As recently as March 27, the WHO was stating on its website that there was insufficient evidence to suggest that the novel coronavirus was airborne except in a handful of medical cases, such as when intubating an infected patient.
An April 2 article published online by the journal Nature reported this information and noted that “airborne” means carried in aerosols with particle sizes less than 5 micrometers (5 μm) in diameter that can linger or travel further in the air, as opposed to the larger, heavier droplets formed when someone sneezes or coughs and which rapidly fall to the ground or intervening surfaces.
On April 1, Dr. Harvey Fineberg, Chair of the Standing Committee on Emerging Infectious Diseases and 21st Century Health Threats of the National Academies of Sciences, Engineering and Medicine, and former President of the Institute of Medicine and Dean of Harvard’s Faculty of Public Health, wrote to Dr. Kelvin Droegemeier of the White House Office of Science and Technology Policy regarding the possibility that the novel coronavirus could be spread simply by a conversation in addition to sneezing and coughing. The letter summarized the limited quantity of recent scientific literature on the subject (discussed below), but concluded that the results were “consistent with aerosolization of virus from normal breathing.” However, it cautioned that the widely used RT-PCR test only detects viral RNA (the genetic contents of the virus particle) in air droplets and aerosols, so this does not necessarily equate with the presence of viable (“live”) virus in amounts sufficient to produce infection.
The conclusions of Dr. Fineberg’s letter were widely reported in the press on April 2. On April 3, CDC, also citing new evidence from recent studies, published new guidance for the general public, reversing its previous stance and recommending the wearing of “cloth face coverings in public settings where other social distancing measures are difficult to maintain (e.g., grocery stores and pharmacies) especially in areas of significant community-based transmission.” The cloth face coverings could be made from household items or other common materials, but should not be surgical masks or N95 respirators, which must be reserved for health care workers and other first responders (and which are, in any case, now generally unobtainable by the general public).
That same day, April 3, Nature Medicine published a Brief Communication by Leung et al. from the University of Hong Kong and the University of Maryland School of Public Health entitled “Respiratory virus shedding in exhaled breath and efficacy of face masks.” (Dr. Fineberg’s letter cited an earlier draft that had not yet been peer-reviewed.) From a group of 246 participants, 50% were randomly selected to wear a surgical face mask during a 30-minute collection of exhaled breath while the other 50% did not wear a mask. (The authors noted that surgical masks were originally introduced to prevent surgeons – the wearers – from passing infections to their patients, and their use was only later extended to protect health care workers from infections by their patients.)
Of the 246 participants in the study, 50% (123) were infected with at least one respiratory virus, of which 90% (111) had among them:
- Seasonal human coronavirus (a form of common cold) – 11 subjects; and/or
- Influenza virus – 43 subjects; and/or
- Rhinovirus (another form of common cold) – 54 subjects
Respiratory droplets (>5 μm) and aerosols (≤5 μm) in the exhaled breath from these infected subjects were separately analyzed for detectable virus.
Overall, viruses were detected in significant percentages of both respiratory droplet samples (26-30%) and aerosol samples (35-56%) from subjects without face masks, the latter result indicating that virus particles may travel further than previously thought. Face masks appeared to be effective at retaining seasonal human coronavirus: it was detected in 30% of droplet samples and 40% of aerosol samples without a face mask, but in 0% of either droplet or aerosol samples when face masks were worn. Face masks greatly reduced the transmission of the influenza virus in droplets, but not aerosols. However, face masks appeared to be very inefficient at retaining rhinovirus when present in either droplets or aerosols.
Dr. Fineberg’s letter acknowledged that Leung et al.’s “findings suggest that surgical facemasks could reduce transmission of human coronavirus.” Overall, however, the number of subjects in the study with seasonal human coronavirus was low, so it is difficult to make more general claims from this small study about the efficacy of the masks at retaining coronavirus, especially since the study was performed on a different strain, not the SARS-CoV-2 strain that causes COVID-19.
Another study cited in the letter was a not-yet-peer-reviewed preprint by Santarpia et al. published March 26 in medRxiv. It involved sampling the air inside isolation rooms of COVID-19 patients at the University of Nebraska Medical Center and in the hallways outside them. Air samples taken in the rooms of two patients (one more than 6 feet from a patient’s bed) were positive for viral RNA, as were personal air samples taken from personnel performing the sampling in both rooms, and two-thirds of samples taken from the hallways outside the rooms. It was reported that neither patient actually coughed while sampling personnel were in their rooms.
The third recent study cited in Dr. Fineberg’s letter was another preprint, not-yet-peer-reviewed, in this case by Liu et al. published March 10 in bioRxiv. The authors tested 35 aerosol samples from patients, medical staff and public areas in two Wuhan hospitals (one an established hospital, the other a temporary field hospital) for the presence of the novel coronavirus. They also tested several other public areas in Wuhan not associated with these two hospitals. The samples were taken in mid- to late February and early March. Of the 11 public areas tested, 8 tested negative while two of the remaining three were hospital-associated (the pharmacy of the field hospital and the outpatient hall of the established hospital) and had relatively low results; the highest of the three was taken from the crowded entrance to a department store and may be due, at least partly, to the presence of asymptomatic carriers in the crowd. In the other hospital areas, the most consistently positive results came from the medical staff areas, which the authors suggested may have resulted from resuspension of particles from personal protective equipment during its removal, the movement of staff, or the resuspension of floor or surface dust. A single sample from a patient bathroom in the field hospital was also significantly positive (which may indicate the creation of aerosol when a toilet is flushed). With respect to the public areas, the authors concluded:
The results showed overall low risks in the public venues but do reinforce the importance of avoiding crowded gatherings and implementing early identification and diagnosis of asymptomatic carriers for early quarantine or treatment. Personal protection equipment such as wearing masks in public places or while in transit may reduce aerosol exposure and transmission.
Overall, CDC’s new recommendation that members of the general public wear face masks in public when appropriate social distancing is challenging (either to protect themselves from others who might be asymptomatic carriers or vice versa) makes a lot of sense, but the scientific evidence to support this position is presently still rather limited. Although proper face protection remains generally unavailable, any barrier, including homemade face masks, that can impede the flow of potentially infectious exhaled droplets and aerosols in confined spaces should be seriously considered.