COVID-19 Aerosols and Airborne Transmission
This article is part of an ongoing collaboration between the Colorado School of Public Health, the Denver Museum of Nature & Science, and the Institute for Science & Policy. Watch the full recording of this session and find all of our previous COVID-19 webinars and recaps here.\
To view full recordings of the National Academies workshop referenced in this piece, visit the organization's session page.
When it comes to the COVID-19 pandemic, scientists and public health officials are still working toward consensus on a fundamental question: To what extent is transmission of the responsible virus (SARS-CoV-2) airborne and what is the size range of infectious particles? Research has shown that an infected person expels viral particles in droplets via sneezing or coughing and that masks, social distancing, and generally being outdoors can reduce the spread of the disease. People also produce small particles that may also be infectious. But much remains uncertain about how long those infectious droplets and smaller particles linger in the air and how far they might travel, particularly in enclosed settings. So what does the current science tell us about exposure indoors? And are there mitigation strategies that homes, schools, and businesses should think about adopting as winter approaches?
The Institute’s Senior Policy Advisor Kristan Uhlenbrock spoke with Shelly Miller, Professor of Mechanical Engineering at the University of Colorado Boulder, and Jonathan Samet, Professor of Epidemiology and Environmental and Occupational Health and Dean of the Colorado School of Public Health to discuss viral aerosol transmission, the emerging importance of adequate building ventilation, and takeaways from a recent interdisciplinary two-day workshop hosted by the National Academies of Sciences, Engineering, and Medicine.
This transcript has been lightly edited for clarity and flow.
KRISTAN UHLENBROCK: Good morning to you both. Jon, could you start by speaking about the science of viral particles?
DR: JONATHAN SAMET: This is a universal topic because of the implications, and I'll start with the historical example of tuberculosis. Tuberculosis control was once about not spitting into spittoons or on the sidewalk. In fact, that was the guidance of the original version the American Lung Association. In the mid-1950s, very convincing experiments showed that tuberculosis was transmitted through the air. That led to a paradigm shift and the move to control airborne transmission. That’s an example of a shift in control strategies because what we learned.
So when we think about SARS-CoV-2, there's a number of different ways that it could be transmitted. One is direct contact between people ─ touching the hand of someone who has touched their face, for example. Another could be indirect contact through objects and surfaces, or so-called fomite transmission. (I don't know why we don't just call it surface contact, but fomites is the word that’s used.) There’s the larger droplet spray, when somebody is coughing or sneezing next to you. And then, there’s the smaller particles which have been referred to as aerosols. Perhaps all are important, perhaps some are more important under certain circumstances, but clearly there are important implications for control.
I'm going to talk about a workshop held last week by the National Academies of Sciences, Engineering and Medicine. I chaired the planning committee for the workshop, which was carried out by a special Environmental Health Matters initiative. Some background: The National Academies was designed just for this kind of controversial question. It was chartered while Abraham Lincoln was President to give guidance science based guidance to the government. The original National Academy of Sciences now been augmented by the National Academy of Engineering and the National Academy of Medicine.
We looked at the deposition of virus particles, or so-called baryons, into the respiratory tract, the nose, the upper airway, and the steps along the way. People generate infectious particles during our regular activities. The larger droplets travel a short distance and have a short residence in the air. The aerosols travel longer distances, particularly in an indoor space, and may mix throughout the space. They can stay in the air for hours and over those hours, some may remain infectious. We generate plumes, and the louder we speak, the more particles we generate. Singing and wind instruments are risky. It's sort of a setup for why bars have possibly been a dangerous or risky locale: they often have music playing in the background and people have to talk loudly. They're taking their masks off to drink, and perhaps alcohol leads to a lack of inhibition and a failure to adhere to distancing.
One thing that came out of the workshop was a good discussion about harmonizing terminology. Aerosols are the smaller particles that travel beyond six feet, and this makes them really important in indoor environments and perhaps less so outdoors where the wind may be blowing. Indoors, they can remain airborne for hours. And as I said, the virus can still be viable, so this is really critical as we think about cutting edge buildings. Dr. Miller is going to talk about behavior and buildings in a moment.
Now just an important conclusion: We still don't really know enough about the amount of virus inhaled into the lung, what risk for disease looks like, or what the disease itself looks like if we inhale more of the virus. There’s some animal work, which gives us some insights, but for people we still lack information. It’s very hard to come by because, obviously, we can't directly challenge people with different amounts of virus to see what happens. Some experiments have done that in the past with more benign viruses, but certainly not for this one.
We need a package of control measures to deal with airborne transmission, not pieces. One thing that came out of the workshop was that much of what is in the air indoors in particular is resuspended dust, and that dust can contain virus. So that’s the importance of keeping surfaces clean, including floors. We wear masks to protect others and we also wear masks to protect ourselves with ventilation and filtration. We still need foot distancing, which helps us with these larger particles to droplets and spray transmission from coughing and sneezing. And then of course, hand hygiene as well. We need all of these modalities.
Bill Nazeroff from UC Berkeley has worked on air quality for a long time and I’ll summarize with his poetic phrasing: Outdoors is better than indoors, of course, because there's dilution. Short term exposure is better than long term exposure. Mask are better than unmasked. Socially distanced is better than too close. A sparse number of people is better than crowded. Quiet is better than loud, because we project more with as we speak louder. General breathing is better than vigorous breathing. And risk can be lowered ─ but not eliminated ─ by improved ventilation and air filtration. So with that, let me turn things to Dr. Miller.
SHELLY MILLER: Thanks for having me here, it's a real treat. I will present to you what I talked about at the National Academies workshop, which is just an incredible experience to be a part of. And I'm really happy to be able to inform the question here, which is about was the role of the built environment in determining exposure. First, I want to alert you to the fact that I have written these two articles that are useful for the community and the public to try to understand the issues at play. The first one was in April on the science of infectious aerosols. Last month, I wrote about how to use ventilation and air filtration to prevent the spread indoors.
There are three things I want to talk about today: ventilation, air cleaners and UV lamps (or air disinfection). Let's go ahead and start with ventilation. The big question on the table for decades has been what ventilation rate is needed when you have an infectious disease outbreak. A study coming out of the University of Maryland looked at acute respiratory infections that are often heavily transmitted during the winter months through dorms and college campuses. They studied in great detail the ventilation of two different dorms. One was a high ventilation dorm and one was a low ventilation dorms. Ventilation rates are measured in liters per second per person. In the high ventilation dorm, they measured six liters per second per person and the outcome was that only one out of 11 residents got an infection over the winter months. In the low ventilation dorm, it was only two liters per second per person, and 47 of 109 got infections. We've seen this kind of data from multiple locations for decades.
A colleague out of Denmark showed us that an outdoor air supply of less than 25 liters per second per person increases the risk of sickness and decreased productivity. So if we can bound our ventilation rate as needed in an indoor environment between five liters per second per person and 25 litersf per second per person, somewhere in there is a is the right ventilation rate for most spaces depending on the use.
Another study which I just find fascinating looked at a tuberculosis outbreak ─ which we know is transmitted by aerosol ─ in in a university setting. They looked at the ventilation rates of these buildings where the outbreak happened. The outbreak started on November 10 with the index case and the subsequent cases followed. The ventilation rate in these buildings was 1.7 liters per second per person ─ very low. So they did a bunch of engineering to increase the ventilation rate and they got it up to 24 liters per second per person, and after that, the outbreak stopped. We also see that carbon dioxide was a good indicator of ventilation. So, when you don't have enough ventilation, your CO2 accumulates indoors and average was 1200 parts per million and maxed out at 3000 ppm. Whereas when you increase the ventilation, you've got it down below 600 ppm.
I also want to explain air change rates, which has been talked about a lot in the media. The air change rate is a normalized value of ventilation and measures the outside air volume added to the space and you normalize it by the volume of that space. So let's take an example of a 500 foot classroom. The American Society for Heating, Refrigerating and Air Conditioning Engineers ─ quite a mouthful ─ recommends 6.7 liters per second per person. In a classroom, their typical occupancy is 35 students per hundred square meters. So if we put this ventilation rate into a classroom with 35 students, and a square area of our 500 foot, we provide 109 liters per second of ventilation of outside air into this classroom. The air change rate is determined by dividing by the volume of the space, and then you have to change the units of time, so we get 3.5 air changes per hour. The time for much of the room air to be exchanged with the outside air is 17 minutes, whereas the time for every bit of air in that room to be put outside and replaced with fresh air takes 51 minutes under that 3.5 air change per hour scenario. (Please note this air change rate varies a lot during the day, even from the morning to the night because of the temperature differences from the number of people inside what they're doing, and so on.)
One of the key reasons that we know aerosols are transmitting this virus is because of the outbreaks/ We've had numerous outbreak reports that can only be explained by aerosol and not by anything else. For example, in this Chinese restaurant in Guangzhou, where we have one pre-symptomatic person at Table A and led to infections at tables, A, B and C. Why did this happen? Because there was no outside air supply and the aerosol was not diluted in concentration. It just built up over time, and these strong air currents from the air conditioner blew the virus across all three of these tables.
An additional outbreak that I've studied extensively was the Skagit Valley (Washington) choral rehearsal outbreak, where we had half of the choir come to practice for two and a half hours and 87% of the attendees were infected and started having symptoms that very next day from COVID-19. In our study, we estimated the average amount of aerosol that it took to get people sick and then we want to explore how we could have prevented this by maybe adding more ventilation.
We measured the probability of infection versus the loss rate coefficient, which includes the ventilation rate, any service deposition, that does happen and then whether the virus was inactivating while it was being airborne. We estimate that their ventilation rate in this space was probably less than one air change. It would take a lot of ventilation, all the way up to 20 air changes per hour to even get this infection rate down to 20%.
We also looked at the duration of the event and we can see immediately that even with one air change per hour, if you don't spend too much time in that space where there's infectious aerosol, your probability of infection goes from 87% to just 10%. And so this is one of the reasons why I've been recommending quite often to not spend more than 30 minutes of time in a heavily occupied space, such as a classroom where there might be aerosols being emitted.
Let's talk about filtration. I want to alert you to some really great resources: you can read the EPA guide to air cleaners in the home, and the California Air Resource Board provides really good information as well on air cleaners and also recommends that you do not buy ozone generating products. The way a filter works is that it basically draws the air that has been contaminated through a fiber filter that kind of looks just like your furnace filter, but a lot more finely woven. And then the clean air is on the other side. The dip in the efficiency of the filter curve is always happens at 0.3 microns and then it goes up from there, meaning that it’s very good at removing very small and very large particles. We know that the SARS-CoV-2 virus is less than 0.10 microns. And so anything above 0.2 to 1 to 10 micron is very efficiently removed. The rating number is just how good the filter is and right now we're recommending for recirculating HVAC systems to use 13 or higher if they can, and the high efficiency particulate air filter which you've heard possibly a lot about is the HEPA filter. That is .997 percent efficient at removing particles at 0.3 microns, meaning it's very efficient.
How do you know what air cleaner to buy? Well, what you want to look at is the clean air delivery rate and the certification seal from the Association of Home Appliance Manufacturers. For example, this certification tells you that this particular air cleaner can work in 120 square foot room and will remove 80% of the dust in the room, 80% of the tobacco smoke, and 80% of the pollen, which is going to work for coronavirus.
I wanted to show you some data from my lab that shows these things work. Air cleaners work for a variety of different organisms just the same. So if we're trying to remove a mycobacteria which is what gives you TB, and that's about one micron, it works just the same as if you're trying to remove a fungi aerosol which could also cause allergies and asthma. And right in the middle of there would be the SARS-CoV-2 virus as well, so it can remove it just the same. In this comparison, air cleaner one has a higher clean air delivery rate than air cleaner two, and that means you can only use air cleaner one in a 30 cubic meter room, and you can only use air cleaner two in a 4 cubic meter room.
My last topic is germicidal ultraviolet UV, which I’ve been studying for over 20 years. It damages the DNA of microorganisms, and they can no longer replicate and cause infection. The wavelength at which that happens is 254 and 260 is where the infectivity is the highest, so if you can inactivate them, they will be really efficient. We really only recommend that they are used in places that can be maintained designed and operated effectively because what you do is you handle UV lights in the upper zone of the space. The lower zone where people are is not irradiated because it can cause skin and eye irritationwith overexposure.
So here are some environments that I recommend considering where you would use them in For example rooms in which you don't have enough outside air ventilation but you need to add additional air changes, or in crowded environments where unsuspected infectious persons may be present (e.g. hospital treatments and isolation wards). As you increase the effective UV irradiance from, say, one microwatts per centimeter squared all the way up to six, you can increase your air exchange rate that you're adding with this UV. We got 17 air changes per hour in our study. Our study data has been published in the CDC guideline for health care settings. We found that among different engineering control measures, UV was the optimal strategy combined with effective isolation of vaccination for containing influenza, measles, and chickenpox.
I want to conclude with some data that's coming out of my lab now on musicians. We're studying aerosol emissions because we think wind instruments as well as singing generates a lot of aerosol. So we have an oboe playing into our particle measuring system, which counts the number of particles that come right out of the bell. We can see that when they play a scale or just play in general, you get a lot of aerosol. But if we put a bell cover on to the bell, it really reduces the particle emission so that it's just almost the same as background or even reading. Singing is similar. But we did find more aerosol released with the oboe.
KU: Thank you, Shelly. Is there a rule of thumb that people can use when it comes to time spent indoors and the ventilation of buildings?
SM: As a member of the public, it's really hard to walk into a space and know anything about the ventilation. What I recommend is first, make sure there's not a lot of people in there, that everyone's wearing masks. But then also if the room feels stuffy and hot, then it's probably not having enough ventilation. If there's no windows or doors open, it's probably doesn't have enough ventilation. Other than that, there's really nothing else you can do unless you want to get a little high tech and take a carbon dioxide detector with you and make sure the carbon dioxide levels are below about 600 to 800 parts per million in unoccupied space. If you have to go there often as a worker, that might be a good tool for you to use. Or, you need to ask facility managers about the ventilation rate and ask ‘are you following ASHRAE guidelines?’ And if they are, then that that should be an adequate amount of ventilation, especially if the space is not occupied as much as it usually is.
JS: We of course don’t walk around with devices that know how much virus is in the air ─ if we did, we could go into contaminated space and then know to leave. But I think this is where you're aware of wearing masks because for those who are in the room, wearing masks reduces the loading of the room with viruses and provides protection. So I think the mask use remains quite relevant to public health protection.
KU: Let's move into people's homes. Someone asked about specifically their swamp cooler and if there is any data or evidence that airborne COVID could be brought in via that or via smoke particles from recent fires?
SM: We don't have any information about how that's going to particularly impact the flow of a virus in indoor environment, I think we want to go back to basic principles which is making sure you have enough outside air. A swamp cooler does need outside air to work. And so that's good compared to if your room is really closed up and you're just running one of those mini split air cleaners.
JS: The question is about the usual particles that are present in the air, probably an unresolved issue at this point. There's been discussion that perhaps viruses tack on to particles and become agglomerated as they come together. Perhaps there's a carrier role. The smoke particles that reach us from the wildfires up to several hundred miles away or thousands of miles away from California are going to be very small by the time they get here. We know they’re here because we smell them. And the nose is a great detector. A swamp cooler brings in outdoor air and will bring in whatever is in that air, including small smoke particles. But I don’t think there’s enough information to say whether those particles could somehow increase the risks of SARS-CoV-2
SM: If I could just add about the resuspension issue, particles eventually will settle on the surfaces. And there has been data that shows that when you sample the floor and different areas of places where there's lots of contamination, you find a lot of virus. And there are implications from resuspension or virus being shed when hospital workers take off their gowns and their PPE, for example. We need an understanding that if you're in a high contamination environment, you could resuspend those particles and be careful. I used to just have “COVID shoes” that I would only go out in. That’s just in case walking around, you picked up some, you just leave it outside of your house. I don't worry so much about that anymore.
KU: Should people consider buying additional air cleaning products?
SM: If we talk about homes versus classrooms, it's a different issue. I recommend people have air cleaners if they're sensitive, if they have asthma, or they have possible COPD, especially in summer with modifiers. It's a really great tool to keep your indoor environment clean from particles that can make you sick. Now, whether it's going to help you in your home, I would say you don't have a lot of potential for transmission unless you have a lot of people in your home that may be going in and out and going to work, and you're worried about them coming back and having been contaminated. So it just depends on what your concern is.
KU: We are moving into some cooler months where more people are going to want to try to congregate indoors more. Are there recommendations that businesses should be thinking about now?
SM: I would like to see business have the ability to share with the public what they have done to improve their outdoor air ventilation and air cleaning. So for example, having some inspection service that comes from the city that says, okay, you have three air changes per hour in this room and so you get a certification. And then the restaurant says, and I've also added a filter that gives me another three air changes, so that you as a patron will know what they've been really trying to do to keep you safe in their space. It's very difficult for patrons and for businesses who are maybe even just renting spaces to express how the air is going to be safe in their environment without a little bit of support around that.
JS: Fall is usually the time when respiratory infections surge. And that's because we go back and forth and of course children returned to school and they very often introduce viruses in the home. Colorado's epidemic right now is in a good place, fortunately, as we head into the fall. School reopening is going on a very, very good way across school systems. So I think we need to continue taking steps with all the measures that we have been.
Labor Day is a time when people will be together and there may be social gatherings. We saw a surge after the July 4th weekend and we hope that won't happen because it would come coincident with it getting cold, like this morning. This is a moment we want to make sure we're doing all that we can to keep the population Colorado as safe as possible.
KU: Jon, could you speak a little bit more to viral load exposure and what is our current understanding on that?
JS: I wish I could! I think one thing learned from the workshop and my own research is that we have animal models that we use to look at viral burden. And as such, they're not people, but they are helpful for fundamental questions about quanta. How much has to get into the lungs? It's just not clear. I think it will be useful to answer those questions, particularly as we move into winter and need to make sure our environments as low risk as possible.
KU: Shelly, I wonder if you could speak to the use of filters when it comes to not killing beneficial microorganisms?
SM: So, unfortunately, we don't actually know which microbes in indoor spaces are beneficial. We have been studying what the microbiome looks like in your home. The bacteria in your home come from who's in your home, like people and dogs. The fungi in your home come from where you live, from outside your home and what comes into your home. And same with businesses. Now, which ones are good for your health and what makes you feel better and what are good probiotics and all that remains to be discovered. We know what the pathogens are. And we know that, for example, in really different environments like healthcare settings, there is this tendency for a pathogen to over because of the high levels of disinfection contamination. So that's a completely different environment. We can't say for sure whether the UV is going to be bad because it kills the good stuff. Yet.
KU: I want to wrap up on whether this has given us some good lessons about the role of science as it evolves in real time and if anything has shifted in your understanding about the role of information?
SM: There are a couple of things that have been really wonderful. The international community has really come together with interdisciplinary scientists to try to help resolve and support the science and the communication of science. We've been really frustrated that the WHO and CDC haven't been out in the forefront of providing the kinds of information that the world needs to understand transmission and understand that there is aerosol transmission happening. And so I appreciate the international science community coming together.
What I'm also understanding, though, is that I need to only speak about what my expertise is. I'm not an epidemiologist, I'm an engineer. And I can tell you a lot about ventilation and filtration, but I'm going to let epidemiologist speak to their expertise, and then work together to really get the clear message out to the public.
JS: A few comments. One is, surprisingly, how little we knew about airborne transmission of viruses heading into this pandemic. Second, the research community has responded quickly in a very interdisciplinary way. You have engineers and aerosol scientists and others speaking to each other in cross-disciplinary fashion which is being driven by the urgency of this pandemic.
Third is just the explosion of papers ─ some peer reviewed literature and some of this preprint realm ─ where there's literally thousands on airborne transmission. There are a few really seminal papers that have come along. It was in this context that the idea of the workshop, under the umbrella of the National Academies, came together to present the science and distill down the findings.
Importantly, it is as much about what we don't know and then asking how we fill the gaps. This type of workshop could easily be repeated for other elements of the pandemic. Perhaps in four or five months we will need another workshop on airborne transmission. I see this kind of workshop as really critical for bringing together all the research done by this extraordinary global set of researchers.
Professor of Mechanical Engineering at the University of Colorado Boulder
Former dean of the Colorado School of Public Health and Professor of Epidemiology and Occupational and Environmental Health
Disclosure statement:
The Institute for Science & Policy is committed to publishing diverse perspectives in order to advance civil discourse and productive dialogue. Views expressed by contributors do not necessarily reflect those of the Institute, the Denver Museum of Nature & Science, or its affiliates.