 The airborne transmission is in fact dominant. There are many, many definitive recorded cases of airborne transmission. And in fact, there are many cases of COVID spreading that can only be explained by airborne transmission. Furthermore, respiratory diseases such as COVID-19 often are spread by an airborne route and this has been established in the past. If you believe, as I do, that airborne transmission is dominant, then you really should be worrying about the air, which means you should be worrying about primarily the exchange of the air. So ventilation is the number one factor. Again, comparing the outside being ideal, but inside still being a situation where you can control ventilation quite significantly. And also looking at other factors such as air filtration, wearing masks is critical. This is Startup to Storefront. Today's guest is Dr. Martin Besant, professor of chemical engineering at MIT and who has recently made headlines for his app which helps calculate safe exposure times and occupancy levels for indoor spaces as it relates to the airborne transmission of COVID-19. Dr. Besant has researched transport phenomena and fluid dynamics as well as the physics of viral spread. So he was uniquely suited to lend his expertise in the fight against the spread of the coronavirus when it became a global threat earlier this year. His research offers a clearer look into airborne spread of the virus and asks that we start thinking more critically about our previous notions of defense against it. So listen in as we cover everything from the origin of the six foot rule, actionable steps you can take to prevent contracting COVID-19 and whether or not you should wear a mask. Actually, I'm gonna answer that one right here. The answer is you absolutely should. Mass work, the scientific consensus is pretty clear on that. The vaccine is coming. Till then, be safe and wear a mask. Now, back to the episode. Welcome to the podcast. Everyone on today's show, we have Professor Besant. Thanks so much for joining. Please introduce yourself to all of our listeners. Hello, I'm Professor Martin Besant from MIT, professor of chemical engineering and mathematics. I wanted to have you on the podcast for so many reasons. I mean, you're doing a bunch of work around modeling around COVID. Can you just give us a quick introduction to some of the work that you're doing, specifically as it pertains to some of the COVID modeling that you're introducing? Sure. My generalized expertise in research and teaching is in the field of transport phenomena in chemical engineering, which refers to the transport of mass or energy or heat through fluid flow and diffusion and processes such as that. And I got into this topic of COVID, specifically last spring when I and my colleague, John Bush as well, felt that the six foot rule and other guidelines being promoted to protect against the spread of COVID-19 were not properly taking into account the physics of how any disease, frankly, is actually spread. And in particular, a respiratory disease such as COVID, there is a potential for airborne transmission through exhaled droplets, which are then inhaled by somebody else. And in fact, there was quite strong evidence from the early days of COVID-19 that that was an important mode of transmission. And that essentially is a transport problem. So, you know, the exhalation of breath, the formation of aerosol droplets, their evaporation and settling to a stable size, the way they're then swept around the room in the ventilation, potentially filtered out, and then ultimately breathed in by other people who can then become infected is really a problem in transport phenomena. So I set out to basically, you know, apply my expertise in modeling those kind of phenomena to COVID-19 specifically. And in particular, not to come up with the most complicated computational model, you know, that of course could be done and in fact would be typical in some of my other research, but rather to come up with something simple that could actually be as simple as possible, but still capture the essential physics of transmission and thereby augment or potentially replace or improve upon the six foot rule or other distancing measures or occupancy caps and things like that that we see even today actually as the primary recommendations to fight the spread of COVID-19. And when it comes to starting this model, right? So typically in modeling, you'd at least have some sort of anchor data around the Spanish flu. I'm sure people didn't have Excel. And so there's nothing to rely on. And so in some way you're kind of trailblazing, but are you looking at things like sneezing, coughing? How does the models begin to take shape? What are some of the things that we need to make or call fixed? And then what are some of the things that we're adding to it in order to get it a little bit more accurate? Yes, so actually first of all, let me just, you know, clarify in all science and especially here, we are standing on the shoulders of giants. So I would definitely not say that, you know, what I've done here is, you know, radically different than all previous work. And in fact, it's very much building on a number of threads of research. I could not have done this work without a lot of important research and knowledge that came first. So you mentioned sneezing and coughing. So there's a long history there going back to least the 1930s in terms of, you know, visualizing sneezes and coughs and understanding the fluid mechanics. My co-author John Bush of the paper and also co-author of the app who's a professor in the math department, I wrote a fairly definitive paper in 2014 doing experiments and also theory on the fluid mechanics of coughing and sneezing. Now, my thinking in approaching this problem though in the spring was different actually. So we all know that asymptomatic transmission is potentially very important for COVID. That's the reason all of us are wearing masks and protecting ourselves in various ways even when we're not sick. And frankly, can any of you remember the last time you saw somebody coughing or sneezing in public? I actually don't think I've seen it once. In the entire pandemic. So the idea that we're somehow protecting ourselves from coughs and sneezes now that may be very important in a hospital ward where there's lots of COVID patients who maybe are not able to handle respirators, you know, and you might be very worried about, you know, their coughs and sneezes and how that transmits disease and certainly it can. But actually that's clearly not the dominant mode of transmission of COVID-19. On the other hand, that thinking based on coughing and sneezing still dominates the public health guidance. In fact, the six foot rule to the extent that it has scientific justification is based on that. So if you look at a cough or sneeze, six feet is roughly where some of the larger droplets are falling out by sedimentation. Although that is a somewhat arbitrary number, as you may know, the World Health Organization recommends one meter, which is closer to three feet. And the country of Netherlands recommends 1.5 meter. So there's not really a consensus on what that distance should be. In fact, it's only the United States actually that says six feet. And in fact, it's that sort of adherence to a number like that and sort of conveying the idea to the public that that's such a critical number that really got me started on this because I was aware that there was sort of limited scientific justification for that specific number and not others. And the strict implementation of that number in the United States led, for example, immediately almost to the closing of our schools and businesses. Because if you have to be six feet apart and no less, essentially all spaces become unusable because no spaces have been designed that way. So if you take the entire classroom area of the United States, you can't accommodate our students at exactly six feet apart. They just don't have it. So basically kids can't go to school and businesses can't be open. On the other hand, three feet is just fine. So the World Health Organization, if you were to follow that guideline, we would have most of our schools open right now, which is the case in most of the world. So I was already aware of that, potentially enormous impact of this rule back in the spring. And so that really, as I saw that, by the way, affect me, myself personally through my kids and my family and my friends and various coworkers and industry but also just seeing the impact of these public health guidelines and also recognizing their somewhat limited scientific basis, it really called me to action and say, I've got to do something about this. I have to study this problem and try to figure out what would be perhaps a more rational guideline for COVID-19 specifically. And then as you're going through the process, so once you begin going down that, which is great because there's so much ambiguity or at least in my head, looking at this from a semi-scientific mind, there's way too much unknown going through this time. You have, and now states are doing whatever they want. So we have Florida, that's operating a little bit different than we're in California. And so you have now like this human frustration where there's no, nothing's linear anymore. Everyone's kind of like doing their own thing, very tangential. So you make the decision to do this or at least take the hypothesis of, does six feet matter, right? I assume that's kind of like a jumping off point. What happens next as it relates to building out the model? Are you using like a, like is it an algorithm that you're using? No, it's just old fashioned pencil and paper. I could have done this model in 1920. So I could not have put it into an app or a spreadsheet, but I could have written down the equation. And in fact, that's exactly what's the motivation here because the reason the six foot rule or for example, an occupancy cap like 10 people or 25 people, the reason those rules are implemented so widely is that they're very easy to convey. Everyone understands what it means. We can go take out a tape measure, start measuring six feet and then we feel we're safe and we know what to do. If I say that to be safe, you have to perform a complicated computer simulation of some model that you'll never understand and that is very specific to a particular space and not somehow as general, that's just not gonna have any impact. And in fact, might be too much. I know from my own work in science, it's extremely important to make models capture the essential physics, but not too much more. You can easily over model something, include too many effects that are very specific to certain assumptions and sort of miss the key physics basically. And so what I did set out to do was to try to understand what are the key physical parameters and derive the simplest possible model that can capture those. Now specifically the starting point though that allows for a simple model to happen is to recognize that especially when people are wearing masks but also when simply they're asymptomatic we're not transferring the disease primarily through coughs and sneezes. It's more normal breathing and in fact through tiny aerosol droplets so that refers to particles which don't settle very quickly that end up in the air just from normal respiratory activities. They're released when you're breathing normally they're released in larger quantities when you're speaking and increases with the volume of your speech singing it turns out as the activity that leads to the largest emissions of aerosol droplets. But in case there are these infectious aerosol droplets that do contain virions which are the capsid form of the virus and they are not able to be seen by the eye. So when you cough or sneeze you can collect on some kind of surface or you can see with your eyes with the right lighting. You can see all these droplets that are at the scale of hundreds of microns or millimeters. So we're talking about sort of things you can see with your eyes but there are many smaller droplets. So it turns out that the most common droplet size emitted by respiration is around one micron or smaller and those droplets you can't see with the naked eye. In fact, you can't even really see them with a normal microscope but they are floating around in the air and so essentially you have infectious air and so they do very slowly settle but they're in the air and so what's more important is to understand the mass balance of the exhalation of an affected person producing those droplets and then those droplets being swept away in the air currents and reaching a mass balance of ventilation which is sort of adding fresh air and removing some of that air possibly filtration if some of that air is going through a filter and droplets are being removed. Also deactivation of the virus which just can happen naturally by a variety of processes within the droplets and ultimately then the inhaling by other people and when you take a model like that and you think about airborne transmission the simplest model is that of a well-mixed room. So we all know that the air is a high Reynolds number flow typically. What that means is that a little bit of movement causes a very complicated flow. Okay, so this isn't like honey or a very viscous liquid where when you move something through that fluid it kind of just has a sort of a very simple flow profile around it. Instead the air has very complex flows which we're very familiar with in cases when we can visualize it. Think of for example smoking or the kind of hot air rising above a candle very complicated flows which do tend to lead to rapid mixing. So this has been well studied in the field of building engineering and indoor air quality and a reasonable first approximation in that field is to assume a well-mixed room. It's also an approximation that has been used in the field of a epidemiology to study respiratory diseases. So I'm not the first person to do that. In fact that goes back to the classic work of Wells and Riley and is called the general approach is called the Wells-Riley model. So essentially I've started with that kind of classical starting point. Is there any difference? So as you know all of a sudden all these places are building these outdoor tents which I don't know if that makes any sense but I was an aspen recently and the indoor is now the outdoor. Like they're literally building additions on the sidewalk and somehow we're getting away with this notion that we're safer. Is there any truth to, if you have no, if you're outside as an example are you radically safer than if you're inside and there's a good or let's call it a normal ventilation system? So yes I believe that it's dramatically safer to be outside and that the greatest risk is in fact indoors and airborne transmission and there's really mounting evidence for that in terms of specific spreading events that have been studied and also statistical analysis of large data sets of how spreading has occurred. But yeah that's my opinion also from just a scientific point of view thinking about how transfer can happen. And we are speaking here of these aerosol transmission think of it like infected air, okay. So I'm not speaking here of coughs or ballistic projectiles or droplets. So if I'm outside and someone coughs in my face like they're still coughing in my face it doesn't matter if I'm inside or outside really. But what does matter is if they're also releasing lots of tiny aerosol droplets those get swept away in the air currents. And outside obviously if it's windy it's extremely safe, everything's blown away. But on top of that even if you're just in a fairly well ventilated especially outdoor space there are just natural air currents that occur because people are moving also just due to thermal buoyancy effects. So the air you breathe out is warmer typically than the air outside if it's not a very hot day. And that leads to the tendency for that air to rise. So in fact, if you visualize the airflow around a person they're almost like a chimney. In fact, even when they're not breathing just the warmth of your body will cause currents of air to be rising around you. And in fact, this is one of the benefits of wearing masks is not only that the mask can filter these droplets but also that it blocks the sort of momentum of breathing itself. So when I speak or I breathe or certainly when I cough I put a lot of momentum in the air I create a plume of air that goes forward which can then lead to a higher concentration of air that can go directly into someone else's face and they breathe it in. But when you're wearing a mask that momentum is blocked and so you're more creating just kind of a cloud of some extra droplets and particles around yourself and those tend to rise and tend to then be swept away by the background air currents. And when you're outside and even in a tent actually that effect can be much stronger than it is indoors. On the other hand when you're indoors with very high ventilation it's also okay. I mean it's important to remember there are laboratories that deal with virus including the famous one in Wuhan China which perhaps didn't do it as well as everybody else but you do understand that when there's a very high air change rate so very high circulation you can deal with very toxic chemical fumes or virus laden droplets in the air and you can be safe actually because the air is so quickly changed around you. The issue arises in normal indoor spaces do we have enough ventilation? Do we have other conditions that make it actually safe or not? And that's where the kind of calculation that I've done and the app actually allow you to make that assessment but generally outside you're gonna be substantially safer. So take us through your model. So I know that one of the parameters is that it presupposes I believe that one person in the room is positive for coronavirus. Is that correct? Well not quite actually so unfortunately a number of news stories have run with that including ones that didn't interview me but actually what it is controlling so here so what I set out to try to do was to drive something very general that would always be a very conservative starting point and so to do that I drew from a concept in a pedi-mology which is how we control diseases in the first place from spreading. So you may have heard in the news discussion of R0 the reproductive number so that's basically for every infected person how many other people do they infect? Okay what's the expected number of new infections per infected person? When that number is bigger than one then obviously the number of infected people is growing so you actually get an exponential increase that's in the early stage of epidemic it's sort of exponential increases until eventually you run out of infected people and then it starts to come back down again, okay? But the key number that people watch for the early stages and it also is how they characterize how infectious a disease really is is by the R0 number, okay? So it's called the reproductive number. So for example in China and Wuhan in January and February early estimates averaged around 3.5 for that number so every one person who had it was infecting around 3.5 others and we all saw how dramatically the infection initially spread and then similar numbers like that have been recorded with some variation in different cities, different countries. So the important thing is again this reproductive number if you want just a blanket statement like this is not contributing to spreading so that was one of the first choices here in terms of deciding if I'm gonna get a formula or a guideline what am I actually gonna try to measure actually try to control? So I said well the simplest thing and the most general thing is to basically give a tolerance which is sort of a risk of a transmission for the indoor reproductive number. It does not mean that the model only applies when one person is in the room who's infected which people have written that unfortunately a lot of news stories it's saying it's a conditional probability if someone enters the room then the expected number of people they infect will be less than some tolerance which you choose to be less than one. So let's say it's 10%, 1%, whatever your risk tolerance is going to be if two people enter the room this is the probability that two people will become infected. So this is not saying that it's only applying to one person it's just saying that for every person that enters they're not gonna infect another person with a probability one it's gonna be a lower probability that so the most common outcome would be if someone enters the room in this amount of time with this occupancy and other factors they won't infect somebody else. So that's the most general thing you can do. Now that's what the app is calculating right now. We are working on trying to keep the simplicity of the app because right now the beauty of it is that there's no math at all. So when you first come to it it just says like you have some dropdowns for kind of what type of space you have how are you gonna use it and says this many people for this many hours can be in this space. But what it's saying is that for every infected person that comes in if you follow these parameters then it's unlikely that that person would infect somebody else. And unless a really large number comes in like if it's half the room is infected that does change the calculation. We did assume we were in the limit of small number of infectors but that's almost always the case because even when we're at sort of a critical level of the pandemic locally in terms of prevalence that's a few percent of the population we start to get very concerned. It's not 80% of the pop you're not gonna have a room where 80% come in sick unless it's some special situation that we're not looking for. On the other hand that conditional probability needs to multiply by the probability that an infected person actually enters the room if you're worried about just controlling the risk of transmission overall. So if you ask yourself like is it safe for me to go to a Christmas party which people are talking about right now using our app or is it safe for me to go to the Walmart? You're not concerned necessarily with the situation assuming there's an infected person or several infected people how many transmission will they be you wanna know is there a transmission and plus times the probability that there is an infected person in the first place. So that has to do with prevalence. So if let's say your gathering has 10 people and the prevalence of infection is 10% which is incredibly high that's one infected but actually it's probably more like 1% in the population locally which means that you could repeat this experiment 10 times and still not see a transmission. In other words that room is actually a lot safer than you think. So the constraints that we give are actually very conservative and if I'm gonna make a single statement like six foot rule this is the statement I wanna make. If you just follow this we're gonna beat this disease like there won't be transmission but that doesn't mean that a space can't open up because after all as the pandemic is subsiding and eventually there's hardly anybody left who's infected why would you keep imposing this rule? At some point you have to decide I'm not gonna impose this rule anymore and that has to do with the second calculation I mentioned what is the total risk of a transmission which includes the probability that there even is an infected person in the first place. So that's a separate calculation but we didn't put it in again for two reasons one to be conservative in the first iteration of the theory in the app and secondly also because people may not know that other parameter I don't want people making sort of medical or public health decisions based on parameters they may not even know. I mean there are tools out there for example Georgia Tech has a very nice online tool for finding the current sort of prevalence of infection in your county in the United States and also in various parts of the world. So there are tools to get that information but I don't think it's fair to assume that every user knows in fact I don't think I know exactly what the prevalence of infection is in my county so I wouldn't know exactly how to enter that into the app although potentially like a policymaker could who's trying to decide like when to open a school or something. Yeah so it's kind of a subtle point and we're trying to in the terms of the app strike a balance between being usable to a broad set of people but also allowing it to be the result to be understood in different ways. Gotcha and it seems like the other thing the other aspect that I was curious about was you talked a lot about how it's spread through air droplets and so I work in the film industry and now we have an entire department that is called the COVID compliance department and their job on set is essentially to enforce a six foot separation policy, masks, face shields but they also go around and they clean highly used surfaces and it seems like a easy thing to do it makes sense everyone's touching it, clean it sanitize it, whatever. But it seems like it should be more of a focus on getting air scrubbers into filming stages and worry more about the ventilation as opposed to high traffic surface areas. You know, do you know are you familiar with the efficacy of cleaning such areas versus getting a properly ventilated area? So this is still an area of scientific debate although I would say that most scientists have come around to the opinion that the dominant mode of transmission is airborne. Public health agencies such as the World Health Organization and the CDC have been very slow to accept that conclusion or to endorse that officially and they've maintained from the beginning even before the pandemic that the standard strategy for dealing with disease is wash your hands, disinfect services and stay apart from sick people. So essentially that advice hasn't really changed except there's that specific six foot version of it in the United States which is not followed elsewhere but nevertheless that has been generally the public health guidance. Now I think personally and I think many, many scientists agree that the airborne transmission is in fact dominant. So the kind of transmission you're talking about in a pedi-mology is called fomite transmission. That's where there's an infectious residue let's say from breathing or coughing which ends up on a surface. You touch that surface and then you touch some part of your body that basically gets it into your body. You touch your eyes, you touch your mouth, you touch your nose and then you transmit. So first of all, the direct evidence for fomite transmission is not that strong and to my knowledge is not a single definitive recorded case of fomite transmission. There are many, many definitive recorded cases of airborne transmission and in fact there are many cases of COVID spreading that can only be explained by airborne transmission. Furthermore, respiratory diseases such as COVID-19 often are spread by an airborne route and this has been established in the past. So for example tuberculosis was treated for decades in exactly the same way. Wash surfaces, you know, think, worry about coughs and finally it was shown that that's the only way of transmitting essentially is airborne because the pathogen, which in that case is a bacteria would be contained in those aerosol droplets I mentioned that needs to get deep into your lungs into the smallest pathways, the alveoli and they are so small that only the smallest droplets can actually get there. So you might ask yourself, you know, let's just say there's an infectious residue sitting on a fork or on a surface and you touch it, how is it exactly gonna get into your lungs? You know, because the virus is not really present in your blood or other bodily fluids aside from the respiratory tract. So even if you were to touch your eye or sometimes you might ask, you know, again, it's the physics of transmission, physically, not biologically, but physically, how is that virus gonna get into your lungs or into your respiratory tract? So it's my opinion that the fomite transmission is not a significant mode of transmission for COVID-19 and in any case, you know, it can't hurt to be overprotective. So, you know, if you're gonna be cleaning surfaces, I don't wanna tell you not to do that, but you're absolutely right that if you believe as I do that airborne transmission is dominant, then you really should be worrying about the air, which means you should be worrying about primarily the exchange of the air. So ventilation is the number one factor, again, comparing the outside being the ideal but inside still being, you know, a situation where you can control ventilation quite significantly and also looking at other factors such as air filtration, wearing masks is critical and, you know, some other mitigation strategies are available as well. But I think that's really the important thing that we kinda need to recognize is the importance of airborne transmission. I'm not the only person saying this actually by far. In fact, there are much greater experts than I am who spent their lives working on, you know, this area of public health who've been saying this. So for example, Lydia Morofska at Queensland University of Technology, Donald Milton here in the United States and Lindsay Maher, many others have been saying that airborne transmission is dominant for COVID-19 since the very beginning. And in fact, over the summer, they got together 239 concerned scientists to sign an open letter essentially to the World Health Organization stating, you know, this is the dominant mode of transmission. The CDC did not accept or even mention airborne transmission until September when they posted some initial guidance which they then took down and then it came back in October and they acknowledged that airborne transmission can occur but did not acknowledge that it could be a dominant mode of transmission but they did at least acknowledge it can occur. World Health also around the time of that letter in the summer did acknowledge afterwards that airborne transmission could occur. So that's where we are right now is that airborne transmission officially according to the public health agencies is a possibility. And so whether you think it's dominant or not, I think it's dominant. So do many, many other scientists. It's important to have a way to protect against it. So the agencies have not yet provided any specific quantitative guys to protect against it which is really what I've been trying to provide again to go beyond six feet as it were because you know, again, if you're talking about airborne transmission the six feet, any distancing measure is really the not the right way to go about protecting yourself. You know, a nice analogy I could give that I think kind of summarizes all of this is smoking. We all know the effects of smoking. If you think of the particles that arise when someone is smoking a cigarette as these infectious aerosols are breathing out, we all know that there is the risk of direct transmission. I could exhale my cigarette directly in your face. In fact, we're also aware of that that most smokers don't do that. They tend to look away and blow somewhere slightly away from someone else's face. You know, so we're aware that that is kind of giving an increased transmission of smoke particles. It's pretty obvious, right? On the other hand, we also know that when you're in the room with even one smoker is that pretty quickly, very quickly, you can smell it and even see it everywhere in the room. And it doesn't matter if there's a face shield. I imagine someone's smoking the face shield. The smoke's just gonna go right around it. Like it's totally useless. It'll block the momentum of that sort of puff, you know, which is useful. But other than that though, the particles get out. They're in the air. And you know, the smoke particles, just like the aerosol droplets, they don't settle. They're in the air. And we also know that, you know, if there were, let's say at the supermarket, there's this kind of barrier, oftentimes a plexiglass between you and the other person. Again, if I'm gonna cough directly on the cashier, that is probably a very good protection, just like a face shield would be. But again, if we're asymptomatic, nobody's coughing, we're just breathing, and we're all wearing masks on top of that, it's not possible to have that sort of projectile of respiration going towards somebody else. Instead, there's this kind of leakage of particles and droplets much like smoke. And what we should be worrying about essentially is like the secondhand smoke. So if you wanna think about it, this is, you know, the theory that I've put forward and the guidelines is sent here a way to protect yourself from a secondhand smoke in the room. In fact, the equations for smoking would be exactly the same. I would just have to change, you know, kind of the number of particles per breath or something like that. But the basic principles of ventilation, filtration, everything else would apply just as much in that case. Makes a lot of sense. I wanted to ask you a question that's all encompassing in some way. So you're wearing so many hats presently or let's just call out your scientific hat and you're running an analysis that for the most part doesn't lead into politics and it's just really straightforward in the sense of a narrow focus. As you're going through this and you've begun to do a bunch of interviews, what has been surprising to you? So you kind of alluded to the fact that some of the press out there is, let's call it just running with certain things and not entirely correct. Are you concerned or worried about how this might be politicized? Cause to some extent someone could run with the headline of, oh, this six feet rule doesn't make any sense. And now there's an MIT professor going on record, creating a model to do this. Oh, I've been, I can't tell you how stressed I've been about this exact topic. I knew that going into it. So I mean, I had no illusions. And that's really, I think one of the barriers for a scientist to decide to take on a problem like this is that it's so much larger than the science. So I'm used to just publishing a paper and generally nobody cares except for like my narrow little community. And maybe years later, we see the importance or not of that result. But this is something where, I didn't really set out here to just try to sign a paper. I really wanted to try to help the world and do it in real time also, which is why I'm doing these interviews before my paper is even published actually and why I posted a preprint, why I put out an app because I'd like to impact, how the world responds to the pandemic while it's still an issue, not have it come out three years later. So I was definitely well aware going into this of how essentially everything is politicized, especially in the United States, but potentially anywhere in the world. And I've been very conscious of the kinds of headlines that could come out or misinterpretations of work. And that's why at every stage of the work, my co-author John and I have been really thinking about being conservative. So if I were to make an error or be misquoted, I would rather it be encouraging others to be more careful than to be less careful and create a risky situation. So we've been very conservative in a number of choices. But in terms of just the outreach then, that's another reason that I'm doing interviews and I'm happy to talk to the press is I want to try to control the narrative and also learn from that. So one thing that surprised me just actually last week, there was a lot of interest in the app. It's kind of gone in waves. The New York Times actually first covered the app back in September and it generated some interest, but not that much for some reason. Just last week, it went crazy actually. In fact, our server crash we're getting, we have had a million users this week actually. So it's been crazy actually. So we've had to actually get another server for the app. Even my personal webpage crashed actually. I couldn't load it actually for a period there was so much interest. So definitely this topic now is like, I guess with all these big second waves happening everywhere and new rounds of shutdowns and possible reopenings, there's tremendous interest in how to protect oneself and also how to make decisions about whether or not to attend a function or for policy makers, whether or not to allow different gathering sizes and all that. So there's tremendous interest right now. So yes, I've just been trying to stay ahead of it. One thing that happened with this recent publicity just to give you an example. And again, I don't mean to criticize any of the, any reporters I think everyone is trying to do their best to kind of get the story right. And when people maybe don't get it quite right, it also has good feedback for me in terms of the messaging and also maybe even the calculation that's presented in our app or even in the paper. So when I was first thinking about this, I was motivated by the problem of schools to be honest, even MIT as a university, like, we also shut down basically because of the six foot rule. And so that was actually my initial motivation. Should we really be doing this or should we be doing something else? Like maybe some spaces are actually safe, others are not safe. We've also sent all our students home, maybe they could actually be here, or maybe they shouldn't be here. So that that was the question that I wanted to address. And so I would say that was kind of the initial thinking of sort of a little bit larger spaces, more people in it, and really understanding the role of ventilation, everything else. The story that came out recently, given the fact that we're on the holidays, Thanksgiving in the US and Christmas coming up as well, people are very concerned about, can I attend my holiday party? And so people were running the app for really small numbers of people, like six, and it might say with no mask, of course you're eating without mask, you could be there only for a short time. And so essentially then it says, we can't have Thanksgiving dinner, we can't have Christmas with our relatives, okay? And I'm necessarily saying you can, you should be aware of the issues that are raised by the app. But you also be kind of careful in how you use it. It comes back to this issue I mentioned earlier, that I was thinking initially more along the lines of like should the New York City public schools open? For example, okay, that's a big question. And there you're really worried about are these people in these spaces, the students, the teachers, contributing to the spread of the pandemic? It's kind of a community spread decision made at a higher level and it's kind of a blanket statement, are we going to be open or are we not? And I think that the information I provide, I think can be very helpful actually in making such decisions. And also I was thinking of cases like nursing homes, where there could be a short but very high risk transmission. And I want to understand when that can be an issue because they're like, it's so sensitive, just it doesn't happen very often but when someone comes like it, this can be deadly actually, so it's with much higher probability. So I was very concerned about those kinds of cases. But when you apply it to the case of say a Thanksgiving dinner, let's say the app gives you for your space, we have very poor ventilation and you're going to have 10 people total, let's say and your nuclear family is maybe five and you're going to have a few relatives over. It might tell you you've got three hours and you might say, oh, maybe I can't really have that dinner. And I would say, perhaps that's the case, we have to look carefully at the situation. On the other hand, that's not necessarily the right way to use the app actually because as I mentioned earlier, we have for a specific incident, what we really care about is the total probability of a transmission, including the possibility that an infected person is there in the first place. So in some parts of the country, the prevalence of infection is very low actually. So like the chance that there is an infected person is extremely small. So what you need to do is multiply the conditional probability of transmission that we're essentially using with the probability that there is an effective person. So you take your gathering size, multiply by the prevalence of the infection which potentially you can look up and find in some database. And then you finally have a lot more time because actually the chances are that there won't be an effective person. Even in fairly stricken areas, like I said, maybe there's a 3% infection rate in the population, it's not 90%. It's still like the chance that an individual person comes in is actually COVID positive and also importantly doesn't know it because of course if they're positive and they've been tested or they have symptoms, they're not gonna come to your party. So these are people coming who think they're healthy. The fraction of those people who actually have the disease and don't know it, it's significant but it's not 100%, it's not even 50%, it's not even 20%, it's probably like a couple percent at most and maybe lower than that. So that's one factor. The second factor, again in terms of using the app in a way that is more kind of tailored for a certain situation, is this idea that we live in a household where we have a certain set of people, let's say like my family is, I have four kids, I have a family of six, a household or more generally people live in pods or other kind of more familiar arrangements, that's a group of people who know and accept the risk of transmission. So another criticism that happened when our app ended up on some stories and websites is if you read the comments, a lot of people are really angry saying, well, this is BS because this professor is saying we're gonna get sick over dinner but we have dinner every night and we're not sick. What's he saying? And we've been living together for three months or no one's gotten sick and now he's saying we're gonna have a dinner with like one extra person and suddenly we're gonna be sick in three hours, no. So in the specific case of a holiday party, what you really should be thinking is there's a certain number of people that are potentially bringing you the infection and it's not everybody because you're, let's say, relatively comfortable with your immediate family, either you're tested or you're spending a lot of time quarantined together or you just simply accept the risk of transmission between you and you're really worried about the outsider. So you're inviting your grandparents or your cousins. So there's some people coming from the outside, they're gonna be joining you for a period of time. So what you really wanna take primarily is that number of people, times that probably they're infected to estimate how many infectors there would actually be or what's the chance there would be an infected person. That's much lower than the probability that's essentially in the app right now which is 100% chance, right? So it's a totally different question. So if you wanna know like, can I actually have a Christmas dinner with a few visitors, chances are you probably can and that requires using the app in a slightly more subtle way. Unfortunately, when those news stories started running, I became the Grinch that's still Christmas. You know, among other things. And so that's, again, not my intention. Now I'm happy to be the Grinch if the real, if the news you need to get is a tough medicine saying, hey, don't have this gathering. Like I might be saying that, but I think many times I'm not saying that. And so it boils down to understanding what the app is calculating, what the theory is telling you and how to use that information. So we're working on improving. So it's actually, this was a good thing, by the way, that just happened this last week. We are right now working on improving the app with just including that effect or prevalence of infection into your decision-making. But we also don't want to be right on the front where you have to enter a bunch of numbers that kind of takes with the beauty of it. So we're trying to find ways to kind of keep the user experience simple, but to convey these kind of two different ways of thinking. One is just sort of from the perspective of like a high level policy maker, you know, let's say in charge of the New York City public schools, you know, wanting to like control the spread of infection in general versus, you know, an individual who's deciding, you know, should I go to a party or should I, you know, go to a store, you know, for a certain period of time? Those are two very different questions. It seems like on an individual basis, your app has been a massive runaway success in terms of getting out to people and getting in the public side. When it comes to high level policy makers or the CDC, the WHO, like the powers that be, who can actually affect change on a massive scale, has anyone reached out to you to infer about your research and your algorithm and your modeling? So this is a very good question. So, and I should also preface my remarks by saying that, again, other scientists, I mentioned some of their names earlier, are also actively advocating for acceptance of airport transmission and development of new guidelines similar to the one I'm presenting with the World Health Organization and with CDC. So, you know, I'm definitely not alone in that regard. Certainly in the early stages of this research, I did reach out to both organizations and have had some conversations and it's still optimistic that eventually changes will happen. I guess, you know, I've come to understand the perspective of those organizations, which is that they are essentially have the huge responsibility and weight of making a recommendation for public health. And so they naturally are very concerned organizations. So they don't wanna jump on something that, you know, might be a little bit less considerate, they're not sure about. And so the safe thing is like keep it as it is until we're absolutely sure that we can go in a new direction. And when we go in the new direction, like for example, they have acknowledged airborne transmission both organizations now, but they haven't told you exactly what to do. They have mentioned that for airborne transmission ventilation is important. Some of these other factors, they just haven't given a quantitative guideline, which is what I've tried to do. But therein lies the challenge that those organizations face is if they're gonna give a quantitative guideline like I've given, you know, I can put a disclaimer that's, which I do on my app, which basically says like I'm not responsible for transmissions that occur. I'm just trying to help the public. This is based on many assumptions. I try to explain what those assumptions are. You know, when the CDC or well teleorganization says something, they're saying, you know, we are the experts, this is what you should do. And so then of course, they also have to take responsibility, you know, if they've gotten, you know, if maybe they've gotten that wrong. And so they're basically very conservative organizations. On the other hand, I think another issue that's at play here is that this problem of transmission is really a problem of physics or engineering. It's really sort of the physical process by which a droplet goes from me to you convicted through the air and all that. It's not really traditional medicine or public health. So, you know, what happens when those droplets are inhaled and the virions enter your respiratory tract is a question for biology and, you know, epidemiology. This question of transmission in terms of the physics of it and what's the role of, you know, the ventilation system, for example, is really much more of an engineering question. So I think that's been another challenge here is it's a truly interdisciplinary topic and it requires, you know, voices from different areas of expertise. So I'm just trying to add, you know, my voice to the mix here, being very open about the things I don't know that well. I'm not a medical doctor, you know, I'm not commenting at all actually on the sort of the progression of disease, things like that. But in terms of, you know, developing a model of transmission, parametrizing that from spreading events, that's absolutely the domain of chemical engineering or related fields. So I think that's another issue. And in terms of what I've been doing, in terms of trying to make that change, it's too much to expect that, you know, the world of telecommunication would listen to somebody like me saying, here's what you should do. They're gonna say, well, who are you? You're not expert in public health for the last 40 years, you know, and all that kind of thing, even though I do understand, you know, transport phenomena very well. And also I think some of these ideas, frankly, are new to some people. Definitely they're not new overall. There are communities that do the kind of modeling that I've been doing and are making, in fact, similar recommendations. But overall, you could say broadly in the field of public health and policy, you know, this way of thinking is not generally champion. And so another thing I've tried to do is outreach. So besides having these kinds of interviews, which is, you know, kind of a sort of a high level, just talking about the basic ideas, I've tried some other, you know, more maybe creative and modern approaches. So I happen to be in my teaching focusing on open learning and massive open online courses. So in fact, I have what I believe is the first graduate level, like really high level Mathematical Engineering class online, actually, which is, I have actually two classes now on transport phenomena, which are graduate classes that I teach exactly on this topic of fluid mechanics and diffusion, all that. Equations all over the place. It's kind of like Khan Academy style. I'm working at the LIPOR, but you know, it's like massively more equations and all that, you know, so that's a medium that I've been working on now for, you know, three or four years. And so I decided, you know, for this topic, what I need to do is to create a massive online class. Actually, it was my producer and former student who now works with me on these MOOCs, Joey Gu, who actually made that suggestion. He said, why don't you just like create a MOOC? Cause I told, the funny thing is I told him, you know what, I've got to get the word out there. It's like, people are just not understanding the science, okay? And my paper is one way to understand the science, but it's aimed at other scientists, very technical. So I thought, maybe I need to write a book. And then Joey told me, he said, no, that's very old fashioned. He said, no one's gonna read your book and it's gonna be published next year. And so he said, you should do a MOOC or you should do Reddit and do and ask me anything. You should do podcasts, you know, and so like I'm starting to kind of wake up to like sort of the modern ways of reaching people and realizing that those ways are much, much more effective than how I would have gone about it as an academic. Again, you know, writing a book, you know, which would come out next year with MIT Press or something, you know, like after the pandemic's over. And so actually what I set out to do just two months ago, actually I had this conversation with Joey in early September. He said, well, let's make the MOOC. So like while I'm actually teaching and doing a million other things, somehow in two months we managed to throw together a MOOC, actually a massive open online course. It just launched a couple of days ago. We have now, I think maybe 600 registered learners and that number is growing. And essentially the MOOC is aimed, you know, not quite at the general public, although the general public could enjoy it and learn from it. It is free, it's self-paced, so it's there, it's on edX. So in fact, I encourage your listeners to take a look. It's called 10S95X, Physics of COVID-19 Transmission. And it's aimed at the general public in a way. It has some interview lectures, some simple lectures that explain the basic concepts, but then it does go into the technical details, mainly aiming for an undergraduate level of background in science or engineering or related fields, including for people that are perhaps, you know, working in industry or in public health or in fact, education or other fields, maybe they had a technical background of some sort years ago. And, you know, so you have to be able to understand some of the equations, okay? But I try to at least explain the concepts in relatively simple terms. And there's a lot of homework problems that help you apply the concepts, behind the theory, behind the app, to practical situations. In fact, you mentioned tense earlier, that's the homework in chapter two, I think, is basically analyzing outdoor tense. Yeah, so if you wanna learn about outdoor tense, sign up for the MOOC and take that. Yeah, in fact, the final exam question is should a school reopen or not? Okay, so it comes back to very practical decisions, but I really feel that in order to understand how to make the decisions, you essentially have to take my class. If you're not, you know, if you're already a scientist and you can read my technical paper, that's great. And you may know exactly what to do. But the vast majority of people don't know that. And it also need to be educated on this whole, on some of the concepts that we've been talking about, these aerosol droplets and indoor air flows. And so the course kinda walks you through that and involves some of these practical examples. So that's one of the more creative ways that I've gone about, you know, trying to get the word out. And I feel that, you know, after there's been an MIT Open Online course, all this other outreach apps, it's kinda hard after a while for the public health agencies to completely ignore this work, you know, cause there's just a larger and larger people globally that are interested. So the app, based on the interest in the app, we had requests globally to have some translations. I started out from certain countries. And, you know, fortunately working in academics, I have even in my own group people from, you know, many different countries. And we, so I easily found a bunch of volunteers with my colleague John Bush also to do translations. So the app is now translated in about 20 languages. And in many of those countries, you know, this is kinda the only game in town. Like they, you know, unlike in the US where we have the CDC is also our national organization for health, in other countries, there's the World Health Organization, but that's not necessarily, you know, directly affiliated with what their country is doing. So they feel a little more freedom. And in fact, if you look at how other countries are dealing with COVID-19, they're not just following WHO or certainly not CDC, they're doing their own thing. And so they're very open and very interested in using this information. For example, in Germany, actually, they're completely on board with airborne transmission. Their public health recommendations now include things such as, or recommendations such as, if you spend 20 minutes in a classroom or office, you need to open the window for five minutes at least. Even if it's winter, you have to get that fresh air in there. And I guess, you know, in Germany, a lot of buildings actually do have nice big windows that can easily be open, but it's just they're directly telling the public, you have to think about the quality of the air and don't be locked in a small space for a long time. We're not getting that message right now clearly enough anyway in the United States. We're just getting, you know, disinfect the surfaces, wash off your fork or your pen, stand six feet apart and put a plexi-class barrier that's about this wide between you and a supermarket cashier. And we're just not thinking about the air. Yeah, you know, it's been really frustrating for me personally as going through this. We do real estate development here. And so there's, and I think this is the issue with the government and the public sentiment now where there's a lot of pressure, there's a lot of vectors coming at the problem at the CDC that relate to, are we choosing the economy, are we not? And there's just this like this growing outrage that continues to happen. And so thank you for doing something to sort of alleviate or bring light to whether it's six feet or just how particles travel through the air. I wanted to just wrap on this, what are you doing? So if you were in charge, let's say, would you keep the six feet? Would you make it a meter? Would you just let, would you focus all of your attention on like in the Germany situation that you just mentioned on ventilation? What would just be like the quick things that you yourself would move forward on if you all of a sudden became, let's call it King or just were in charge of a small community? What are some of the things that you would say this is, these are the measures that I've seen that would be effective? Yes, well, so the first thing I would say is that in situation we're worried about the transmission, we should be wearing masks, first of all, it really does help. So I think there are no doubt are valid arguments about the filtering capability of masks for these aerosol droplets. Now there are some misleading aspects, they're in the sense that the virus is really small. It's about 120 nanometers. And you, but I don't think you find the virus just freely floating in air. It's always in droplets, which are significantly bigger than that. And the typical size for a dried up sort of mucus droplet that's submitted by respiration is more like a micron, which is like a 10 times larger than a virus. At that scale is right about the tipping point where most filtration media are going from being, you know, very effective at larger sizes to being much less effective, smaller size. So certainly masks are not perfect filters, but they help. One reason is that the factor, the effect of the mask comes in squared in the sort of safety formula, because there's a mask protecting, you know, sort of blocking the transmission of the infected person, but also you're wearing the mask at the source, you know, or at the target, basically where someone's breathing it in. Compared to air filtration, which just sort of filters there somewhere farther away from the people, it's impossible to completely filter all the air. So I can transmit a droplet from me to you and still have very effective filtration in the room, because you can't filter everything. In fact, you can't recirculate the air forever, because there has to be outdoor air coming in so you can breathe. We need oxygen, we need to remove carbon dioxide. So there's always air coming through the room, and so it's much more effective to block at the source or sink than to, you know, kind of filter it. So masks are important. Also, they block that short range transmission I mentioned. But then once you get people wearing masks, then I think you find, you know, using the theory or the app that in fact, many places we currently shut down are in fact safe and that there's nothing particularly important about six feet, I would make that three feet, actually. And just generally, I call it natural social distancing. People don't like to be in someone else's face. Unless you're in a nightclub, you know, or maybe you're at a bar where you wanna be really close to other people, I don't know, but generally, like in a classroom, people are not in each other's faces. They tend, you know, if you and I were in the same room and we didn't have to do this through Zoom, I would probably be three feet apart from you, okay? And I probably wouldn't be like breathing in your face over and over, I'd probably be turning my head. And so I think especially when you're wearing masks, this idea of strict distancing, I think can really go away. I don't think it's that valuable. I don't think it's adding that much, but it's causing tremendous issues, as you well know, and we've been discussing, you know, closures of schools and businesses, oftentimes they're simply predicated on not having enough area, basically. You can't have the people in there because of six foot rule. And I don't think there's very strong basis for that, especially if people are wearing masks. But then instead you say, well, what is effective? Ventilation is the number one thing. Air filtration is second, but not nearly as effective as increasing the ventilation rate. And that could mean as simply as opening windows, turning off fans, but if you have mechanical ventilation, you know, use it. So yeah, so those are some of the recommendations, you know, that I would make. I love it. Well, look, thank you for coming on the podcast and sharing your expertise with us. Super informative to me. We'll try to get this episode out immediately so people can take a listen to it. You mentioned a couple of things and I just like to recoup around the class that you just tell everyone again about the class that's out there. I think it's 10S95X and then where they can find your app. Yeah, so first of everything is linked from my website. If you just search by name, you'll find that I have a COVID page which is kind of an outreach page that links everything that I'm doing, including the course. You can also find the course on edX, edX.org, which is the free repository of open online courses that was started by MIT and Harvard. So the course is on there. If you just, in fact, if you type COVID, it's one of the first courses you'll find, but it's called Physics of COVID-19 Transmission from MITX. That would be, I would say the main outreach. I'd love for your listeners to take a look at. Again, if there's no pressure to actually do the course or turn in homework, you don't have to be graded. You can just look at it and then walk away, but you may find something interesting there after listening to this podcast. Otherwise, also my paper actually, which gives you the kind of deep technical dive into the material is also a link from my website and can be found on Medical Archive and I hope will soon be published, but it's still in peer review. So if I were waiting for peer review, none of this work would have come out yet. We wouldn't be having this conversation. So I'm also grateful that there are pre-print servers that can help get the word out, such as Medical Archive. I love it. Well, thanks so much for coming on the podcast. I appreciate it. Yeah, thank you, Martin. Yeah, thanks. Bye-bye.