Getting Back to Work—Shield Immunity vs Herd Immunity: A Conversation with Joshua Weitz
Steven Cherry Hi, this is Steven Cherry for TTI/Vanguard.
Perhaps there’s no more pressing question in the United States and around the world than, “When can we all get back to work?”
If a return to normal living is months, maybe a year, or even more, away, how do we maximize a return to work—work and play, because after all one person’s restaurant visit or vacation is another person’s employment—while minimizing the rates of infection and death?
My guest today has given this question a great deal of thought. Joshua Weitz is a Professor at the Georgia Institute of Technology, and Director of its Interdisciplinary Ph.D. in Quantitative Biosciences. The self-declared mission of the Weitz group is to understand “how viruses transform human health and the fate of our planet.” The group includes physicists, computational biologists, mathematicians, and bioinformaticians, and their work includes research and modeling of disease dynamics and epidemiology. So it should come as no surprise that he has some novel ideas for dealing with the novel coronavirus.
Joshua, welcome to the podcast.
Joshua Weitz Thanks for having me, Steven.
Steven Cherry Joshua, you and a number of colleagues at Georgia Tech and elsewhere, recently published a paper describing a model of what you call shield immunity. First, let's talk about what the paper calls the existing strategies of mitigation and suppression.
Joshua Weitz Sure. As is probably well known, the response to Covid-19, has led jurisdictions, whether at the state or national levels, to impose lockdowns of various kinds in the purpose of these lockdowns and shelter-in-place edicts is really to reduce transmission, to minimize the incidents in which an individual who susceptible to Covid-19—and I should note that almost all of us are susceptible except for those who've been unlucky enough to have been infected, lucky enough to have been recovered—interacts with someone who is infectious. And that can lead to a new transmission event. But those mitigation and suppression policies have the potential and have already had a tremendous positive impact with respect to reducing disease burden and prevalence. However, they come with another consequence, which is severe socio-economic impacts and disruptions to social order, not just work–life balance, but truly social order.
So that I think is the context in which many of us are thinking about how can we break through this perhaps false dichotomy between, on the one hand, lockdown, everything is stopping in order to stop the disease, on the other hand opening up, which would then put us back at risk for new infections. And that's really where we began to think about a different approach.
Steven Cherry Okay, so what is shield immunity?
Joshua Weitz Right. The concept of shield immunity is to leverage individuals who have been infected and recovered as a means to essentially substitute for potential interactions that could lead to transmission. It is based on an assumption that people who've been infected and recovered, are protected from re-infection, in other words, are immune. That may not be immunity that lasts multiple years, but maybe multiple months. There's a lot that we still have to know about the extent and duration of potential immunity. But the core idea is that in other coronaviruses, like SARS and MERS, those who have been infected and recovered are protected from re-infection over a two- or even three-year period. To the extent to which that prior lesson applies here, we believe there's an opportunity to identify such individuals, many of whom—in fact, most of whom have never been patients—they were infected, maybe they had mild or even maybe severe symptoms, but they weren't confirmed as having Covid-19. Now that they've recovered, if it's possible that they could be identified—we can talk about how one does that and so on. They can form the basis for, in some sense, substituting, shielding interactions between susceptible and infectious individuals and instead replacing those kinds of risky interactions with interactions between a recovered individual who is not going to either make others sick nor get sick themselves.
Steven Cherry Yeah, I think we should note, if I understand correctly, although it could very well be much longer, your model assumes a strong level of immunity for four months?
Joshua Weitz And that, I think, is actually a conservative assumption, vis-a-vis, SARS and MERS, but it's not with other seasonal coronaviruses that obviously have far less of an impact. In our study, we assumed, at first that the duration of immunity lasts a year or more, but then did evaluate what would happen if immunity were shorter-lived. And we found at least that if the process of using serological tests to check for antibodies and then use that antibody information as a means to identify those individuals who've recovered and have them in essence act as shields to protect not just them, but serve the collective good. I think that's quite important.
So in essence, mitigation and suppression, these lockdown policies are telling everyone to interact less. But a shield-immunity approach says that a subset of people who've recovered could actually interact more, not just go back to normal business but actually, interact more, in some sense, help us all, not just themselves, return to normal, but help all of us return to normal.
So these shield-people interpose themselves basically whenever possible between people who haven't yet had the virus, I'm sort of picturing that a restaurant chef, say, could go back to work if all of the waiters and waitresses have had the virus. A heart surgeon maybe could perform surgery if everybody else in the O.R. Had immunity. Is that the sort of picture I should have?
Joshua Weitz Well, there are different ways to think about it, and I think there are steps to get there. We should do these in steps in order to build towards shield immunity in practice, meaning there are many individuals who've been infected and not been identified via PCR. We're increasingly knowing that through serological assays that test for antibodies. And then we also need to know that those people sero-convert, which means they actually develop antibodies. We know now that the vast majority people do. And we have increasing evidence that these antibodies have the potential to be neutralizing and confer protection, but that hasn't been fully confirmed yet. We still don't understand the entirety of heterogeneous responses. We don't have evidence of re-infection. But it's not to say it could not happen. So we still need to know more.
But to answer your question, with respect to when we get to that phase, there are priority sectors like healthcare workers, emergency-response workers, those who work with at-risk individuals like in nursing homes. And I'd put those even up before food-service workers. So that's critical as well. So, for example, in a hospital setting, if you had a choice between an E.R. doctor who had antibodies being on the Covid-19 ward, and one who did not, knowing that is critical that allows them to operate with significantly more freedom of movement and less risk to them than otherwise. The same thing applies to a nursing home where you have high-risk individuals, there's a large burden of mortality in nursing homes and other elder care facilities. Thinking carefully about testing for antibodies among staff as well as residents is going to be critical to figuring out how does one change, in some sense substitute, within a class of workers—not replacing someone's job, but actually shifting responsibilities to prioritize interactions that may involve extended periods with other individuals of unknown status to those who have had been infected, recovered and have confirmed antibodies.
Steven Cherry So this all requires, (a) that people develop that strong level of immunity for at least a period of time and that we can with great accuracy test for that. And then finally, what level of testing would really be required for this to work. And I've seen estimates that go as high as 25 or even 35 million tests per day, which is wildly above what we're doing.
Joshua Weitz So we thought about this question because like all research, we're trying to build evidence moving at a pace that I think in some ways is unprecedented given the scale and consequences of not taking actions, but also making sure that the actions that people do take are effective in not causing harm. Modeling has a key role there, one of the modeling studies that we've done, you've alluded to one which is this paper on the principle shield immunity, which came out in Nature Medicine. But together with Ben Lopman's group at Emory University, we asked that exact question how much and how accurate tests do we need in order to turn this into practice? And what we found was that once we get to the point where tests are increasingly accurate, scaling up makes a lot of sense. And so the landscape here has changed. I apologize if I'm mixing two things together at once, but I need to combine both ideas to answer the question.
There are tests that are now increasingly specific, meaning that a positive result denotes the fact that you do in fact have antibodies over 99.5, if not 99.8 percent of the time. And at that level of specificity, we estimate that if one were to test essentially everyone in America—I know that's a big number—but once a month. So there are about 330 million Americans, divide by about 30, you got about 10 million tests per day nationwide. But that essentially put another way, it sounds like a very large number. It really means once a month, one does an antibody test. The consequence of that is you begin to recruit more and more people who have recovered and enable them to take on increased roles and also therefore protect everyone else who has not yet been infected. The purpose here is to avoid getting new transmission. In other words, to avoid—which topic we may get into today, of herd immunity—by essentially enabling shield immunity, by enabling those recovered individuals contingent upon the fact that they have protective levels of antibodies to dilute otherwise risky interactions between susceptible and infectious individuals.
Steven Cherry This really depends on you being, in effect, a pool of people who test positively for having the antibodies—for having had exposure and now have immunity. I mean, how would this work in, say, some county in New Mexico that is maybe only 1 percent of the population exposed to the virus yet versus New York, which statewide might be at 14 or 15 percent—New York City, it's close to double that, perhaps.
Joshua Weitz So you're right that the efficacy of this strategy is going to work best in areas in which there's the greatest burden. In some ways you can think about that is that's an opportunity to leverage the unfortunate spread in certain areas to actually halt that spread in a faster way. So you're right that in a county with very low prevalence and sometimes there are going to be counties with low apparent prevalence, that doesn't mean we shouldn't test using population-based serological assets to find out true prevalence. I think it's very important to distinguish between the number of lab-confirmed cases that might be reported on a state-level Department of Public Health website versus the true spread in a jurisdiction so that we are almost certainly significantly undercounting cases that could be 5, 10, perhaps even 20 times undercounting cases. It does mean, however, that in such places we might think about targeted types—in other words, examining types of jobs, categories, types of work within a job sector as a means to identify groups that we think for various reasons may be at greater risk for having been exposed and therefore moving forward could potentially serve a role in terms of reducing the risk for everyone else. But in practice, once the percentages get to be 5 to 10 percent, or above, the population-level impacts of that can be significant. And unfortunately, there are many places and where we've already passed that level or we're approaching it.
Steven Cherry So maybe we could talk a little bit about herd immunity since you brought it up and how this differs from that.
Joshua Weitz Many folks listening to this probably heard the term herd—and I should say, since I'm I just said "herd" twice: H-E-R-D as in a group—and the idea of herd immunity in epidemiological terms. Is that when a disease spreads to a certain level in the population, then an infectious individual will come into contact increasingly with people who have recovered and therefore, for example, if I were to interact with you, if I'm sick, you're already have been sick and have recovered. I can't pass it on to you and therefore you can't pass it onto some of your contacts. In some sense, by having been infected and recovered, you are protecting other people who may not be in my range, but maybe in your range. If that makes sense. The problem with herd immunity for a disease that has as severe an outcome as Covid-19 is that it takes something like 70 percent or so of us to be infected for this to kick in. And with an approximately 1 percent infection fatality rate, there is a severe consequence to having epidemics spread to that level. I should say there are some technical debates about whether herd immunity is actually 70 percent or slightly less because of some heterogeneity, but that's a bit in the weeds for the sake of now. The key point is that herd immunity only happens insofar as the disease spreads, you can think, almost unchecked through a population. If we think about New York City, then the severity of the outcomes there with fatalities and consequences that have happened up to a point where 15 to 20 percent of the population have been infected. We want to avoid the outcomes that could happen if we were to reach 60, 70 percent or above. And I should point out that even when the prevalence reaches about 70 percent, there can still be what's called overshoot. There can still be new cases, even though the case rate goes down.
So the point of shield immunity and other mitigation approaches are to avoid a situation in which the population is protected only because it has suffered, because people have been infected and recovered, but rather to drive down new transmissions from the outset to a point where other strategies can be used to control cases. Like in New Zealand, they're nowhere near herd immunity and have, I think recently zero new cases recently in a day. I don't know if that's up to date with respect today, but there are so few cases and that's not because people in New Zealand are not any different in some sense immunologically than us. They are also entirely susceptible to infection. But they've controlled the outbreak. So that is the destination we're trying to aim for here and not a case where we're protected only in so far as many have been sick, many have been hospitalized, and far too many have died.
So I mean, herd immunity, is something where, as you said, and we don't even know exactly what percentage would be large enough to give us a sort of group protection and it wouldn't be perfect anyway. So I guess we could talk about sort of like levels of herd immunity and the idea of shield immunity is to sort of simulate that by increasing the number of interactions, so that—just to make up some numbers—maybe at only 10 percent of the population having immunity, you would get the benefit of 40 or 50 percent. Is that fair to say?
That's right. The concept of herd immunity is that, let's say I have 10 contacts that are going to be potentially infectious on average as long as I'm infected. And there's another concept, which is related to these herd immunity estimates, called the basic reproduction number or R0, which is the average number of new cases caused by an infectious individual in an otherwise susceptible population. And that number changes because we're not entirely susceptible. But at the start, we think it's something about 3 on average for Covid-19, which means a sick person causes 3 people to be sick and so on. Now there's time over which that happens.
Once we think about my 10 contacts, if we get to the point where seven of them have already been infected, then it's possible that I don't even write interact with those three who are susceptible any more because seven are recovered. And so instead of me infecting three new people, out of the 10 who I interact with, I only get to infect essentially 30 percent of the three, which gets much closer to one new case. And you can see now instead of cases increasing. I have no effect. Just one person on average, approximately. And once that number gets above 70 percent, then I infect less than one person on average and cases start to decrease. But shield immunity picks a different approach. The idea is that let's say there were a few people in my network who had recovered, let's say just 2. But if I go out to have some, whether essential, service or interaction, and those 2 people take on a larger role, so I interact with them more, then my contacts are preferentially with those recovered people. And in some sense, from the virus's perspective, their wasted contacts, they're not facilitating spread. And that's exactly what we want to see happen. So in some sense, it's a way of augmenting interactions that we view as less risky as opposed to mitigation and suppression, which suppresses all interactions in order to reduce transmission. And so in that way, mechanistically, it enables a smaller subpopulation of people—depending on how accurately we can identify them, deploy them—to serve the collective good to in some sense shield in the way that herd immunity does, but without the same consequences.
A recent article in The New Yorker by Atul Gawande argues that a sort of cocktail of 4 tactics—hygiene (handwashing, wiping down surfaces, and so forth), distancing, screening, and masks—is a sufficient regime to get people out of their homes and moving again. And I guess that's pretty much a description of what New Zealand has done successfully, a bunch of other Asian countries seem to as well. Shield immunity would be an additional tactic. It doesn't seem essential in New Zealand and some of those other places. Do you think they would still benefit from it?
I think that the clear benefit is going to be in places in which there has been significant enough spread that we need a greater suite of tactics to try to drive cases down, and I view shield immunity as one of them. In some sense we should think about trying to reduce risk whenever possible and by breaking it up in some sense into wedges. Thinking about this number 3 [i.e., R0—ed] how can we reduce 3 new transmissions to 2 or to 1 or even to less than one? And the more that we bring to bear on this problem, the more information we have with respect to not only who's sick, but also is recovered, that's going to help. So in framing this, one of the major points of our articles is that it can be used in combination with other methods. So synergistically with things like social distancing and I lump them in social distancing, not just the fact that we are physically apart, but taking steps through hand-washing, or things like mask-wearing, to reduce transmission in the event that one is in close proximity to others. So, in a place like New Zealand where the number of recovered individuals is very low, this would not be a priority strategy. In other places, including the United States, especially in harder-hit areas, it should be considered as part of the strategy set. And to prepare for that is precisely why we need to ask, well, how would we implement it insofar as we'd begin to keep learning more about these steps? Right. That there is this ascertainment bias. Many more people have been infected than we thought. Second, the people seroconvert. Third, that we can measure antibodies—and everything we know about SARS and MERS—it seems to suggest we're on the path. We're still waiting for a bit more of that evidence to show what levels of seroconversion and therefore what levels of antibodies and what kinds of antibodies are going to confer protection from re-infection. But in those places that are at that 5 percent or above level—or in sectors in which that's the case—like health care workers, like food service professionals, like those who work with at-risk people. Then there's a chance to rethink the work environment. Again, to reduce the chances that an individual who has many contacts comes in contact with someone sick and then transmits on to many other individuals.
Steven Cherry So places like the United States don't seem to be doing very well at most, maybe all of those four tactics that we mentioned. How effective would shield immunity be or how much help would provide for a country that wasn't doing well on those, in those other ways?
Joshua Weitz That's a very challenging question and it's challenging because there's the risk that if we're not doing good at those other measures, why all of a sudden are we going to be doing good at this new approach? So first of all, I'd like to just back up and say that we need to be doing better on these other measures too. The efforts that I think it would take to have a social norm of acceptable, and if not, essentially required, even if not by law, but as a social norm, to have mask-wearing whatever indoors and in close contact—that's something that's an accessible point. That's something that could be valorized and valued. The testing levels, there's been a lot of critique, have gone up to some extent, and you mentioned screening. Well, there's a very big difference between testing as a form of just checking in to see where things stand and testing as a gateway to action-taking. I view, serological testing as well as PCR testing as a gateway to action-taking. The predominant idea is that one should use PCR tests, which check the question, are you infected by checking for viral shedding and not just using that say, OK, well, that person needs to be treated, which is important one needs to support and respond to that person, but as a means to initiate contact tracing, to figure out that the person who's sick needs to be isolated so they don't infect others. But the persons that might have interacted with those infected individuals recently need to be pulled that quickly. So that they can be quarantined and also tested because much of the transmission happens in what is called a pre-symptomatic, or early symptomatic phase. So if we just use this test as a way to have a dashboard or a statistic or a static view, we're not going to keep driving, intentionally, cases down. These tests are an opportunity to stop change-of-transmission. PCR-based tests in isolation, stop trains, chains of transmission that have begun, and then try to stop others from being. A serological base test is trying to use information about who's already infected, so the chain is now done but to stop other new chains from beginning. So again, if we start to use serological tests, they can be helpful to the extent to which we understand how far the disease has spread. They can be helpful in so far as it helps us to estimate things like infection fatality, because we can compare fatalities. Even that is difficult to do sometimes versus the extent of the disease. But the point of shield immunity just like in the case of test–trace–isolate measures is to try and use that information actively to stop new chains of transmission from beginning.
Steven Cherry Even with a lot of testing, we would still need some sort of governmental policy or regulation or some other direction toward shield immunity?
Joshua Weitz I think the answer is yes. We need to think about it carefully. And one of the ways I think we can think about it carefully is to think about messaging related to the duration over which a positive test for serology should imply a sustained change in behavior. In other words, enhanced interactions because one thinks one is protected from re-infection, it will be critical to view things through the lens of what I prefer to call immunity visas in so far as they should be viewed as temporary, conferring some benefits, but that need to be reviewed in order to be renewed. In the sense that at present it is literally unknowable. Even if we were to find that the individuals who've been infected so far, there's no evidence of re-infection—that would basically tell us that we know that you can't be reinfected over a timescale of approximately four months. It is unknowable if it's going to last one year. So we should not presume that somehow an antibody test now means carte blanche for one full year or even six months. So I think it will require being quite careful. We have to avoid things like bug-seeking in the sense that there could be perverse incentives in order to get privileges with respect to having recovered. I think that it should be apparent to most that this is a serious and severe disease that would strongly disincentivize any kind of intentional effort to get sick or to have the benefits because the negative consequences are tremendous. But one still has to think about that. I think with respect to implementation, as we take a step forward, it makes more sense to try to think within sectors that already understand many of these issues as a means to pilot and evaluate how this could work in practice and those that understand most about exposure, infection, infection-control in general, health care workers—and also those to some extent working with at-risk individuals. I think those sectors are the ones in which could benefit from such a strategy, deal with individuals who are infected, have other critical services to provide and thoughtful interfacing, let's say between that sector and government, is going to be essential to make sure we're on the same page with what does it mean when one gets a positive anybody test? How long should we think about that as being actionable? How do we review it? And in fact, by reviewing it, we're also going to learn more about the duration of antibody titer that could then benefit all of us collectively from the scientific side as well.
Steven Cherry It's funny that you mention the idea of people seeking out exposure to the virus, because it has occurred to me that there would be an incentive and maybe it's not such a bad decision for people who are young and have no co-morbidities. I am reminded of—and I don't know this firsthand by any means—but I'm told that there used to be chickenpox parties where somebody in a neighborhood would get chickenpox and people would bring their children to get exposed to the chickenpox and sort of get it over with. This would be a little bit like that, but with a higher risk attached to it wouldn't it?
Joshua Weitz And because it's higher risk attached to it would be nothing like that. We should not be having such Covid-19 parties. No one should have them. And there are many reasons for that. One of the reasons is simply that the overall infection–fatality rate here at a population scale is about 1 percent. It's probably a little lower—there's a lot of uncertainty around it—that is significantly higher than influenza. And as far as we can understand, there are many things we don't understand about consequences.
So for individuals who might feel that they are healthy and young enough to be outside of a risk group, they might not know all of their potential co-morbidities or they may have no co-morbidities, but simply be unlucky with respect to the fact that there are fatalities of young adults and adults of my age in the mid-40s and obviously older adults as well. And people who ... If you are infected, there's a chance that you interact with someone who has such risk. So in other words, you are then risking other people who may suffer grave consequences of being infected. And finally, I would only add that these are complex systems. There are things that we don't yet understand, like the clusters of Kawasaki-like illnesses among some children that have popped up in Italy and in New York that are just beginning to be understood and may be related to Covid-19 in ways that I'm not an expert in, but certainly seem worrisome enough to pay attention to, as the father of two kids.
And so I think once you start going in the direction of increasing cases, you increase complexity, increase unknowns. And not only are there more dangers and issues to the individual, but there are many more problems. I would much rather be in a situation like New Zealand where they took extreme measures from the outset—again, it's an island nation, smaller, they did it earlier; they also had better leadership, let's be frank, with respect to the kind of strategies that they've used. That is a better destination to aim towards. So whatever level we're at now, a strategy that tries to drive down cases to zero reduces the scale of the problem from a human perspective, lives saved, also, the healthcare system has a capacity to deal with it. And also in terms of the contact tracing and control measures, meaning our public health interventions can scale. If cases start to increase and increase all the things that we can use and all the methods, become overwhelmed, not just the hospital beds, but our ability to track and trace and measure and do all the things we need to do to drive down cases.
I don't want to be ... actually, I did want to be that inside emphatic because I think it needs to be said that I understand that there is a potentially perverse incentive ... I want to reinforce the fact that our strategy here is to leverage the fact that many have been unlucky enough to be infected, but lucky enough to have recovered. Many of those individuals are not patients. They never were patients. They didn't know or maybe suspected that they were infected. But now that that's happened. How do we leverage that, to make sure that other people don't get infected in the first place, given the risk and severity of Covid-19.
Even if it were a rational decision to seek out exposure—and I'm convinced by your arguments that it's not—but even if it were, if too many people made that rational decision individually, you could overwhelm hospitals and other medical systems. So it would be some sort of horrible tragedy of the commons.
Steven Cherry I'm curious about one thing ... the level of testing that you argued for involved testing people once a month. Because it relies on people testing positively, there's no risk to doing it more or less often. But if testing were more often, you would know sooner that somebody was immune and was able to go out there and be exposed to other people safely. I hate the the war metaphors that surround this virus and the epidemiology of it, but it would be like being able to train soldiers faster to get them to the frontlines. Wouldn't it?
Joshua Weitz So with respect to serological testing, if you test negative today, there's no point in testing you tomorrow. There's a period in which people seroconvert—in other words, going from being infected, which then itself takes a few days to become PCR positive—to get to the point where there's levels of antibodies in which a serological test, of which there are a number of them are now increasingly accurate, would lead to a positive result. And because of that, that process takes multiple weeks. So unless someone has had then the day after they got sick, but even then it wouldn't show up as a positive result. Serological tests, when looking for antibodies, are not really diagnostics. And so you want to inevitably take what is going to be a resource—we would like it to be less of a scarce resource, but it's not an unlimited resource. And there's a cost to getting people in and the infrastructure required to do that. Logistics and otherwise. So we view a monthly testing rate as seemingly compatible with logistics. It's an idea I think people can wrap their heads around. The models suggest, with these 99.8-percent-specific tests coming out that it could be actionable. And given the fact that it takes maybe about that long to seroconvert, it's not a bad timescale either.
Steven Cherry And then as well, the 2 or 5 percent or whatever people who have tested positive don't need to be tested every month, although you do think they do need to be tested every four months or so for safety's sake.
Joshua Weitz Well, I think that the advantage that's something that I think we're still debating with the advantage of retesting is actually to start building information on variation in antibody titer and the more that we learn about how that correlates with protection, the more benefit we can get, the subsense, subsampling and repeat testing, we're going to learn different information. Right. In some sense, again, if you view it through a visa analogy, right, immunity visa, then retesting helps to confirm, helps to keep you in that sense, but also has a scientific benefit with respect to, we're accumulating knowledge overall. I should also say there's other benefits of serological testing. We're getting more information on candidates for convalescent plasma therapies and so on. So a whole bunch of benefits potentially that come ... that are ancillary when we start to think more carefully about systematic and population-scale serological testing.
Steven Cherry Well, I'm convinced and I hope the world gets convinced that this idea of shield immunity would be a big benefit as we struggle through life with the virus.
Joshua Weitz I think that I am glad to hear that. But I also think that, for someone listening, I view this as a process. So it's critical to think about serological testing as a potential intervention. And the more that we prepare for new evidence to come in, the better off we'll be rather than waiting too late to take action.
Steven Cherry Very good. Well, thank you and all of your colleagues at Georgia Tech and elsewhere for your work. And thanks for your time today.
Joshua Weitz Thank you, Steven. I enjoyed the conversation.
We've been speaking with Georgia Tech Professor Joshua Weitz about a scheme for reducing the virus exposures and deaths that are inevitable once we relax our policies of sheltering-in-place. This interview was recorded March 14th, 2020. Our audio recording and engineering was done remotely by Gotham Podcast Studio. Our music by Chad Crouch. For TTI/Vanguard, I'm Steven Cherry.
This interview was recorded March 14, 2020.
Audio engineering by Gotham Podcast Studio, New York, N.Y.
Music from the song “Algorithms,” by Chad Crouch.
We welcome your comments @ttivanguard and @techwiseconv
Note: Transcripts are created for the convenience of our readers and listeners. The authoritative record of TTI/Vanguard’s audio programming is the audio version.
Modeling shield immunity to reduce COVID-19 epidemic spread
By Joshua S. Weitz, Stephen J. Beckett, Ashley R. Coenen, et al. in Nature Medicine, May 7, 2020
Protecting the population with immune individuals
By Ole F. Norheim in Nature Medicine, May 7, 2020
Quantitative Viral Ecology: Dynamics of Viruses and Their Microbial Hosts
By Joshua S. Weitz (Princeton Univ Press, 2016)