Study: COVID-19 risk slightly lower for mask wearers
Click Here to Manage Email Alerts
The CDC recently updated its guidance to state that face masks provide some personal protection against COVID-19 for those who wear them, not just for those around them.
New data from a randomized controlled study published in Annals of Internal Medicine showed that wearing a mask was slightly more effective in preventing COVID-19 than not wearing a mask in situations where other preventive measures like physical distancing are recommended.
Henning Bundgaard, MD, DMSc, a professor and consultant in cardiology at the University of Copenhagen and coauthor of the study, told Healio Primary Care that it is important to note that the researchers only assessed the effect of masks as personal protection — not “source control.”
“The identified potential positive effect of the mask for prevention — although small — might still be of interest considering COVID-19 is a very serious disease,” he added.
Bundgaard and colleagues conducted a randomized controlled study in Denmark from April 3 to June 2, where participants who spent at least 3 hours per day outside of their home were randomly assigned to wear a mask or to not wear a mask when away from home.
Those asked to wear a mask were given 50 high-quality surgical face masks with a 98% filtration rate.
At the time of the study, Danish authorities did not recommend the use of masks, and less than 5% of people outside of hospitals used masks, according to the researchers. However, authorities recommended other preventive measure during the study period such as quarantining people with COVID-19, physical distancing measures, limiting contact with people, frequent handwashing and cleaning and limiting visitors in hospitals and nursing homes. Denmark’s cafes and restaurants were closed from the start of the study period until May 18.
Among participants, the researchers evaluated the prevalence of SARS-CoV-2 infection at 1 month based on positive antibody results, polymerase chain reaction or diagnosis at a hospital.
A total of 3,030 participants were assigned to wear masks and 2,994 were assigned to not wear masks.
Bundgaard and colleagues found that 1.8% of those who wore masks and 2.1% of those who did not wear masks tested positive for SARS-CoV-2 infection. They determined that the difference between the groups was -0.03 percentage points (OR = 0.82; 95% CI, 0.54-1.23), with the findings in favor of those who wore masks.
The researchers noted that additional analyses accounting for participants with missing follow-up data yielded similar results.
They said that while the difference between the two groups was not statistically significant, the data were “inconclusive,” with mask wearing demonstrating a potential 46% decrease to a 23% increase in COVID-19.
Bundgaard said that previous observational studies have shown that masking is effective, but the studies are challenging to interpret due to the limitations of observational data. However, he added that laboratory research has shown that wearing masks appears to be effective.
He noted that additional randomized studies on mask wearing will be difficult to perform because face masks are mandatory in most places.
In an editorial accompanying the study, Christine Laine, MD, MPH, editor in chief of Annals of Internal Medicine, and colleagues stressed that the findings do not address the effectiveness of masks in preventing transmission in areas were most people mask, and that the findings do not disprove the effectiveness of masking.
Laine and colleagues said CDC guidance recognizes that mask wearing can reduce transmission when worn by everyone. They added that the study by Bundgaard and colleagues suggests that when masks are worn by only a minority of people, those who wear them are not “invulnerable” to COVID-19, and that the impact of mask wearing on personal protection is probably small.
“All who worry about the COVID-19 pandemic should carefully consider these findings for what they show and refrain from viewing them as evidence that widespread mask-wearing is ineffective,” they wrote. “While we await additional evidence about the effectiveness of masks as source control of SARS-CoV-2 transmission, we are going to do our part to protect everyone by masking up and hope that those around us do the same.”
In another editorial published alongside the study, Former CDC Director Thomas R. Frieden, MD, MPH, and Shama Cash-Goldwasser, MD, MPH, of Resolve to Save Lives, an initiative of Vital Strategies in New York, highlighted several limitations of the study, including that it was conducted in an area with relatively low transmission and that antibody tests were used to diagnose COVID-19.
“Although no single strategy can control the pandemic, widespread masking in the community can mitigate spread as part of a comprehensive approach,” they wrote.
Frieden and Cash-Goldwasser added that “if everyone wears a mask when near others, everyone is safer.”
References:
Bundgaard H, et al. Ann Intern Med. 2020;doi:10.7326/M20-6817.
CDC. Scientific brief: Community use of cloth masks to control the spread of SARS-CoV-2. https://www.cdc.gov/coronavirus/2019-ncov/more/masking-science-sars-cov2.html. Accessed Nov. 18, 2020.
Laine C, et al. Ann Intern Med. 2020;doi:10.7326/M20-7448.
Frieden TR, et al. Ann Intern Med. 2020;doi:10.7326/M20-7499.
Perspective
Back to Top
The study team — cardiologists in Denmark — should be commended for performing a randomized controlled trial (RCT) of a nonpharmaceutical intervention — mask wearing — in the early days of a pandemic. However, there are major design flaws to this study that will impair our ability to draw conclusions from DANMASK. The study was a nonblinded study of providing 50 surgical face masks for community participants spending 3 hours or more outside of the home in April to May (intervention group) vs. the standard of care of Denmark: not recommending widespread use of face masks in the community at the time (although other European countries and the U.S. CDC had provided this recommendation in earl April). Of note, this study did not have institutional review board (IRB) approval as the local IRB apparently said approval was not needed. The study shows only a small reduction in incidence of SARS-CoV-2 to those randomized to the mask intervention as measured by PCR or antibody testing.
The major design flaws are as follows:
1. Underpowering: This study was conducted during a time of declining incidence in Denmark. The country was on partial lockdown. Therefore, the study was actually underpowered (classic Type II error) to look at the outcome. Indeed, there were very few outcomes in this trial, which demonstrates this limit in statistical power, regardless of the authors’ sample size calculations.
2. Randomization at individual rather than community level: Other studies have shown that the most common source of infection with SARS-CoV-2 is from household contacts. By designing the study as an individual-level rather than a cluster-randomized household or community-level RCT, investigators significantly diminished the ability of the study to find an effect of mask-wearing. Household members were not randomized to the same intervention. Household members generally do not wear masks in the home, so the individual randomized to the intervention would be at higher risk of infection from their household members (not in the intervention) who remain at higher risk of infection.This individual-level design becomes an Achilles heel for the intervention (the statement that participants reported few infections within their households does not assuage this concern — many infections and transmissions are asymptomatic and PCR testing was limited at this point in the pandemic, so most study outcomes were assessed via positive serology tests)
3. Flawed outcome metrics: Use of a home point-of-care antibody test with limited sensitivity (sensitivity is likely even less than 83% given that samples used in validation studies typically come from symptomatic participants with positive PCR tests) decreased the power of the study and limited the number of outcomes further. Moreover, the authors mention the hypothesis (advanced by our group and others) that mask wearing could increase the proportion of asymptomatic infections if infection does occur, but this study is not designed to show this difference. Antibodies can fade quickly and may decrease to undetectable levels after asymptomatic infection more quickly; T-cell assays are more useful to detect past infection. Participants had to prick their own finger and perform home testing and then mail their results back in (subject to underreporting, forgetting). Finally, for the more limited PCR-based tests that were recorded, the authors used a platform involving self-collection of oropharyngeal of nasopharyngeal swabs (instead of collection by health care personnel) which can be limited by familiarity with swabbing and is non-standardized.
4. Self-reporting of results introduces bias: Use of self-reported results on this test introduces further measurement error and potential bias, in an unknown direction, which is impossible to sort out without confirmatory testing.
5. Low incidence will bias results to the null: Given low community-level adherence to masking during this time period, masks were at a disadvantage, given that they are more effective at preventing infection from an infectious source.
6. Community norms and global recommendations could influence mask wearing in each group, leading to misreporting: Early in the study, it is apparent that participants in the intervention experienced social harms from wearing the masks, possibly leading to low adherence and over-reporting of mask adherence. Later in the study, the control group was likely to be influenced by increasing promotion of masking as a nonpharmaceutical intervention worldwide, and likely underreported their mask wearing (and there was no assessment of this). Twenty percent of participants did not complete the study.
7. Assessment of risk: There is no assessment of the nature of SARS-CoV-2 exposure in the two groups. They are spending less than 3 hours out of the house? Doing what? Occupational? The country was under partial lockdown at the time, so it is very helpful to understand their exposure.
8. Definitions of adherence: Adherence to mask wearing in the intervention group is quite low (47%) and we know from many other fields and studies that self-reported adherence usually overestimates actual adherence. Moreover, we have no assessment of adherence in relation to exposure. There is just no granularity. Forty-seven percent adherence is one thing if it is during times of likely exposure vs. not.
9. Exploration of viral inoculum hypothesis: In terms of exploring whether mask wearers could have more milder infections, there are no details on the clinical course of the detected infections. Therefore, although mentioned, this doesn’t appear to be investigated at all.
These factors, when taken with the low-test sensitivity, bias in self-reporting, and lack of masking at the household or community level, resulted in a trial that was ill-equipped to study the impact of masking on prevention of SARS-CoV-2 transmission. Moreover, testing this question via an RCT (rather an observational or implementation design) is likely not possible because of the contamination issues; it is easy for those randomized to not wear masks to “drop in” and adherence in those randomized to wearing masks is both hard — and critical — to assess. Of note, the authors of this study acknowledge many of the limitations above.
Since mentioned in the article, our group’s hypothesis that social distancing and masks reduce the severity of disease via reducing the viral inoculum, even if transmission occurs, is gaining more evidence, but requires further study. In conclusion, there is increasing evidence that community-level masking reduces transmission by protecting oneself (personal filtration effect) and others (source control), ranging from a series of physical sciences studies to observational studies that show that universal mask wearing reduces transmission.
References:
Bielecki M, et al. Clin Infect Dis. 2020;doi:10.1093/cid/ciaa889.
CDC. Scientific Brief: Community Use of Cloth Masks to Control the Spread of SARS-CoV-2. https://www.cdc.gov/coronavirus/2019-ncov/more/masking-science-sars-cov2.html. Accessed November 10, 2020.
Chan JF, et al. Clin Infect Dis. 2020;doi:10.1093/cid/ciaa644.
Gallaway MS, et al. MMWR Morb Mortal Wkly Rep. 2020;doi:10.15585/mmwr.mm6940e3.
Gandhi M, et al. J Gen Intern Med. 2020;doi:10.1007/s11606-020-06067-8.
Gandhi M, Rutherford GW. N Engl J Med. 2020;doi:10.1056/NEJMp2026913.
Hendrix MJ, et al. MMWR Morb Mortal Wkly Rep. 2020;doi:10.15585/mmwr.mm6928e2.
Imai M, et al. Proc Natl Acad Sci U S A. 2020;doi:10.1073/pnas.2009799117.
Lyu W, Wehby GL. Health Aff (Millwood). 2020;doi:10.1377/hlthaff.2020.00818.
Sims MD, et al. Clin Infect Dis. 2020;doi:10.1093/cid/ciaa1684.
University of Kansas. Do Masks Matter in Kansas? https://ipsr.ku.edu/covid19/images/MaskMandateUpdate.pdf. Accessed November 10, 2020.
Vanderbilt University Medical Center. Study finds areas without mask requirements have larger increase in COVID-19 hospitalizations. https://news.vumc.org/2020/10/27/study-finds-areas-without-mask-requirements-have-larger-increase-in-covid-19-hospitalizations/. Accessed November 10, 2020.
Wang X, et al. JAMA; 2020;doi:10.1001/jama.2020.12897.
Collapse
Read more about
Click Here to Manage Email Alerts