By Lambert Strether of Corrente
I was considering using “All that is fomites melts into air,” but I couldn’t bring myself to, so count yourself lucky (and anyhow, it’s not really true). From the beginning of the #COVID19 pandemic, we’ve been washing our hands, masking up, cleaning surfaces, and social distancing. These measures have worked (especially masking), but now we know more. There’s mounting evidence that airborne transmission indoors is a key — perhaps the main — pathway to SARS-COV-2 transmission. In this post I want to look at why that’s so, give examples, and suggest a simple heuristic to stay safe. Material like this might also be used to inform public policy (here; here) by reducing superspreader events in enclosed spaces like churches (airborne transmission via singing), restaurants (loud talking, especially if room is noisy), bars (ditto), nursing homes (shouting), gyms (grunting), meat-packing plants (shouting), call centers (talking), offices generally (air conditioning), and other hot spots, but working that polucy out is not the object of this post (see here for engineering controls for airborne transmission, and here for covid-proofing public spaces).
Airborne Transmission of SARS-CoV-2 Indoors
This article from PNAS seems to be the index publication for airborne transmission. From “The airborne lifetime of small speech droplets and their potential importance in SARS-CoV-2 transmission“:
Speech droplets generated by asymptomatic carriers of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) are increasingly considered to be a likely mode of disease transmission. Highly sensitive laser light scattering observations have revealed that loud speech can emit thousands of oral fluid droplets per second. In a closed, stagnant air environment, they disappear from the window of view with time constants in the range of 8 to 14 min, which corresponds to droplet nuclei of ca. 4 μm diameter, or 12- to 21-μm droplets prior to dehydration. These observations confirm that there is a substantial probability that normal speaking causes airborne virus transmission in confined environments.
That experiment was done inside a box. Vox translates to real world terms:
A crowded indoor place, then, with poor ventilation, filled with people talking, shouting, or singing for hours on end will be the riskiest scenario. A sparsely populated indoor space with open windows is less risky (but not completely safe). Running quickly past another jogger outside is on the other end of the spectrum; minimal risk.
(In other words, the problem is not density or proximity; the problem is transmission of the virus, through the air, by human vocalization (of which coughing and sneezing are a small, and symptomatic, subset.)
That would explain why masks have worked. (One could argue that masks need only be worn indoors, but most people are constantly moving from the outdoors to the indoors and out again, which would involved touching the mask constantly to remove and replace it; better to wear it all the time. In any case, minimal risk, to others, is not no risk). From the South China Morning Post, “Coronavirus: hamster research shows effectiveness of masks ‘huge’ in Covid-19 battle, Hong Kong scientists say”:
Hong Kong scientists conducting research on hamsters have offered the first proof of what many residents have believed all along – that wearing surgical masks can significantly reduce the rate of airborne Covid-19 transmission.
The study, which the team called the first of its kind, found the rate of non-contact transmission – in which the virus was transmitted via respiratory droplets or airborne particles – dropped by as much as 75 per cent when masks were present.
(See also “If 80% of Americans Wore Masks, COVID-19 Infections Would Plummet, New Study Says,” from Vanity Fair.)
Examples of Airborne Transmission of SARS-CoV-2 Indoors
So far as I can tell, there are two main villains: Air conditioning, and vocalization:
Air conditioning. Here is the very first article I spotted on airborne transmission, back on March 9, 2020. From the South China Morning Post, “Coronavirus can travel twice as far as official ‘safe distance’ and stay in air for 30 minutes, Chinese study finds”
A passenger, known as “A”, boarded a fully booked long-distance coach and settled down on the second row from the back.
The passenger already felt sick at that point but it was before China had declared the coronavirus outbreak a national crisis, so “A” did not wear a mask, nor did most of the other passengers or the driver on the 48-seat bus.”It can be confirmed that in a closed environment with air-conditioning, the transmission distance of the new coronavirus will exceed the commonly recognised safe distance,” the researchers wrote in a paper published in peer-review journal Practical Preventive Medicine last Friday. They said the study proves the importance of washing hands and wearing face masks in public places because the virus can linger in the air attached to fine droplet particles.
Here is a seating chart of the bus:
After reading this, I altered my practice not to mask up, which I was already doing, but to avoid (air-conditioned) public transportation entirely, and indeed air-conditioned spaces entirely.
Here is a second example, a Chinese restaurant. From the CDC’s Emerging Infectious Diseases, “COVID-19 Outbreak Associated with Air Conditioning in Restaurant, Guangzhou, China, 2020“:
We conclude that in this outbreak, droplet transmission was prompted by air-conditioned ventilation. The key factor for infection was the direction of the airflow. Of note, patient B3 was afebrile and 1% of the patients in this outbreak were asymptomatic, providing a potential source of outbreaks among the public (7,8). To prevent spread of COVID-19 in restaurants, we recommend strengthening temperature-monitoring surveillance, increasing the distance between tables, and improving ventilation.
Here again is a seating chart:
I was already not going to restaurants because of the Chinese bus episode, if I had been, I would have stopped. One can’t wear a mask while eating!
Vocalization. We have several examples of vocalization (singing, shouting, talking, grunting, etc.) causing transmission.
On March 29, we had an event at the Skagit Valley Chorale’s rehearsal in Mount Vernon, WA. Here is the report, again from the CDC: “High SARS-CoV-2 Attack Rate Following Exposure at a Choir Practice — Skagit County, Washington, March 2020“:
Following a 2.5-hour choir practice attended by 61 persons, including a symptomatic index patient, 32 confirmed and 20 probable secondary COVID-19 cases occurred (attack rate = 53.3% to 86.7%); three patients were hospitalized, and two died. Transmission was likely facilitated by close proximity (within 6 feet) during practice and augmented by the act of singing.
(Sadly, privacy concerns forbid a seat diagram.) I don’t agree that proximity in and itself transmits anything; it seems clear to me that singing was the issue (although transmission through fomites was possible, as choir members put away chairs, etc.). Supporting evidence from Missouri’s Daily Journal, “A surprising way you may risk getting Covid-19“:
The possibility that singing might help transmit infectious diseases is not a new concept. A 1968 article, “Singing and the Dissemination of Tuberculosis,” described an elaborate box that volunteers could talk, sing and cough into, allowing investigators to measure the number, size and length of time airborne of individual infectious droplets they breathed out. And a few TB outbreaks have featured singing, including one in a New Jersey church choir in 1995.
(This is good local journalism, too.) Science confirms, in “Why do some COVID-19 patients infect many others, whereas most don’t spread the virus at all?“:
Some situations may be particularly risky. Meatpacking plants are likely vulnerable because many people work closely together in spaces where low temperature helps the virus survive. But it may also be relevant that they tend to be loud places, [Gwenan Knight of the London School of Hygiene & Tropical Medicine] says. The report about the choir in Washington made her realize that one thing links numerous clusters: They happened in places where people shout or sing. And although Zumba classes have been connected to outbreaks, Pilates classes, which are not as intense, have not, Knight notes. “Maybe slow, gentle breathing is not a risk factor, but heavy, deep, or rapid breathing and shouting is.”
We also have the following case in Chicago. From the CDC, “Community Transmission of SARS-CoV-2 at Two Family Gatherings — Chicago, Illinois, February–March 2020“:
This report describes the cluster of 16 cases of confirmed or probable COVID-19, including three deaths, likely resulting from transmission of SARS-CoV-2 at two family gatherings (a funeral and a birthday party)
Here, instead of a seating diagram, we have a timeline:
I’m guessing “Happy Birthday” was sung at the birthday party, hence the greater number of cases originating from it.
Here is the case of a South Korean call center. From the CDC, “Coronavirus Disease Outbreak in Call Center, South Korea”:
We described the epidemiologic characteristics of a COVID-19 outbreak centered in a call center in South Korea. We identified 97 confirmed COVID-19 case-patients in building X, indicating an attack rate of 8.5%. However, if we restrict our results the 11th floor, the attack rate was as high as 43.5%. This outbreak shows alarmingly that severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) can be exceptionally contagious in crowded office settings such as a call center. The magnitude of the outbreak illustrates how a high-density work environment can become a high-risk site for the spread of COVID-19 and potentially a source of further transmission. Nearly all the case-patients were on one side of the building on 11th floor. Severe acute respiratory syndrome coronavirus, the predecessor of SARS-CoV-2, exhibited multiple superspreading events in 2002 and 2003, in which a few persons infected others, resulting in many secondary cases. Despite considerable interaction between workers on different floors of building X in the elevators and lobby, spread of COVID-19 was limited almost exclusively to the 11th floor, which indicates that the duration of interaction (or contact) was likely the main facilitator for further spreading of SARS-CoV-2.
Here we do have a seating diagram:
It seems unlikely to me that air conditioning was the major factor, because otherwise — HVAC mavens in the readership will correct me — the cases would have been distributed throughout the floor. However, what call center personnel do is talk, a lot. Hence I would urge that vocalization is the driver, not mere proximity.
And finally, we have the case of a South Korean gym. From the CDC, “Cluster of Coronavirus Disease Associated with Fitness Dance Classes, South Korea”
Characteristics that might have led to transmission from the instructors in Cheonan include large class sizes, small spaces, and intensity of the workouts. The moist, warm atmosphere in a sports facility coupled with turbulent air flow generated by intense physical exercise can cause more dense transmission of isolated droplets (6,7). Classes from which secondary COVID-19 cases were identified included 5–22 students in a room ≈60 m2 during 50 minutes of intense exercise. We did not identify cases among classes with <5 participants in the same space. Of note, instructor C taught Pilates and yoga for classes of 7–8 students in the same facility at the same time as instructor B (Figure; Appendix Table 2), but none of her students tested positive for the virus. We hypothesize that the lower intensity of Pilates and yoga did not cause the same transmission effects as those of the more intense fitness dance classes.
Here is a really neat map of the cases (full size version):
We see that “high intensity” classes accounting for all the transmission; there was no transmission from Yoga and Pilates classes. Here the vocalization would be heavy breathing, huffing and puffing.
Heuristic of Avoid Transmission of SARS-CoV-2 Indoors
One salutary result of focusing on indoor transmission of SARS-CoV-2 is that we don’t have to get all het up  about photos like this anymore:
The foreground/background compressed photos of people enjoying outdoor spaces are becoming an irresistible genre. This one from the Washington Post shows a few hundred people in the space of a few hundred meters. In other words, reasonable numbers. The moral panic will backfire. pic.twitter.com/MBBNNMVcXP
— Pinboard (@Pinboard) May 25, 2020
Photographer’s tricks aside, these people are outdoors; the risk is minimal (though I still won’t answer for, say, small groups of people sitting on a beach blanket, sharing beers and singing old songs). I would avoid groups like this, but then I would anyway, virus or no virus.
The Japanese seem to have had some success by focusing on indoor transmission as well, following a rule called “The Three C’s.” Here is a poster:
Bloomberg explains, in “Did Japan Just Beat the Virus Without Lockdowns or Mass Testing?“:
Experts are also credited with creating an easy-to-understand message of avoiding what are called the “Three C’s” — closed spaces, crowded spaces and close-contact settings — rather than keeping away from others entirely.
“Social distancing may work, but it doesn’t really help to continue normal social life,” said Hokkaido University’s Suzuki. “The ‘Three C’s’ are a much more pragmatic approach and very effective, while having a similar effect.”
However, I think that knowing what we know now, we can add two more almost-C’s to avoid: Air onditioning, and “Voalization (or perhaps acophony?) However, all in all, I think the best heuristic is offered by one Ángela Caída’s Twitter account:
This “Three C’s environment is the same as a crowded, musty cave full of tightly packed, chattering bats, which makes sense, because that’s where the coronavirus evolved.
Caves are also cool, like air conditioned spaces, and while bats vocalize, I doubt they transmit SARS-CoV-2 as well as we do.
So, to avoid SARS-CoV-2 airborne transmission, don’t be like a bat! Avoid bat caves! Also, learn sign language?
 Thanks to the NC commentariat for a really useful discussion on shouting in nursing homes.
 There is a big debate over whether vocalization produces big and small droplets, or a continuum of droplets (with “aerosols” at the small end) but I don’t think that matters for the purposes pf this post.
 “Much of the cacophony in a bat cluster, the researchers suggest, is bats voicing their annoyance with those in very close quarters around them.” Bat vocalizations are “ultrasonic“, “repeated bursts of only a few hundred milliseconds.” So making the assumption that lower-pitched and longer human communication emits more virus, one might speculate that bats would be less vulnerable to airborne transmission of #COVID-19 than humans.
 Terry Pratchett, The Truth:
“Mrs. Tilly, I think you wrote a lovely well-spelled and grammatical letter to us suggesting that everyone under the age of eighteen should be flogged once a week to stop them being so noisy?”
“Once a day, Mr de Worde,” said Mrs. Tilly.