment, C Raina MacIntyre and Quanyi Wang² stated that, based on those findings, N95s should be standard of care for all health-care workers working on COVID-19 wards and that guidelines³ be promptly reviewed. This statement disregards the important limitations of observational studies (eg, recall bias and limited ability to control for additional exposures), analytical shortcomings, and that the difference in protection between masks was statistically significant only when accounting for aerosol-generating procedures, consistent with guidelines.³ Also ignored were multiple systematic reviews of N95s versus surgical masks that consistently found no significant differences in risk of respiratory illness, influenza-like illness, or the most robust outcome of laboratory-confirmed viral infection (including non-influenza respiratory viruses), after adjustment for clustering.4, 5

Responsible policy recommendations should weigh the totality of available evidence, with major consideration given to quality. To call for sweeping changes in policy² based on low-certainty findings that are highly susceptible to bias and contradict higher quality evidence is scientifically unjustified, and it does a disservice to front-line health-care workers relying on balanced, evidence-informed recommendations to guide use of personal protective equipment.

JC reports a grant from Pfizer for a randomised controlled trial with a Staphylococcus aureus vaccine versus placebo in vertebral spinal surgery with instrumentation, for which all funds were provided to the University of Calgary, outside this Correspondence and grants from the Canadian Institutes for Health Research on acute and primary care preparedness for COVID-19 in Alberta, Canada, awarded February, 2020. He serves as chair of the WHO Infection Prevention and Control (IPC) Research and Development Expert Group for COVID-19 and the WHO Health Emergencies Programme Ad-hoc COVID-19 IPC Guidance Development Group, both of which provide multidisciplinary advice to WHO and for which no funding is received. He is also a member of the Cochrane Acute Respiratory Infections Group and reports support from the Centers for Disease Control and Prevention to attend an Infection Control Think Tank Meeting in 2019. RC reports grants from WHO, outside this Correspondence. AV and MJS declare no competing interests.