Pushkin Pushkin from Pushkin Industries. This is Deep Background, the show where we explore the stories behind the stories in the news. I'm Noah Feldman. Before we start today, I want to let you know that this is the last episode in our season about power. After this, we're going on a hiatus for the next few months to work on our next season. If you have thoughts or suggestions, please let me know. You can find me on Twitter at Noah R. Feldman, or you can reach me on my website Noah Dashfeldman dot com, where I also have a newsletter which you can use so that we can keep in touch while I'm off the air. It's been a joy to explore behind the headlines with you all these last two seasons, and I want to take a moment to thank you for joining us on this journey over the years. Okay, on with the show. For our last episode of the season, we're returning once more to the biggest story of them all, the global pandemic. Joining us once again on just about the exact second anniversary of his first time on the show, we have Mark Lipsich, the brilliant and influential epidemiologist from Harvard University's School of Public Health. Mark is a professor of epidemiology and he directs the Center for Communicable Disease Dynamics. It's no exaggeration to say that Mark has been the go to person for interpreting the trajectory of the COVID nineteen pandemic and indeed for trying to make sense of its future. Mark, thank you so much for joining me. It's almost exactly two years since we first had a conversation in the very early days of realization of what was having with COVID, and we've checked in periodically since. And I would like nothing better than for this to be a valedictory conversation on the topic, so that next time we could talk about some other aspect of disease dynamics. But somehow it won't quite go away. So how many more times are you and I are going to do this? I mean, but put another way, how many more letters of the Greek alphabet are we going to need to learn? Is this is the variant process, one that is probabilistically likely to just come back again and again and again. Well, until we see a protracted period without new variants, we have to assume they will keep coming. What their characteristics will be we can make some speculations about but those speculations would not have probably included omicron in the form that it had, so we should be humble about our speculations as well. I mean, I think when you have tens to hundreds of millions of people infected in any given a short period of time around the globe with a virus, there is an opportunity for natural selection and genetic drift to play their tricks and for new variants to evolve, and our immune systems impose selective pressure. Our transmission behavior impose a selective pressure. So the ingredients are there. We all knew sort of intellectually that variants were a possibility. There was some hope that maybe the evolutionary rates were a little bit lower in coronaviruses, but that has not turned up to be true. And in fact, the big puzzle in some of my friends minds right now is why is omicron so different from from the last one. So I think we're going to be at this COVID game for a long time and the real question what the impact of that on health is and what the properties of these variants are, which is I think the hardest thing to predict other than likely being more and more transmissible, and the easier thing to predict but still not certain, is what is the state of us in terms of our immunity and the level to which vaccine and prior infection confer on us lasting immunity. I would love to learn more in our conversation Mark about what are the kinds of logics that you and your colleagues use when trying to talk about what those properties might be, understanding that they're not perfectly predictable. So you mentioned transmissibility, that seems like the obvious one. What are other criteria of fitness that you usually ascribe to viruses that would lead one to make any kind of speculations? And it might be interesting just to hear some examples of what you expected to happen how omicron differs from those. Yeah. Well, so transmission for the virus is sort of, roughly speaking, a two part job. It has to find new naive hosts, and as those become less and less common because more people are immune from prior infection or vaccination or both, it also needs to transmit to immune hosts. So one sort of generalization is that initially the selection is for more transmissibility. When almost everyone as naive, and as more people are immune, it's always good to be more transmissible. But there's more and more of a premium on being able to escape immunity, particularly for infection and transmission, Meaning there's not direct selection to escape immunity against severe outcomes that may happen as a side effect of selection for immune escape. That's not the primary goal of the virus. So I mean market for one quick second, I mean it's never the primary goal of a virus to make you very sick. It's just that the virus will do whatever it takes to make more of the virus, and if it happens to be then making you very sick, does that the virus will do that at least in so far as it doesn't kill you and therefore make it harder to spread to more people. Right, That's right, And that's why everybody's watching very carefully to see whether the immune escape of omicron, for example, is to one extent it also escapes the immunity to severe disease that people have, and it's less but to some extent it does escape even that level that type of immunity. Okay, so sorry interrupted, but go go ahead back to what you're saying. So as you say, the virus wants to make more copies of itself once in an evolutionary sense, to make more copies of itself. There's been a long standing discussion in the evolutionary biology community and parallel, mostly separate discussion in the medical and public health community about what that does to the severity of viruses. And in fact, that was what I wrote my PhD thesis on long time ago, was making models of that interaction between the severity of the infection and the direct targets of selection, which are more transmission. So in the old days, people said, in a sort of hand wavy way, without much quantitative reasoning, that if you kill somebody that you in fact then you can't spread if you're a virus, And so viruses try to become over evolutionary time, less harmful to their hosts, and at the same time the hosts become immune. It was then, on sort of conceptual grounds, noted that that's true up to a point. But if a virus is trying to grow in a host and does so so weakly that it doesn't make more copies of itself, that's the other extreme. So killing the host is too much growth, and not even making many copies of itself to transmit is the other extreme. And so the notion of that viruses evolved to an intermediate level of virulence to their host was sort of what many models predict, and then factors that can modulate that include how it's transmitted and the relationship between how many copies of itself it makes and the transmission rate. So the old idea they always try to become more mild is not widely held by people who know what they're talking about, but is widely held by the general public. Right, yeah, So is it correct even to say that, as a probabilistic matter, of the viruses we know about, they are more likely to evolve to being less harmful over time, or is even that which I've sometimes heard as a fallback statement, Is even that not empirically observably true. I don't know if we have enough of the history of viral virulence to look back at and watch. The most famous experiment was the release of mixomatosis, a disease of rabbits caused by maxomavirus in Australia, which was done as an effort to kill off the Australian pest rabbit population, and indeed that started out as an exceedingly virulent virus and did evolve to become less harmful beautiful. Experiments done by Frank Fenner, who was better known for a smallpox eradication, he showed very conclusively that it did become milder. It didn't become mild, it became milder, and the counterbalancing problem was that if it became very, very mild, it just couldn't grow in the rabbits. So that's probably the best documented example of watching evolution in practice, because rabbits and that virus both reproduce quickly, so you had a lot of generations each year. There is also a sort of mechanistic consideration that seems to be true in flu viruses and might be true also for omicron, which is that replicating deep in the lungs is a different property from replicating further up the respiratory tract. Replicating deep in the respiratory tract down in the lungs is more likely to cause severe disease, and replicating up at the top of the respiratory tract is more likely to spread. So to the extent that that's a tradeoff. There might be some selection for transmission that is indirectly selection for mildness. But you can imagine a situation where that's not true. Is it fair to say? Then? So I'm trying to get what one should say around educated people. Somebody says, well, you know, the great thing about omicron is that it shows you that, like all viruses, this is getting more mild. I mean, the first thing to say is that's an old view and it's not on conceptual grounds. Is not necessarily true. And then there are certainly counterexamples of viruses that have not evolved to be substantially less virulent. Are there any that come to mind? What happened with smallpox? By the way, which you mentioned, my layman senses that smallpox was still pretty virulent up to the time when it was eliminated. Yeah, that's right. What's difficult is that proving a negative is hard, and we didn't see the initial emergence of most viruses when they became human pathogens or pathogens of whatever organisms they In fact, that's why Maxsoma introduced into Australia was such a great example. You could watch the whole time course. But I think it's hard to settle the history, especially because we usually also can't separate the changes in the host from the changes in the virus. In other words, as we become more immune, even the same virus will cause less severe disease. Mark we want to warn readers not to generalize for the mixed homa case to thinking that's true of all viruses. Tell me again so that I understand it. The conceptual reason that we cannot assume the way people did in the old days through hand waving. As you said that overtime virulence would necessarily be reduced. In this kind of Dayton picture, virulence is selected along with transmissibility to maximize transmissibility. And so when there is a trade off between making more copies of the virus and harming the host, the virus will choose making more copies of itself. When there's a reinforcement between making more copies of itself and harming the host, the virus will choose making more copies of itself. So what it does to us, as you said at the beginning, it doesn't really care. It's trying to maximize its own function. It is true, I think where there have been observations of evolution. The virulence has gone down in several examples, and I can't think of an example where it started mild and became more virulent that we can really document, but that would be hard to see. If it did happen, we might just see when it jumped out. So I think the thing to say is that there are some examples where viruses have become more mild, and there are some reasons why that might happen, But more mild and completely harmless are very different things, and that really the part that we can control is our immunity. We can control to some extent, and that is almost always a beneficial thing in terms of virulence of the virus. So that's the part where at least we can do something about it. You mentioned that one thing that variance want to do is of course they want to find new naive hosts, but over time that gets harder and harder, and so then they want to be able to get around sound immunity. Omicron seems to be pretty good at getting around immunity. Is there a sense a clear sense now of how it does so? And if so, is that a strategy that is likely to be replicated in future variants. Yeah, I mean, I think what's striking about omicron is that it not only has some mutations in its spike protein, which is a major target of antibodies, but it has a lot of mutations, more than it should given the amount of time that's passed if it were a typical virus out there. So I think the mechanism is still being understood in detail. But pridly speaking, the shape of its protein is different from the shape that the antibodies recognized which we're generated to previous variants, and so those antibodies don't bind as well, they don't neutralize as well, and therefore don't work as well. Any sense, people must be theorizing about this. You must be theorizing about this out why the jump seems to be as big as it is from earlier variants. Yeah, the evolution was faster than is typical in typical humans, and so the two leading hypotheses are it happened in a not typical human, meaning a human who perhaps had an immune system that was just good enough to exert selection but not good enough to end the infection. Part of why evolution doesn't happen too much during a typical infection is that there's a lot of selection to escape it, but not much time before the virus is gone, and dead viruses don't mutate, so once you clear an infection, it doesn't have the chance to evolve further. Immuno Compromised people in various forms often have much longer infections. If they're deeply immunocompromised, then they they may not exert much selection on the virus, and they may not survive for very long because they won't be able to control the virus. But if they are then this is just speculation. I mean compromised in some particular way that allows the virus to persist, but maintains that selection pressure that could accelerate. The other class of explanations is that, as we've seen, the Saris Kobe two can infect other mammals. For example, there was a lot of interest in the white tailed deer being infected. They were minks in Denmark that were infected earlier in the pandemic. So entering a new host species is often a trigger for accelerated evolution because the virus has a lot of new problems to solve to get good at infecting that host and So those are the two major hypotheses that people have thought about. We'll be right back. Mark word that one hears more and more is COVID is on its way to being endemic. Sometimes you hear people getting fancy and talking about a state of endemicity. My question is, first of all, what's the working definition that epidemiologists use for that? So what does endemic actually mean to an epidemiologist? And the second is in practice, what does that mean if and when we are there? I think we're The dirty laundry of the epidemiology community continues to be aired, and different people use these words differently, but I think the core idea of endemicity is that an infection is present at some level, almost constantly in a population, and it may go through say seasonal swings, and sometimes you call that seasonal upturn an epidemic, and sometimes people call averaging over the whole year to be endemic an endemic state. But it's a situation in which most people, by the time they reach some reasonably young age, have been exposed to the infection. That's an important characteristic, and where the swings are either very small, as with some infections that are just sort of pretty similar all year round, like some respiratory bacterial infections that are a little bit seasonal but basically kind of always there at similar levels, or they're predictable swings that are every winter or every summer or something like that. So it has aspects of predictability of most people having prior exposure. And therefore the reason why it's sort of a comforting word is that the two things that have really been miserable about sarskovi two have been the very large and unpredictable swings in the number of people infected, and the fact that it's preying on a population of us that is mostly naive to it and therefore at greater risk of having bad outcomes then we will be when we've all had the infection and or vaccination. The upshot that if I understand you correctly, is that even though endemic and epidemic are obviously etymologically related in terms of their relationship to the public, they're not in any very very sharp relationship to each other. Because something can be endemic and yet there can be epidemics that are seasonal. Is that right? So that those two and yet or you can have an epidemic that is not endemic, or you can sending it is endemic and is not epidemics. So it seems to have almost all the possible logical relationships. But I think I think the more helpful contrast probably is pandemic versus endemic. So the signal features of pandemics, although these are also revised as we as we have more of them, because they're each one is special. But the signal features of pandemics are a very naive population that is experiencing something for the first time, and being very widespread and having large swings and incidents and prevalence because of those two things. Yeah, so that's super helpful. I mean, that's super helpful because one can sort of think a pandemic nobody has it at first, then lots of people get it in bunches, and that's the pandemic endemic. Everybody's been exposed to it from the time that they were pretty small children, but it's still out there in the population. That seems to be very, very helpful. The thing that is a bit confusing to me at least is if we think about the relationships, say of omicron, to the earlier versions of COVID, it seems like even though in principle we should be moving to a place where almost everybody has been exposed or vaccinated one of the two that lots of people who were previously exposed to those other things or were vaccinated are getting o macron because of its capacity to evade immunity. And if that's the case, do you guys start thinking of it as though it were a separate disease and so therefore it doesn't really matter if let's say it becomes endemic. Let's say omicron becomes endemic, but something a new variant comes along that evades immunity, then aren't people sort of a naive population again with respect to that new variant if they're evading immunity, Yeah, that is the big question. The reason why omicron has been bad but not worse than it was is that the combination of somewhat lower severity and growing amounts of population immunity to severe disease, which still to a large degree though not perfectly hold against omicron, has meant that there are enormous numbers of cases, but not correspondingly many deaths and hospitalizations. And that's a kind of new thing. We haven't had a flu experience that's quite like that, and we haven't had a far as Kobe two experience. It's quite like that. So this is a new combination of properties to deal with. One way that I find it helpful to think about what's ahead in the next few years is that, as you say, the bad option will be if new variants continued to come and they really are either so good at escaping immunity and transmitting that the cases go shooting up, as they have with Omicron, or they are really good at escaping our immunity to severe disease, which was not that high at the beginning of Omicron, should be higher after Omicron, we hope. But if they're good at escaping our immunity to severe disease, then even modest numbers of cases could be a problem, and that would be like almost like a new flu pandemic. It would be an evolutionary jump that is highly consequential and is what we used to talk about with flu pandemics. Right. It's the reason that the flu vaccine, it's a good thing to get it, but it doesn't guarantee that you won't get the flu, right because the variants comes so fast. Right, partly and partly because they just aren't fully protective even against what they're supposed to talk and nearly as good as these these vaccines that I say, I kind of categorize like the two possible futures are that we have changes like the usual changes between years of flu, or we have changes like new flu pandemics that just are more frequent, And those are the two broad categories that we really really hope the first happens, but we have to consider the possibility in the second. Mark I'm trying to figure out what the right kinds of policy responses are at this moment, partly in order to get sense of what they're going to look like going forward if we have more variants. So, for example, at our mutual employer, Harvard University, on the one hand, we have classes at least at the law school in person, but we are all told not just to wear masks, but to wear a ninety five masks and possibly to wear two masks. And there seems to be a heightening of concern about transmission because I guess of the great transmissibility of omicron. On the other hand, we're in person, which we weren't obviously at an earlier stage, and What's more, everyone is mandatorily vaccinated and boostered, So what's the policy logic if you can reconstruct it behind that kind of heightened prevention coupled with diving right in and having classes. Well, another part of it, at least at our school public health, is that contact tracing is basically now left to the individual. There's no centralized contact tracing. So putting that together with what you just described, my interpretation is that people are trying to do the things that are efficient and relatively low impact that reduced transmission, while reallocating resources from very very costly in terms of time efforts like contact tracing, and trying to return to doing as much of what we're supposed to be doing as a university as possible. So that's my interpretation. And an environment where let's say almost everybody or everybody has been vaccinated and boosted, then the goal would be try to avoid too many people getting omicron so that they have to miss class and stay home and give more people omicron, but nevertheless still try to push forward with as much normalcy as can be mustered under these circumstances. And I guess the question that leads me to is and It's not that I think either has the answer to it, but I want to talk about it. Is this the new normal? Right? I mean, is there any reason if people can bear going to class in masks or teaching in masks? Will there be a policy argument to you me going forward that even if as the omicron surge continues to decline, as it's doing in Massachusetts, we should just stick with this because it's a good preventive measure. To get people to comply with something like that, there would have to be a continuing real threat of severe outcomes in large numbers in the populations doing those behaviors, and or of overwhelming healthcare, which are closely related, of course, although not perfectly correlated. Because the burden on hospitals from omeocron has been enormous, the severity has been less, and that's because so many people were getting it at the same time. So I can't imagine that in a few years we will all be wearing masks for our daily life. I don't expect it, because I think, in my mind, the most likely scenario is that our immunity to severe disease continues to get ampleified by probably continued vaccination, and also continued circulation of the virus, and that this becomes a vaccine preventable disease, at least for the severe manifestations, and therefore we will start to dial back these control measures as they are dialing back. In fact, in many parts of the country already well in some parts of the country, they never really dialed up, right, They were dialed back from the very beginning, and they never stopped. I have a good friend in Oklahoma who this is long ago in this process. I asked him something about a public lecture and I said, he mentioned public and I said, there's going to be a public lecture, and he just laughed and he said, we have never stopped doing these things, he said, And by the way, my wife and I and all of our kids have had COVID, so's there's clearly a lot of local variation around the country. Yeah, But I mean, I think it's important to consider all the ways things could go bad, and it's really hard to do that at the saint and actually really put policy energy towards preparing for those and at the same time to have I think a reasonable level of optimism, like a realistic level of optimism that those things might well not happen, and we probably will not be wearing masks and a few hears, I think, and maybe even much sooner than that. So I mean, I think it's even possible that in say April, the amount of omicron immunity will be so great that will have a long rest from this. But it's also possible that we will have a new variant that creates bigger problems. Last question mark one of the optimistic thoughts. I don't want to say fantasies, but hopes that when sometimes hears about response to potential future variance, is one vaccine to rule them all, like some sort of a vaccine that is sufficiently protean and powerful that all it'll affect almost all coronaviruses, maybe all coronaviruses, such that you know, whatever these variants may be, they will probably be coronaviruses, and then we'll fall under that rubric. I guess what I'm wondering is, from the standpoint of epidemiology, do you have any thoughts on what are the probabilities of such a thing being possible? Are there other instances or areas where we've had effective vaccines that cover like a broad range of phenomena under some rubric like say coronavirus. Yeah, there have been. And the approach so far that has been the most effective with the vaccine that I used to study before COVID, which was the new Mecoco vaccine, and with the polio vaccine and with the flu vaccine is to just try to vaccinate people against everything that's around at once, and then with flu to change it over time. So those precedents exist, but other than flu, those are for things where there's a lot of variation and you need coverage of a whole bunch of different variants, but you have a list of them. It's not a changing list. It's just the list, or maybe a new one's discovered periodically, but it's kind of a fixed list, especially if they're ones that are worst. So that approach is well worn. There is a very strong effort right now underway in many quarters to create paying coronavirus vaccines. There will be different approaches of trying to put in a mix of different types, or of trying to find parts of the virus that are more conserved where immunity will be effective. And there's reason to be optimistic, but it still has to be proven. So I think it's a very worthwhile effort and we have to see if it works. Mark, thank you so much again for your insights. And I would say that I look forward to more conversations, and that would be true about this substance of the conversations, because a conversation with you is always illuminating. I hope we don't need too many, Moore, but my guess is we will be checking in in one form or another. Thank you so much for your time, look forward to talking about something else. Sometimes we'll be right back. As always, when talking to Mark Lipsitch, there are clear takeaways. First, as a probabilistic matter, it is overwhelmingly likely that variants are going to keep coming, possibly for a long time. The reason for that, Mark explains is that there is virus out there in the population, and while virus is out there among a large number of people, the probability that evolution will continue to do its work and develop new variants remains. Next, Mark points out that omicron has been a surprise in various respects, most particularly in how different it is than the variance that came before it, thereby suggesting that for various possible reasons that we're not yet sure of the evolutionary process is happening faster than we previously expect it. Third, and this is to my mind very important, we should not accept the viewpoint which we sometimes here expressed by non experts, that all viruses necessarily evolve in the direction of becoming less virulent. In fact, the virus will do whatever it takes to replicate itself, seen in evolutionary metaphorical terms, and that means that if it turns out that the virus can spread more effectively while being less virulent, it will do so. But if the virus needs to be more virulent in order to spread more effectively, it will do that too. Consequently, we cannot generalize and assume that we are on the way to things automatically gradually getting better via omicron Fourth, we are certainly approaching some quality that could be called endemicity, or a state of being endemic for COVID nineteen, defined to mean that almost everybody in the population will have been exposed to the disease, often relatively early in life. However, and this is the big however, whether that matters in the long run in terms of making it less likely that people get very sick depends entirely on how good future variants are at avoiding immunity. If people get sick when young and retain immunity, then that is a good result in terms of the effects of endemicity. But if future variants on the virus can still infect people who have nevertheless been exposed, the serious and indeed severe problems associated with the healthcare system and COVID nineteen could in principle continue even when the disease is endemic last and very much not least. Mark says that the possibility of a vaccine that would effectively block future variance by virtue of it being general enough to prevent all forms of the coronavirus from spreading or alternatively from getting the people who get them very sick, is worth pursuing. Whether it will succeed or not, of course, remains an unknown and not something on which it's possible to put a probabilistic numerical judgment. At the human level, Mark says he thinks it's very unlikely that in a couple of years or even sooner than that will all still be walking around wearing masks. That, for one leaves me some reason to continue to be hopeful. Ultimately, and this is me talking, not Mark. The power of this virus has been to shape the way we in the world respond to it in a broad set of complicated ways that themselves deploy our own conception of social power. Disease has power, Humans have power in responding to disease. All of it involves us in a set of complex trade offs and choices. The difference between us and evolution is that evolution is making those choices blindly, with no normative vision of how it wants to be or how the world should be arranged. We as humans are in a different position. Our judgments can be thoughtful, strategic, and ethically inclined, and with any luck, they will continue to be. As I mentioned at the beginning of the show, we hear a deep background are going to take a hiatus for a few months as we work on our next season. We'll miss you, but I very much hope to hear your voices through reaching me on Twitter at Noah are Feldman or perhaps even more efficiently, on my website at Noah Dashfeldman dot com. Until the next time I speak to you, breathe deep, think deep thoughts, and, if at all possible, under the circumstances of our world, try to have a little fun. Deep Background is brought to you by Pushkin Industries. Our producer is mo La Board, our engineer is Ben Taliday, and our showrunner is Sophie Crane mckibbon. Editorial support from noahm Osband. Theme music by Luis Guerra at Pushkin. Thanks to Mia Lobell, Julia Barton, Lydia, Jean Coott, Heather Faine, Carlie mcgliori, Maggie Taylor, Eric Sandler, and Jacob Weisberg. You can find me on Twitter at Noah R. Feldman. I also write a column for Bloomberg Opinion, which you can find at Bloomberg dot com slash Feldman. To discover Bloomberg's original slate of podcasts, go to Bloomberg dot com slash podcasts, and if you liked what you heard today, please write a review or tell a friend. This is Deep Background.

Pushkin Pushkin from Pushkin Industries. This is Deep Background, the show where we explore the stories behind the stories in the news. I'm Noah Feldman. Before we start today, I want to let you know that this is the last episode in our season about power. After this, we're going on a hiatus for the next few months to work on our next season. If you have thoughts or suggestions, please let me know. You can find me on Twitter at Noah R. Feldman, or you can reach me on my website Noah Dashfeldman dot com, where I also have a newsletter which you can use so that we can keep in touch while I'm off the air. It's been a joy to explore behind the headlines with you all these last two seasons, and I want to take a moment to thank you for joining us on this journey over the years. Okay, on with the show. For our last episode of the season, we're returning once more to the biggest story of them all, the global pandemic. Joining us once again on just about the exact second anniversary of his first time on the show, we have Mark Lipsich, the brilliant and influential epidemiologist from Harvard University's School of Public Health. Mark is a professor of epidemiology and he directs the Center for Communicable Disease Dynamics. It's no exaggeration to say that Mark has been the go to person for interpreting the trajectory of the COVID nineteen pandemic and indeed for trying to make sense of its future. Mark, thank you so much for joining me. It's almost exactly two years since we first had a conversation in the very early days of realization of what was having with COVID, and we've checked in periodically since. And I would like nothing better than for this to be a valedictory conversation on the topic, so that next time we could talk about some other aspect of disease dynamics. But somehow it won't quite go away. So how many more times are you and I are going to do this? I mean, but put another way, how many more letters of the Greek alphabet are we going to need to learn? Is this is the variant process, one that is probabilistically likely to just come back again and again and again. Well, until we see a protracted period without new variants, we have to assume they will keep coming. What their characteristics will be we can make some speculations about but those speculations would not have probably included omicron in the form that it had, so we should be humble about our speculations as well. I mean, I think when you have tens to hundreds of millions of people infected in any given a short period of time around the globe with a virus, there is an opportunity for natural selection and genetic drift to play their tricks and for new variants to evolve, and our immune systems impose selective pressure. Our transmission behavior impose a selective pressure. So the ingredients are there. We all knew sort of intellectually that variants were a possibility. There was some hope that maybe the evolutionary rates were a little bit lower in coronaviruses, but that has not turned up to be true. And in fact, the big puzzle in some of my friends minds right now is why is omicron so different from from the last one. So I think we're going to be at this COVID game for a long time and the real question what the impact of that on health is and what the properties of these variants are, which is I think the hardest thing to predict other than likely being more and more transmissible, and the easier thing to predict but still not certain, is what is the state of us in terms of our immunity and the level to which vaccine and prior infection confer on us lasting immunity. I would love to learn more in our conversation Mark about what are the kinds of logics that you and your colleagues use when trying to talk about what those properties might be, understanding that they're not perfectly predictable. So you mentioned transmissibility, that seems like the obvious one. What are other criteria of fitness that you usually ascribe to viruses that would lead one to make any kind of speculations? And it might be interesting just to hear some examples of what you expected to happen how omicron differs from those. Yeah. Well, so transmission for the virus is sort of, roughly speaking, a two part job. It has to find new naive hosts, and as those become less and less common because more people are immune from prior infection or vaccination or both, it also needs to transmit to immune hosts. So one sort of generalization is that initially the selection is for more transmissibility. When almost everyone as naive, and as more people are immune, it's always good to be more transmissible. But there's more and more of a premium on being able to escape immunity, particularly for infection and transmission, Meaning there's not direct selection to escape immunity against severe outcomes that may happen as a side effect of selection for immune escape. That's not the primary goal of the virus. So I mean market for one quick second, I mean it's never the primary goal of a virus to make you very sick. It's just that the virus will do whatever it takes to make more of the virus, and if it happens to be then making you very sick, does that the virus will do that at least in so far as it doesn't kill you and therefore make it harder to spread to more people. Right, That's right, And that's why everybody's watching very carefully to see whether the immune escape of omicron, for example, is to one extent it also escapes the immunity to severe disease that people have, and it's less but to some extent it does escape even that level that type of immunity. Okay, so sorry interrupted, but go go ahead back to what you're saying. So as you say, the virus wants to make more copies of itself once in an evolutionary sense, to make more copies of itself. There's been a long standing discussion in the evolutionary biology community and parallel, mostly separate discussion in the medical and public health community about what that does to the severity of viruses. And in fact, that was what I wrote my PhD thesis on long time ago, was making models of that interaction between the severity of the infection and the direct targets of selection, which are more transmission. So in the old days, people said, in a sort of hand wavy way, without much quantitative reasoning, that if you kill somebody that you in fact then you can't spread if you're a virus, And so viruses try to become over evolutionary time, less harmful to their hosts, and at the same time the hosts become immune. It was then, on sort of conceptual grounds, noted that that's true up to a point. But if a virus is trying to grow in a host and does so so weakly that it doesn't make more copies of itself, that's the other extreme. So killing the host is too much growth, and not even making many copies of itself to transmit is the other extreme. And so the notion of that viruses evolved to an intermediate level of virulence to their host was sort of what many models predict, and then factors that can modulate that include how it's transmitted and the relationship between how many copies of itself it makes and the transmission rate. So the old idea they always try to become more mild is not widely held by people who know what they're talking about, but is widely held by the general public. Right, yeah, So is it correct even to say that, as a probabilistic matter, of the viruses we know about, they are more likely to evolve to being less harmful over time, or is even that which I've sometimes heard as a fallback statement, Is even that not empirically observably true. I don't know if we have enough of the history of viral virulence to look back at and watch. The most famous experiment was the release of mixomatosis, a disease of rabbits caused by maxomavirus in Australia, which was done as an effort to kill off the Australian pest rabbit population, and indeed that started out as an exceedingly virulent virus and did evolve to become less harmful beautiful. Experiments done by Frank Fenner, who was better known for a smallpox eradication, he showed very conclusively that it did become milder. It didn't become mild, it became milder, and the counterbalancing problem was that if it became very, very mild, it just couldn't grow in the rabbits. So that's probably the best documented example of watching evolution in practice, because rabbits and that virus both reproduce quickly, so you had a lot of generations each year. There is also a sort of mechanistic consideration that seems to be true in flu viruses and might be true also for omicron, which is that replicating deep in the lungs is a different property from replicating further up the respiratory tract. Replicating deep in the respiratory tract down in the lungs is more likely to cause severe disease, and replicating up at the top of the respiratory tract is more likely to spread. So to the extent that that's a tradeoff. There might be some selection for transmission that is indirectly selection for mildness. But you can imagine a situation where that's not true. Is it fair to say? Then? So I'm trying to get what one should say around educated people. Somebody says, well, you know, the great thing about omicron is that it shows you that, like all viruses, this is getting more mild. I mean, the first thing to say is that's an old view and it's not on conceptual grounds. Is not necessarily true. And then there are certainly counterexamples of viruses that have not evolved to be substantially less virulent. Are there any that come to mind? What happened with smallpox? By the way, which you mentioned, my layman senses that smallpox was still pretty virulent up to the time when it was eliminated. Yeah, that's right. What's difficult is that proving a negative is hard, and we didn't see the initial emergence of most viruses when they became human pathogens or pathogens of whatever organisms they In fact, that's why Maxsoma introduced into Australia was such a great example. You could watch the whole time course. But I think it's hard to settle the history, especially because we usually also can't separate the changes in the host from the changes in the virus. In other words, as we become more immune, even the same virus will cause less severe disease. Mark we want to warn readers not to generalize for the mixed homa case to thinking that's true of all viruses. Tell me again so that I understand it. The conceptual reason that we cannot assume the way people did in the old days through hand waving. As you said that overtime virulence would necessarily be reduced. In this kind of Dayton picture, virulence is selected along with transmissibility to maximize transmissibility. And so when there is a trade off between making more copies of the virus and harming the host, the virus will choose making more copies of itself. When there's a reinforcement between making more copies of itself and harming the host, the virus will choose making more copies of itself. So what it does to us, as you said at the beginning, it doesn't really care. It's trying to maximize its own function. It is true, I think where there have been observations of evolution. The virulence has gone down in several examples, and I can't think of an example where it started mild and became more virulent that we can really document, but that would be hard to see. If it did happen, we might just see when it jumped out. So I think the thing to say is that there are some examples where viruses have become more mild, and there are some reasons why that might happen, But more mild and completely harmless are very different things, and that really the part that we can control is our immunity. We can control to some extent, and that is almost always a beneficial thing in terms of virulence of the virus. So that's the part where at least we can do something about it. You mentioned that one thing that variance want to do is of course they want to find new naive hosts, but over time that gets harder and harder, and so then they want to be able to get around sound immunity. Omicron seems to be pretty good at getting around immunity. Is there a sense a clear sense now of how it does so? And if so, is that a strategy that is likely to be replicated in future variants. Yeah, I mean, I think what's striking about omicron is that it not only has some mutations in its spike protein, which is a major target of antibodies, but it has a lot of mutations, more than it should given the amount of time that's passed if it were a typical virus out there. So I think the mechanism is still being understood in detail. But pridly speaking, the shape of its protein is different from the shape that the antibodies recognized which we're generated to previous variants, and so those antibodies don't bind as well, they don't neutralize as well, and therefore don't work as well. Any sense, people must be theorizing about this. You must be theorizing about this out why the jump seems to be as big as it is from earlier variants. Yeah, the evolution was faster than is typical in typical humans, and so the two leading hypotheses are it happened in a not typical human, meaning a human who perhaps had an immune system that was just good enough to exert selection but not good enough to end the infection. Part of why evolution doesn't happen too much during a typical infection is that there's a lot of selection to escape it, but not much time before the virus is gone, and dead viruses don't mutate, so once you clear an infection, it doesn't have the chance to evolve further. Immuno Compromised people in various forms often have much longer infections. If they're deeply immunocompromised, then they they may not exert much selection on the virus, and they may not survive for very long because they won't be able to control the virus. But if they are then this is just speculation. I mean compromised in some particular way that allows the virus to persist, but maintains that selection pressure that could accelerate. The other class of explanations is that, as we've seen, the Saris Kobe two can infect other mammals. For example, there was a lot of interest in the white tailed deer being infected. They were minks in Denmark that were infected earlier in the pandemic. So entering a new host species is often a trigger for accelerated evolution because the virus has a lot of new problems to solve to get good at infecting that host and So those are the two major hypotheses that people have thought about. We'll be right back. Mark word that one hears more and more is COVID is on its way to being endemic. Sometimes you hear people getting fancy and talking about a state of endemicity. My question is, first of all, what's the working definition that epidemiologists use for that? So what does endemic actually mean to an epidemiologist? And the second is in practice, what does that mean if and when we are there? I think we're The dirty laundry of the epidemiology community continues to be aired, and different people use these words differently, but I think the core idea of endemicity is that an infection is present at some level, almost constantly in a population, and it may go through say seasonal swings, and sometimes you call that seasonal upturn an epidemic, and sometimes people call averaging over the whole year to be endemic an endemic state. But it's a situation in which most people, by the time they reach some reasonably young age, have been exposed to the infection. That's an important characteristic, and where the swings are either very small, as with some infections that are just sort of pretty similar all year round, like some respiratory bacterial infections that are a little bit seasonal but basically kind of always there at similar levels, or they're predictable swings that are every winter or every summer or something like that. So it has aspects of predictability of most people having prior exposure. And therefore the reason why it's sort of a comforting word is that the two things that have really been miserable about sarskovi two have been the very large and unpredictable swings in the number of people infected, and the fact that it's preying on a population of us that is mostly naive to it and therefore at greater risk of having bad outcomes then we will be when we've all had the infection and or vaccination. The upshot that if I understand you correctly, is that even though endemic and epidemic are obviously etymologically related in terms of their relationship to the public, they're not in any very very sharp relationship to each other. Because something can be endemic and yet there can be epidemics that are seasonal. Is that right? So that those two and yet or you can have an epidemic that is not endemic, or you can sending it is endemic and is not epidemics. So it seems to have almost all the possible logical relationships. But I think I think the more helpful contrast probably is pandemic versus endemic. So the signal features of pandemics, although these are also revised as we as we have more of them, because they're each one is special. But the signal features of pandemics are a very naive population that is experiencing something for the first time, and being very widespread and having large swings and incidents and prevalence because of those two things. Yeah, so that's super helpful. I mean, that's super helpful because one can sort of think a pandemic nobody has it at first, then lots of people get it in bunches, and that's the pandemic endemic. Everybody's been exposed to it from the time that they were pretty small children, but it's still out there in the population. That seems to be very, very helpful. The thing that is a bit confusing to me at least is if we think about the relationships, say of omicron, to the earlier versions of COVID, it seems like even though in principle we should be moving to a place where almost everybody has been exposed or vaccinated one of the two that lots of people who were previously exposed to those other things or were vaccinated are getting o macron because of its capacity to evade immunity. And if that's the case, do you guys start thinking of it as though it were a separate disease and so therefore it doesn't really matter if let's say it becomes endemic. Let's say omicron becomes endemic, but something a new variant comes along that evades immunity, then aren't people sort of a naive population again with respect to that new variant if they're evading immunity, Yeah, that is the big question. The reason why omicron has been bad but not worse than it was is that the combination of somewhat lower severity and growing amounts of population immunity to severe disease, which still to a large degree though not perfectly hold against omicron, has meant that there are enormous numbers of cases, but not correspondingly many deaths and hospitalizations. And that's a kind of new thing. We haven't had a flu experience that's quite like that, and we haven't had a far as Kobe two experience. It's quite like that. So this is a new combination of properties to deal with. One way that I find it helpful to think about what's ahead in the next few years is that, as you say, the bad option will be if new variants continued to come and they really are either so good at escaping immunity and transmitting that the cases go shooting up, as they have with Omicron, or they are really good at escaping our immunity to severe disease, which was not that high at the beginning of Omicron, should be higher after Omicron, we hope. But if they're good at escaping our immunity to severe disease, then even modest numbers of cases could be a problem, and that would be like almost like a new flu pandemic. It would be an evolutionary jump that is highly consequential and is what we used to talk about with flu pandemics. Right. It's the reason that the flu vaccine, it's a good thing to get it, but it doesn't guarantee that you won't get the flu, right because the variants comes so fast. Right, partly and partly because they just aren't fully protective even against what they're supposed to talk and nearly as good as these these vaccines that I say, I kind of categorize like the two possible futures are that we have changes like the usual changes between years of flu, or we have changes like new flu pandemics that just are more frequent, And those are the two broad categories that we really really hope the first happens, but we have to consider the possibility in the second. Mark I'm trying to figure out what the right kinds of policy responses are at this moment, partly in order to get sense of what they're going to look like going forward if we have more variants. So, for example, at our mutual employer, Harvard University, on the one hand, we have classes at least at the law school in person, but we are all told not just to wear masks, but to wear a ninety five masks and possibly to wear two masks. And there seems to be a heightening of concern about transmission because I guess of the great transmissibility of omicron. On the other hand, we're in person, which we weren't obviously at an earlier stage, and What's more, everyone is mandatorily vaccinated and boostered, So what's the policy logic if you can reconstruct it behind that kind of heightened prevention coupled with diving right in and having classes. Well, another part of it, at least at our school public health, is that contact tracing is basically now left to the individual. There's no centralized contact tracing. So putting that together with what you just described, my interpretation is that people are trying to do the things that are efficient and relatively low impact that reduced transmission, while reallocating resources from very very costly in terms of time efforts like contact tracing, and trying to return to doing as much of what we're supposed to be doing as a university as possible. So that's my interpretation. And an environment where let's say almost everybody or everybody has been vaccinated and boosted, then the goal would be try to avoid too many people getting omicron so that they have to miss class and stay home and give more people omicron, but nevertheless still try to push forward with as much normalcy as can be mustered under these circumstances. And I guess the question that leads me to is and It's not that I think either has the answer to it, but I want to talk about it. Is this the new normal? Right? I mean, is there any reason if people can bear going to class in masks or teaching in masks? Will there be a policy argument to you me going forward that even if as the omicron surge continues to decline, as it's doing in Massachusetts, we should just stick with this because it's a good preventive measure. To get people to comply with something like that, there would have to be a continuing real threat of severe outcomes in large numbers in the populations doing those behaviors, and or of overwhelming healthcare, which are closely related, of course, although not perfectly correlated. Because the burden on hospitals from omeocron has been enormous, the severity has been less, and that's because so many people were getting it at the same time. So I can't imagine that in a few years we will all be wearing masks for our daily life. I don't expect it, because I think, in my mind, the most likely scenario is that our immunity to severe disease continues to get ampleified by probably continued vaccination, and also continued circulation of the virus, and that this becomes a vaccine preventable disease, at least for the severe manifestations, and therefore we will start to dial back these control measures as they are dialing back. In fact, in many parts of the country already well in some parts of the country, they never really dialed up, right, They were dialed back from the very beginning, and they never stopped. I have a good friend in Oklahoma who this is long ago in this process. I asked him something about a public lecture and I said, he mentioned public and I said, there's going to be a public lecture, and he just laughed and he said, we have never stopped doing these things, he said, And by the way, my wife and I and all of our kids have had COVID, so's there's clearly a lot of local variation around the country. Yeah, But I mean, I think it's important to consider all the ways things could go bad, and it's really hard to do that at the saint and actually really put policy energy towards preparing for those and at the same time to have I think a reasonable level of optimism, like a realistic level of optimism that those things might well not happen, and we probably will not be wearing masks and a few hears, I think, and maybe even much sooner than that. So I mean, I think it's even possible that in say April, the amount of omicron immunity will be so great that will have a long rest from this. But it's also possible that we will have a new variant that creates bigger problems. Last question mark one of the optimistic thoughts. I don't want to say fantasies, but hopes that when sometimes hears about response to potential future variance, is one vaccine to rule them all, like some sort of a vaccine that is sufficiently protean and powerful that all it'll affect almost all coronaviruses, maybe all coronaviruses, such that you know, whatever these variants may be, they will probably be coronaviruses, and then we'll fall under that rubric. I guess what I'm wondering is, from the standpoint of epidemiology, do you have any thoughts on what are the probabilities of such a thing being possible? Are there other instances or areas where we've had effective vaccines that cover like a broad range of phenomena under some rubric like say coronavirus. Yeah, there have been. And the approach so far that has been the most effective with the vaccine that I used to study before COVID, which was the new Mecoco vaccine, and with the polio vaccine and with the flu vaccine is to just try to vaccinate people against everything that's around at once, and then with flu to change it over time. So those precedents exist, but other than flu, those are for things where there's a lot of variation and you need coverage of a whole bunch of different variants, but you have a list of them. It's not a changing list. It's just the list, or maybe a new one's discovered periodically, but it's kind of a fixed list, especially if they're ones that are worst. So that approach is well worn. There is a very strong effort right now underway in many quarters to create paying coronavirus vaccines. There will be different approaches of trying to put in a mix of different types, or of trying to find parts of the virus that are more conserved where immunity will be effective. And there's reason to be optimistic, but it still has to be proven. So I think it's a very worthwhile effort and we have to see if it works. Mark, thank you so much again for your insights. And I would say that I look forward to more conversations, and that would be true about this substance of the conversations, because a conversation with you is always illuminating. I hope we don't need too many, Moore, but my guess is we will be checking in in one form or another. Thank you so much for your time, look forward to talking about something else. Sometimes we'll be right back. As always, when talking to Mark Lipsitch, there are clear takeaways. First, as a probabilistic matter, it is overwhelmingly likely that variants are going to keep coming, possibly for a long time. The reason for that, Mark explains is that there is virus out there in the population, and while virus is out there among a large number of people, the probability that evolution will continue to do its work and develop new variants remains. Next, Mark points out that omicron has been a surprise in various respects, most particularly in how different it is than the variance that came before it, thereby suggesting that for various possible reasons that we're not yet sure of the evolutionary process is happening faster than we previously expect it. Third, and this is to my mind very important, we should not accept the viewpoint which we sometimes here expressed by non experts, that all viruses necessarily evolve in the direction of becoming less virulent. In fact, the virus will do whatever it takes to replicate itself, seen in evolutionary metaphorical terms, and that means that if it turns out that the virus can spread more effectively while being less virulent, it will do so. But if the virus needs to be more virulent in order to spread more effectively, it will do that too. Consequently, we cannot generalize and assume that we are on the way to things automatically gradually getting better via omicron Fourth, we are certainly approaching some quality that could be called endemicity, or a state of being endemic for COVID nineteen, defined to mean that almost everybody in the population will have been exposed to the disease, often relatively early in life. However, and this is the big however, whether that matters in the long run in terms of making it less likely that people get very sick depends entirely on how good future variants are at avoiding immunity. If people get sick when young and retain immunity, then that is a good result in terms of the effects of endemicity. But if future variants on the virus can still infect people who have nevertheless been exposed, the serious and indeed severe problems associated with the healthcare system and COVID nineteen could in principle continue even when the disease is endemic last and very much not least. Mark says that the possibility of a vaccine that would effectively block future variance by virtue of it being general enough to prevent all forms of the coronavirus from spreading or alternatively from getting the people who get them very sick, is worth pursuing. Whether it will succeed or not, of course, remains an unknown and not something on which it's possible to put a probabilistic numerical judgment. At the human level, Mark says he thinks it's very unlikely that in a couple of years or even sooner than that will all still be walking around wearing masks. That, for one leaves me some reason to continue to be hopeful. Ultimately, and this is me talking, not Mark. The power of this virus has been to shape the way we in the world respond to it in a broad set of complicated ways that themselves deploy our own conception of social power. Disease has power, Humans have power in responding to disease. All of it involves us in a set of complex trade offs and choices. The difference between us and evolution is that evolution is making those choices blindly, with no normative vision of how it wants to be or how the world should be arranged. We as humans are in a different position. Our judgments can be thoughtful, strategic, and ethically inclined, and with any luck, they will continue to be. As I mentioned at the beginning of the show, we hear a deep background are going to take a hiatus for a few months as we work on our next season. We'll miss you, but I very much hope to hear your voices through reaching me on Twitter at Noah are Feldman or perhaps even more efficiently, on my website at Noah Dashfeldman dot com. Until the next time I speak to you, breathe deep, think deep thoughts, and, if at all possible, under the circumstances of our world, try to have a little fun. Deep Background is brought to you by Pushkin Industries. Our producer is mo La Board, our engineer is Ben Taliday, and our showrunner is Sophie Crane mckibbon. Editorial support from noahm Osband. Theme music by Luis Guerra at Pushkin. Thanks to Mia Lobell, Julia Barton, Lydia, Jean Coott, Heather Faine, Carlie mcgliori, Maggie Taylor, Eric Sandler, and Jacob Weisberg. You can find me on Twitter at Noah R. Feldman. I also write a column for Bloomberg Opinion, which you can find at Bloomberg dot com slash Feldman. To discover Bloomberg's original slate of podcasts, go to Bloomberg dot com slash podcasts, and if you liked what you heard today, please write a review or tell a friend. This is Deep Background.