[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address Please contact the publisher to request reinstatement.
[Skip to Content Landing]
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    9 Comments for this article
    Analysis of Cases in the General Public Should be Provided
    Ben Apgar | Public
    In the discussion, the authors write "the case number continued to increase in Massachusetts throughout the study period." The reference provided (Massachusetts Department of Public Health COVID-19 dashboard—April 30, 2020), however, is only raw data, not an analysis of the data.

    My unweighted linear regression of that data for the period of April 11 to April 30 shows daily cases decreasing. While I didn't do the work to determine the positivity rate, the number of tests performed increased over this period, suggesting positivity rate was also decreasing over this period.

    This appears to contradict their statement in the
    discussion that their results suggest "the decrease in the SARS-CoV-2 positivity rate in MGB HCWs took place before the decrease in the general public." It appears to have happened at the same time, though perhaps masks had an additional effect.

    Additionally, Massachusetts cases were subject to a data error on April 24th, as reported in the Boston Herald. I am not sure if this is reflected or corrected in the data, but it may deserve a mention to avoid confusion.

    Regardless of my simple analysis, I believe, at minimum, it shows that raw data is insufficient to support the statements in the discussion and their analysis of the cases in the general public needs to be included.
    Confounding of Effects of Masking and Hospital Visits on SARS-CoV-2 Infections in HCWs
    Knut Wittkowski, PhD, DSc | ASDERA, LLC
    Wang et al. (2020) report an increase in the proportion of positive PCR tests to 10% on March 24, which coincides with an increase in proportion of hospital visits with influenza-like illnesses in MA to 6.6%, as reported by the CDC's ILInet, which peaked in week 12, March 16 to 22 (1).

     As the number of ILI visits declined, but the number of hospitalized patients is expected to have increased (although those numbers are not reported), the proportion of positive tests increased at a lower rate and declined during the time when patients are expected to seroconvert (seroconversion events
    are also not reported). As the changes in the number and infectiousness of patients and the effects of masking requirements, first by HCWs and then by patients, overlap, further studies are needed before conclusions can be drawn about the effectiveness of masking requirements in health care settings (by the authors) or for the general public (by the CDC in the accompanying Editorial).


    1. https://gis.cdc.gov/grasp/fluview/fluportaldashboard.html
    Article Title is Deceptive
    Lew Jacobson, PhD | University of Pittsburgh
    The article title posits an association between universal masking and SARS-CoV-2 positivity, but the results reported in the article show the opposite -- masking reduces positivity. This contradictory titling is a repeating error. It could be fixed by changing the title to "Association between universal masking and reduction of SARS-CoV-2 positivity."
    Absolute Event Rates Just as Important as Proportion of Positive Tests
    James Siegler, MD | Cooper University Hospital
    There is little doubt that the use of face masks and other personal protective equipment can slow the rate of new infections with SARS-CoV-2 among healthcare workers (HCWs). But equally important to the proportion of symptomatic HCWs testing positive is the overall number of new positive cases of SARS-CoV-2. How many fewer HCWs are acquiring the infection now, as opposed to before the implementation of universal masking?

    If one were to discount the period during which the rise in new cases were occurring (all dates before March 25th), one may consider the relative plateau of new positive HCW cases
    to take place between March 25th and April 11th. As the authors stated, April 11th is an appropriate threshold at which the effect of mask-wearing could reduce the transmission of SARS-CoV-2. In my crude deconstruction of the figure, where the mean daily testing rate was selected from the available ranges (e.g., 10 for a test rate <20, 60 for 20-100, 150 for 101-200, and 250 arbitrarily selected for days with >200 tests), including all available dates after which the first healthcare worker tested positive, the mean daily rate of new diagnoses fell by 27.5% from a mean of 30.6 ± 9.8 cases/day to 22.2 ± 12.4 between these periods (p=0.03 by independent samples t-test).

    Not knowing the true daily test counts renders this oversimplified analysis limited, so I followed the first comparison with a sensitivity analysis that would intentionally attenuate the effect of mask-wearing. In this analysis, assuming the daily test rate was at the 25th percentile of the reported range before April 11th (e.g., 5 tests on days when the test rate was <20, etc., and 225 for dates with >250 tests)—and the 75th percentile for dates after April 11th (e.g., 15 tests when the test rate was <20, and 275 for dates with >250 tests), the difference was no longer significant (26.8 ± 9.5 cases/day vs. 25.5 ± 13.1, p=0.73). This discrepancy with the former analysis highlights the limitations of the data that are presented. Knowing the daily absolute rate of new infections with SARS-CoV-2 among HCWs would clarify this issue, and a follow-up analysis could (potentially) put to rest the question of whether masks may reduce transmission of SARS-CoV-2 to HCWs.
    Positivity Rate in What Population?
    Carson Kent, PhD Candidate | Stanford University
    From the description of how the positivity rate is calculated, it is unclear what population the positivity rate is being calculated for. If the authors wish to measure the positivity rate within the population of HCWs which are susceptible to SARS-CoV-2 on the i-th day, then the denominator of the positivity rate is the number of HCWs who never tested positive prior to day i. This is not the denominator that is described.
    Wearing a Face Mask is a Sine Qua Non to Stop the Pandemic
    Giovanni Ghirga, MD | Pediatrician, Civitavecchia, Italy.
    This is one of the most important articles published on COVID-19. While in many areas of the world the battle against the SARS-CoV-2 rages in favor of the virus scientists are trying in every way to find a vaccine, a drug, or something else to combat the disease. Nevertheless, this research very clearly shows and confirms to all of us that we already have a highly effective weapon against the virus: “The face mask!” The latter retains the ability to block the survival of the virus by preventing the infection of susceptible individuals. In fact, if new individuals are not infected sooner or later the SARS-CoV-2 will be overwhelmed by the immune system and die. Wearing a face mask when you are even in low-risk areas is a sine qua non to stop the pandemic.

    Giovanni Ghirga, MD, Pediatrician, Civitavecchia, Italy
    The 11% Who Contracted Covid after Universal Masking
    Natasha Mitchell, BSN, MS | University of Pennsylvania
    Assuming, for the sake of argument, that the study's reported 11% incidence of COVID-19 among health care workers after implementation of universal masking is in fact higher than that of the general population at the time of the study, the statistic is intriguing. To what does the group attribute that persistent incidence?

    Particularly in light of the CDC case report on the apparent total efficacy of universal masking in the prevention of COVID transmission between two COVID+ hair dressers and 139 clients they serviced (1) and the fact that universal masking did appear to dramatically reduce transmission rates
    at the Brigham study hospital, that persistent 11% infection rate could hold keys to efficacious transmission prevention even with continued social interaction. Put another way, what is the difference between 11% transmission and 0% transmission? Did 11% of hospital employees contract COVID during the odd moments they weren't wearing masks eating with other staff, drinking, etc.? Did they contract it from patients who weren't wearing masks? Did they contract COVID because the concentration of aerosolized disease was relatively higher than it would be in, say, a hair salon, and therefore an N95 mask would have been required to drive the in-hospital transmission rate towards 0?

    Within these data, and further study, lies the potential for dramatic insight into how to precisely control spread of the virus while allowing people to continue with much of "normal" life. If we could compare disease incidence, for instance, between staff who were rigidly adherent to masking policy (ate alone and outside, never took their masks off during work day) and staff not so strict (ate in the presence of others, with masks down, intermittently pulled mask down for moments of "fresh" air ) in similar work environments, we would perhaps identify high-risk activities which folks could relatively easily avoid without having to avoid human interaction all together (i.e. go ahead and hang out...but don't even think about pulling the mask down so that you can enjoy a meal together); rates of of staff COVID infection where relatively higher proportions of patients wear masks vs. those where relatively lower proportions of patients wear masks; rates of infection in places where aerosolized viral load is is likely to be higher (e.g. hospital ER or ICU) vs. place where it's likely to be relatively lower (hair salons): we might find, for instance, that it really is safe to go to the salon so long as you're wearing a well-fitting, simple face covering...but everyone in a hospital ought to be wearing an N95.

    I am not in a position to pursue such research but I imagine others here are (and perhaps such study is already under way).


    1. https://www.cdc.gov/mmwr/volumes/69/wr/mm6928e2.htm?s_cid=mm6928e2_w&fbclid=IwAR1ID1xVHAYH_mF8qE9SjbwsNidSJufpGqCtmpvOOBLSapyeC6H7xEXbF4o
    Moral Arguments for Masks
    Eta Berner, EdD | University of Alabama at Birmingham
    In their editorial comment on this article, Brooks et al. recommend “appealing messaging that normalizes mask wearing.”(1) However the major thrust of most arguments to persuade the public about masks is that although the recommended cloth mask may not provide virus protection to the individual wearing it, the mask can prevent transmission to others. The assumption is that the general public will be persuaded by this argument of public good, even at the cost of personal inconvenience. However, the pictures and stories of US citizens ignoring the recommendations should make us reconsider that assumption.
    /> Over 50 years ago, the late Lawrence Kohlberg formulated six stages of moral reasoning that are still used in studies of moral reasoning today (2-4). Kohlberg’s ideas may provide some direction to persuade a larger portion of the public about the importance of wearing masks. Data have shown that most people reason at the two stages of what Kohlberg calls the conventional level, where the reasoning focuses on doing what is polite, “nice”, being a good person, conforming to community norms, and, at the higher conventional level, following laws and other rules. Furthermore, individuals are able to comprehend arguments only slightly above their usual stage of reasoning (2). The problem with the argument that individuals should inconvenience themselves for the public good is that the argument reflects the higher stages of moral reasoning (post-conventional), which may not be persuasive to the many individuals who do not normally reason at that level. The higher moral (and scientifically accurate) argument promoting the public good should still be made, since some individuals may respond to it. However, an additional argument that emphasizes that a good, responsible, considerate person will follow the rules may be more effective in reaching a larger number of people in a way that they can understand and may be the best way to persuade the largest number of people to do what is best for the public good.


    1. Brooks JT, Butler JC, Redfield RR. Universal masking to prevent SARS-CoV-2 transmission—The time is now. JAMA. Published online July 14, 2020. doi:10.1001/jama.2020.13107
    2. Rest J, Turiel E, Kohlberg L. Level of moral development as a determinant of preference and comprehension of moral judgments made by others. J Pers. 1969 Jun;37(2):225-52.
    3. McLeod-Sordjan R. Evaluating moral reasoning in nursing education. Nurs Ethics. 2014;21(4):473-483.
    4. Fang Z, Jung WH, Korczykowski M, et al. Post-conventional moral reasoning is associated with increased ventral striatal activity at rest and during task. Sci Rep. 2017;7(1):7105.
    A Possible Confounder
    Abigail Zuger, MD (1); Kent A. Sepkowitz, MD (2) | (1) None (2) Memorial Sloan Kettering Cancer Center
    Wang et al (1) seek to demonstrate that a hospital mandate for universal masking reduced COVID-19 infection rates among its staff. They report that at baseline 21% of symptomatic staff seeking SARS-CoV-2 testing were confirmed to be infected; this rate declined to 15% after staff masking was mandated, and then to 11% after patient masking was mandated. The authors note that their results may reflect the effects of other hospital and municipal infection control measures enacted during the study period. Still, they conclude the data strongly support the effectiveness of universal masking in health care settings -- and other authors extend this conclusion to other settings. (2)

    We would like to propose an important confounding variable in their observation: the strong possibility that clinical staff at a tertiary care academic medical center in the Northeast US in late March 2020 were already using masks well before any official hospital policy went into effect. The risks of nosocomial COVID-19 may have been small in Boston at that time, but the disease's transmission patterns were clear, and Boston's influenza season was still active (3). We suspect that many or most medical and nursing staff, transporters, housekeepers and others were masked, at least when caring for patients with respiratory infections, by the time masking became compulsory.

    If this is the case, other explanations for the sharp decrease in employee infection rates noted after March 26 deserve consideration. Perhaps internal SARS-CoV-2 testing criteria were informally liberalized during the study period to include some of the worried well. Perhaps a single cluster of workplace or community infection drove the spike in infection rates seen at the end of March. An analysis of the specific duties and home zip codes of infected employees might have clarified some of these questions. A precise quantitation of hospital mask use before and during the study period would also have informed the argument.

    Anyone who has worked in a hospital has learned that -- particularly during emergencies -- nothing influences behavior quite so powerfully as do informal negotiations and the group survival instincts born of long camaraderie. Although we may all wish for some numerical proof of the efficacy of universal masking, official hospital policy is unlikely to be the best place to find it.

    Abigail Zuger MD
    New York City

    Kent A. Sepkowitz, MD
    Deputy Physician in Chief, Quality & Safety
    Memorial Sloan Kettering Cancer Center
    New York City


    1. Wang X et al. Association between universal masking in a health care system and SARS CoV 2 positivity among health care workers. Jama online 7/14/2020. https://jamanetwork.com/journals/jama/fullarticle/2768533

    2. Brooks JT et al. Universal masking to prevent sars-CoV-2 transmission -- the Time is Now. Jama online 7/14/2020. https://jamanetwork.com/journals/jama/fullarticle/2768532

    3. Massachusetts Department of Public Health weekly influenza update for March 20, 2020, accessed 8/4/20. https://www.mass.gov/info-details/weekly-flu-report-march-20-2020
    Views 216,549
    Citations 0
    Research Letter
    July 14, 2020

    Association Between Universal Masking in a Health Care System and SARS-CoV-2 Positivity Among Health Care Workers

    Author Affiliations
    • 1Division of Cardiovascular Medicine, Brigham and Women’s Hospital, Boston, Massachusetts
    • 2Department of Medicine, Brigham and Women’s Hospital, Boston, Massachusetts
    • 3Center for Clinical Investigation, Brigham and Women’s Hospital, Boston, Massachusetts
    • 4Occupational Health Services, Mass General Brigham, Boston, Massachusetts
    JAMA. Published online July 14, 2020. doi:10.1001/jama.2020.12897

    The coronavirus disease 2019 (COVID-19) pandemic, caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has severely affected health care workers (HCWs).1 As a result, hospital systems began testing HCWs2 and implementing infection control measures to mitigate workforce depletion and prevent disease spread.3 Mass General Brigham (MGB) is the largest health care system in Massachusetts, with 12 hospitals and more than 75 000 employees. In March 2020, MGB implemented a multipronged infection reduction strategy involving systematic testing of symptomatic HCWs and universal masking of all HCWs and patients with surgical masks.4 This study assessed the association of hospital masking policies with the SARS-CoV-2 infection rate among HCWs.