A new meta-analysis raises questions about efficacy of respirators and masks despite known poor adherence and other limitations highlighted by a secondary analysis.

The COVID-19 pandemic has challenged the concept that N95 respirators and surgical masks can lower risk for respiratory virus infections (RVIs). A new Cochrane review on this topic has reported that masking is not beneficial despite the authors' cautions about study interpretation and limitations. In an update of four Cochrane Library meta-analyses on differing physical interventions to limit RVIs, Jefferson and colleagues analysed 78 randomised controlled trials (RCTs; 6 of which were conducted during the COVID-19 pandemic). Study settings included suburban schools, inpatient hospital wards, and inner-city venues; adherence to interventions was often low.

Analysis of 19 trials of hand hygiene (71,210 participants) found a relative reduction of 11% in respiratory illness. Similar assessment of 12 trials (10 in a community setting) of surgical masks found no apparent benefit in preventing influenza-like illness (ILI) or laboratory-confirmed RVI. Analysis of 5 studies (all before the COVID-19 pandemic) comparing N95 respirators with surgical masks found a non–statistically significant benefit of N95 respirators in preventing ILI but not laboratory-confirmed illness. The authors note that, while the results showed no benefit of N95 respirators or masks, high risk for bias and low adherence to the interventions limited the conclusions.

Although the initial study found no efficacy difference for the two forms of protection, in this analysis, the researchers sought to determine if household exposure to preschool children could have affected risk for RVIs in the study population. In 4476 HCW-seasons, while household exposure to preschool children had no effect on incidence of influenza infections, overall incidence of RVIs in HCWs progressively rose with exposure to 1 or ≥2 preschool children.

The RCT is often considered the gold standard of clinical trials, but as the study by Most et al. shows, unrecognised factors can still affect the results — particularly for studies of RVIs, where individuals can be exposed to infection during other parts of their day and not simply in the controlled setting of a clinical trial. The conclusions of the Cochrane review by Jefferson et al. are further limited in that it only assessed the benefit of respiratory protection for the wearer, and not for preventing viral transmission to others.

Also, the review's design led to the exclusion of laboratory studies on mask efficacy and well-executed non-RCTs, including a test-negative case-control study of respiratory protection and a difference-in-differences analysis of the impact on COVID-19 incidence of discontinuing universal masking in public schools. Both studies found notable benefits of mask or respirator use for protection against COVID-19.

The body of evidence overwhelmingly supports the benefits of wearing masks; the Cochrane review might be best interpreted as showing that masks don't provide protection when they are not worn.

Source: Journal Watch

Source link