Large Review Finds Good Evidence for Masks, Distancing in Stopping COVID-19

Non-pharmaceutical interventions, such as physical distancing of more than about 3 feet and wearing a medical mask and eye goggles, may be associated with reductions in COVID-19 transmission, a systematic review and meta-analysis found.

A review of at least 172 studies on SARS, MERS, and SARS-CoV-2, the virus that causes COVID-19, found transmission of viruses was lower with physical distancing of more than about 3 feet versus less than about 3 feet, reported Holger J Schünemann, MD, of McMaster University in Ontario, Canada, and colleagues of the COVID-19 Systematic Urgent Review Effort.

Both face masks and eye protection could be associated with reduction in risk of infection, although the evidence for both was “low certainty,” the authors wrote in the Lancet.

Notably, they added there have been no randomized trials for COVID-19 about the effectiveness of these interventions, nor were any randomized trials included in this analysis.

Schünemann said in a statement that government and the public health community can use these results to provide advice for both community settings and healthcare workers about how to reduce risk.

“Our findings are the first to [synthesize] all direct information on COVID-19, SARS, and MERS, and provide the currently best available evidence on the optimum use of these common and simple interventions to help ‘flatten the curve’ and inform pandemic response efforts in the community,” he noted.

In an accompanying editorial, C. Raina MacIntyre, PhD, of University of New South Wales in Sydney, and Quanyi Wang, PhD, of the Beijing Center for Disease Control and Prevention in China, highlighted several of the review’s findings, including a subanalysis that found N95 respirators were 96% effective, considerably more so than other mask types (77% effective).

“For healthcare workers on COVID-19 wards, a respirator should be the minimum standard of care. This study … should prompt a review of all guidelines that recommend a medical mask for health workers caring for COVID-19 patients,” the editorialists wrote.

The Infectious Diseases Society of America’s current guidelines for healthcare providers recommend use of either N95 respirators or surgical masks as part of routine patient care when dealing with patients with suspected COVID-19, though for aerosol-generating procedures, N95 respirators are recommended over surgical masks.

MacIntyre and Wang also noted the finding that respirators and multi-layer masks are more protective than single-layer masks, which is “vital to inform the proliferation of home-made cloth mask designs, many of which are single-layered.” They added a well-designed cloth mask should have “water resistant fabric, multiple layers and a good facial fit.”

“This study supports universal face mask use, because masks were equally effective in both health-care and community settings when adjusted for type of mask use,” the editorialists argued, citing a recent study out of China that found masks worn at home prevented secondary transmission if worn before symptom onset of the index case.

Schünemann and colleagues examined 172 observational studies across 16 countries and six continents, as well as 44 relevant comparative studies in healthcare and non-healthcare settings, for a total of around 26,000 patients. They used the GRADE approach to rate the certainty of evidence.

Overall, 29 unadjusted studies (RR 0.30, 95% CI 0.20-0.44) and nine adjusted studies (aOR 0.18, 95% CI 0.09-0.38) found “a strong association” between proximity and risk of infection, including six studies on COVID-19, with “moderate certainty.”

For face mask versus no face mask, 29 unadjusted studies (RR 0.34, 95% CI 0.26-0.45) and 10 adjusted studies (aOR 0.15, 95% CI 0.07-0.34) on face masks versus no face masks found a “large reduction in risk of infection,” albeit with “low certainty.”

Eye protection, such as face shield and goggles, was associated with reduced risk versus no eye protection in 13 unadjusted studies (RR 0.34, 95% CI 0.22-0.52), but with “low certainty.”

The study’s primary limitation was that none of these were randomized trials, and may have suffered from recall and measurement bias.

“Globally collaborative and well conducted studies, including randomised trials, of different personal protective strategies are needed regardless of the challenges,” the authors concluded.