IE 11 is not supported. For an optimal experience visit our site on another browser.

Breaking down how pandemics end with Ellie Murray: podcast and transcript

Chris Hayes speaks with Ellie Murray, epidemiology assistant professor at the Boston University School of Public Health, about where we are in the COVID crisis.

The COVID pandemic has been going on for more than two years. Will it ever end? It’s a question that doesn’t have a straightforward answer, as much as we all desperately want one. Dr. Ellie Murray is an epidemiology assistant professor at the Boston University School of Public Health. She also runs the Murray Causal Decision Lab and co-hosts the “Casual Inference” podcast in partnership with the American Journal of Epidemiology. Murray joins WITHpod to discuss the importance of smart public health messaging, ethical and sociological concerns regarding the determination of “acceptable” virus case numbers and misconceptions of what endemicity means. She also talks about strategies for protecting those most at risk and why it’s possible that COVID could mutate into something much more dangerous if too much complacency continues.

Note: This is a rough transcript — please excuse any typos.

Ellie Murray: It's absolutely within the realm of possibility, that if we continue to do what we're doing, basically, most people back to normal, lots and lots of virus out there, it's allowed to mutate. We're not really tracking cases. We're not doing very much virus sequence surveillance in the U.S. It's entirely plausible that this could mutate into something really quite a lot worse.

Chris Hayes: Hello and welcome to "Why Is This Happening?" with me your host, Chris Hayes.

I've had the experience a few times a lot, actually, in the last few weeks, where I have started a sentence by saying like, “Well, since the pandemic is over, or since the pandemic ended,” or something like that, and then I stopped myself because pandemic, of course, is very much not over. And every day, there's someone else I know tested positive, “Oh, we were supposed to have the cousins. We’re all coming to town for the reunion. But now, I've tested positive and isn't that a pain.”

In the vast majority of cases, at least personally, in the world that I've been in, and people that I know and I love, almost all of them are vaccinated and often boosted as well. They have not led to health complications. They have not been severe cases that have led to hospitalization. This even includes people who are immunocompromised, have survived cancer or things like that. There's a variety of people's health statuses.

By and large, in this part of the pandemic, I haven't known people that have been hospitalized for severe illness. And the numbers basically bear that out, even though we have pretty robust transmission happening. Basically, the combination of prior infection, which produces some antibodies and immunity with vaccination and boosters, has produced what I've heard referred to as an immunity wall, which means we're not in the same place we were in 2020 when the whole reason everyone was freaking out was that the, quote, “novel Coronavirus” was a new virus for which no human on Earth had any protection. That was precisely what was so terrifying. It was taking a match to utterly, utterly, utterly dry tinder, right?

So that's not the case anymore. That said, the virus continues to mutate. There's no law of nature or science that it couldn't mutate into something much more deadly or much more severe, unfortunately. So we're like in a very weird middle space. It doesn't feel like anyone is throwing like a V-E Day Parade. Like, oh, it's done. We're done with COVID. But it's also just demonstrably not the place we were before. It is a far lower threat to your health than it was certainly in 2020 pre-vaccine. And I would even say during the delta wave, I think that the numbers bear out that in the aggregate, it is a less dangerous thing that is floating around, and yet still there.

But it's weird to try to kind of conceptualize both from a public health perspective and a kind of sociological perspective, like, so is this it? Like, are we done? Is this the kind of quasi-permanent status quo? Is it just the thing like when I have my 60th birthday, it will be like “Test for COVID. Oh, shoot, my friend can't come. He just got COVID.” Like, is that going to be what it's like? My 60th, oh, my God, it makes me so depressed just to say that, even though I guess it's possible. We're going to find out.

We're going to get the definitive answer to what's going to happen in 17 years from today's guest, Dr. Ellie Murray. She caught my eye because she wrote this really, really smart, long thread on Twitter about precisely this question like when is a pandemic over? What does it mean for it to be over?

She's an epidemiologist, assistant professor at Boston University School of Public Health, and runs the Murray Causal Decision Lab in the Department of Epidemiology there, along with her team. She's a social media editor at the American Journal of Epidemiology. She also partners with the American Journal of Epidemiology, where she co-hosts the “Casual Inference” podcast, which features chats about epidemiology, and statistics, and data science. And Dr. Murray, it's great to have you on the program.

Ellie Murray: Great to be here. Thanks for inviting me.

Chris Hayes: First, just tell me about yourself. You're an epidemiologist. My understanding is your area of specific focus is kind of how do we determine cause and effect in large pools of data around disease and viral transmission.

Ellie Murray: Yes, exactly. So my research focus is really about the sort of methodological. How do we make sense of data on population health? Largely, the questions we want to ask are not things we can do interventions for, because interventions are great if you have some kind of treatment, if you want to study something that's harmful. You can't give people COVID and see what happens. And so we need a lot of other tools. And so that's really where my area of expertise is, in how do we get sort of the truth from that kind of data. And also, how do we communicate to people in a way that they can use it to make decisions.

Chris Hayes: Well. That's pretty relevant, I think, to this chapter of the pandemic because a lot of it is about trying to figure out what is and isn't effective, and also about what the levels of risk are at population levels, when it's not like we're running vaccine trials, right? So everyone is kind of out doing their thing. And we're trying to take the data pool, the datasets we have, to try to make determinations about how dangerous is it? How much risk is it? What is or isn't helping on the margins?

Ellie Murray: Right, exactly. And so, for example, last summer, the big question was, do we need boosters yet? And that was something where we were kind of stuck with observational data, what does it look like is happening to people who have and haven't gotten the vaccine? And does that seem like it's changed from the trials? And then, eventually, Israel did more of a sort of formal, like, “Let's assign people to get boosters and we can get an answer.”

But at first, there was a lot of trying to make sense of observational data, where you have this problem that the people who got vaccinated the earliest in our vaccine rollout were the people who are highest risk of exposure or highest risk of bad outcomes?

Chris Hayes: Right. So that's a pretty confounding variable.

Ellie Murray: Right. So the whole reason those people got early access to the vaccine means that we think that they're very likely to get COVID. And so if we see them getting COVID, compared to people who have not been vaccinated, is that just the reason we have vaccinated them at the beginning, or has the vaccine started to wear off?

Chris Hayes: So we're now in this position where it's all kind of a muddle and a mess. I mean, I don't know, I don't even know how, how would you describe where we are in this pandemic? Let me ask you this, so you're the expert.

Ellie Murray: So I've been calling this like Schrodinger’s wave, because it feels like one of those things where it's depends how you look at that data. It feels like you're in a wave or you're not. I think, like you, so many people I know have recently been infected this past spring. I do know some people who've had some pretty serious consequences of it. We're hearing all these stories about absenteeism in workplaces and things like that. But the testing numbers, I mean, they're high, but they're not quite what you would expect if everyone is getting COVID.

On the other hand, people are taking rapid tests, and they're not going for PCR tests, and the rapid tests aren't getting counted in the statistics. Omicron, among unvaccinated people, it's still roughly putting people in the hospital at the same rate. But for vaccinated people, they're much less likely to be hospitalized. And when they are hospitalized, their stays are much shorter. So if you just look at the total hospitalization numbers, they're way down. But the new hospitalization numbers are not down as much. And so, there's all this conflicting information out there. So Schrodinger’s wave is my shorthand for that.

Chris Hayes: Well, I think the question I wanted to sort of wrestle with today with you is like, what does it look like for the pandemic to be over? When will it be over? But maybe it's easier if we start at the beginning of like, I remember when it was the headlines, WHO declares, COVID-19 a pandemic. What is starting? What does it mean when a pandemic start? Like, what does that designation mean?

Ellie Murray: So this sounds like it should be an easy question. But the thing that I think is making this whole process really kind of complicated is that all of our systems have been set up in a time where the only type of pandemic we've ever had really is an influenza pandemic. And so, all of our pandemic definitions are specifically about influenza, and pandemic influenza criteria. And so you can see the WHO has this nice, like six levels of a pandemic. And if you're in Level 5 and 6, it's definitely a pandemic. If you're in Level 4, you're maybe in a pandemic. But the definitions like specifically say the influenza cases are, blah, blah, blah.

And when the WHO declares a pandemic, what they actually declare is a public health emergency of international concern. And that's not necessarily the same thing as just a pandemic. Anyway, it's not necessarily the same thing as what the public thinks when they hear pandemic.

Chris Hayes: Well, let's talk in non-technical terms. Let's just talk in like what is it? Like, let's get to the most basic question, right? There's lots of viruses that are going around at any time. And there's lots of possible illnesses a person gets. We're talking about communicable diseases, right, and things that you can catch from another person. There's a bunch that are going around anytime. There's the normal Coronaviruses that make up the cold. There is various strains of influenza and flu.

There's sexually transmitted infections and diseases that people can get from each other. There's mononucleosis. I mean, there's just a whole bunch of stuff that people can give each other, right? And that's just part of living on the planet. So what does it mean when like something pops out of that background level of transmissibility to become an epidemic or a pandemic?

Ellie Murray: Yeah. So I think that this is kind of a good way of looking at what's normal and then what's abnormal.

Chris Hayes: Right. Yeah.

Ellie Murray: Because we have a pretty good understanding of what we mean by the idea of an outbreak, which is pretty much synonymous with epidemic. And an outbreak is when you have an unusually high number of cases of a disease for a given population, time, area, context. So if we saw a single case of bubonic plague in New York City, then that would count as an outbreak because we don't expect any. Zero is the usual number.

Chris Hayes: Right.

Ellie Murray: But if we saw a single case of syphilis, we'd be like, “Wow, New York is really syphilis free.

Chris Hayes: Right.

Ellie Murray: And so it depends on the disease, the population, the time, place, et cetera. So if that's what an outbreak is, then a pandemic should be that but basically everywhere. And it doesn't necessarily need to be everywhere all at once, but more the idea that anywhere you go has the potential to be imminently in an outbreak setting for that disease.

Chris Hayes: That's a very useful way to think about it. We know what an outbreak is, right? So like if you were like me of the generation pre-chickenpox vaccination. When there was an outbreak of chickenpox in your class, like, you know what an outbreak, right? There's an outbreak of chickenpox. It's very, very communicable. Everyone in the class got it. Nine people were out.

My son, there's a stomach bug going around his class. And like, he told me this amazing story of like being in class, like, yeah, there were 13 people in class today out of 24. Like, 11 of them were home, and then like four more went home that day. That's a bug that’s racing around the class, right? That's like the norovirus will do that. So we know what that looks like, right? In small situations, when something just burns through a place, that's an outbreak.

Ellie Murray: Right.

Chris Hayes: So what you're saying is what a pandemic is, is that there's some kind of communicable disease that's out there, that means that everyone everywhere is just like possibly about to have an outbreak.

Ellie Murray: Right. And I think that if we think about it in that way, then really what that kind of boils down to is that there is some disease which, basically, in most places of the world is just not currently controlled by the things that we're doing. So that means that its level at any given time is unpredictable, and we don't have a good grasp on what's going to happen a month from now, what's going to happen 17 years from now at your 60th birthday party.

Chris Hayes: Right. Well, so what are examples of things that were outbreaks and then not, right? So like, I'm trying to think of things, HIV is a great example, right? HIV/AIDS, where there was completely uncontrolled spread. It was centered on certain populations in the beginning. The risk factors were incredibly disproportionate throughout much of the AIDS epidemic. But now, it's not that anymore. And the question of like when it went from the epidemic to not is a little hazy to me.

Ellie Murray: Yeah. So in some ways, by some criteria, you could still call HIV and AIDS a pandemic because span of human history, it's relatively new. So we would generally expect there to be zero cases. But if we look on a much more smaller timescale, for the U.S., for the U.K., we have a pretty good handle on HIV. We have preventive medications. We have preventive behaviors. We have great treatments. We have good detection systems. We need to do a little bit better at getting people into detected and into care.

But in general, our numbers are staying pretty static over time, so that we know how many new HIV doctors we need. We know how much prep we need to order for the next six months. These are easy and unpredictable. And so, that would sort of put it into the realm of not a pandemic anymore.

Chris Hayes: So this is key. Like, this notion of steady state, steady state equilibrium, predictability, a line that you can track. Generally, there's going to be this much tuberculosis in a population, this much syphilis, this much mononucleosis, this much HIV. Like, those are all things that are communicable, that have posed very severe and acute public health problems, that require both sort of pharmaceutical and non-pharmaceutical interventions in terms of dealing with. So they have all those features that are similar to COVID-19. But what they don't have is it's not like crazy growth, crazy outbreaks happening all the time.

Ellie Murray: Yeah, exactly. And so, then we can sort of say, okay, well, syphilis is generally pretty consistent over time. HIV is generally pretty consistent over time. What about something like norovirus, or the flu where there's a season when that happens? And we do expect that the sort of normal expected thing with norovirus is that in the winter, there'll be a certain proportion of classrooms that are like your sons, where everybody ends up being sick home with that.

Chris Hayes: That thing is unreal. I mean, I have one family vacation, once early into the virus, it's like my wife's family and all. It is a truly monstrously infectious situation, if you're like in a household with a bunch of people.

Ellie Murray: Right. And so, I mean, imagine if anywhere you went at any given time, you could be about to enter a norovirus outbreak situation.

Chris Hayes: Right. But why is that not the case? Is that not the case just because of seasonality?

Ellie Murray: So norovirus is seasonal. And we monitor it and we have controls on it. One of the places it spreads is in the healthcare settings. And so all of these doctors washing their hands kind of campaigns, things like that help reduce norovirus that spreads really, really well in child care settings. And so, keeping schools as clean as possible, keeping kids as clean as possible. These kinds of things are important. And then like in the summer, they can get outside and play, and they're not like spreading their fecal and all each other, just basically how norovirus transmits. I'm sorry. But those activities all together keep it low.

And so we see things like in a cruise ship, where it's really, really, really hard to keep all surfaces clean and pristine, with people throwing up all over the place, and like stumbling down the halls, touching everything as they like, the boat rocks, or whatever. We see this huge outbreaks all the time, because that's a setting where we don't have a good system of control. But in general, our systems of control work pretty well. Maybe they could work better, but it seems like nobody has an interest in making them work better. And so, by controlled, we kind of mean we're happy with the level it's at.

Chris Hayes: Right. And this gets to, I think, this like real profound, almost like sociological question, right, which is that the answer to this question is like, what do we tolerate? Like, how do we define normal? There's not actually like a scientific answer to the question. There's not a public health answer to the question. It's like, what can we abide? What can we live with? And like, we couldn't live with like cruise ship level norovirus in office buildings year round all the time. Like, it would be unbearable. Like, you could have people calling in sick. Like, that wouldn't be a bearable thing. We can live with it like once a year.

Ellie Murray: Right.

Chris Hayes: So that's just kind of the difference, right?

Ellie Murray: Right.

Chris Hayes: It's like what we will or won't tolerate, what we can or can't abide at some level at a broad sociological sense.

Ellie Murray: Yes, absolutely. And so this way, I always say like as a scientist, as an epidemiologist, I can tell you how to go about figuring out how to control an infectious disease. I can go about telling you how to figure out whether an infectious disease is controlled, but I can't tell you what the level it should be controlled at. That's a political decision.

And so, one of the things there is, that I have been really saying over and over again, and pushing for is more of a public conversation about what is our target, right?

Chris Hayes: Yeah.

Ellie Murray: And how many deaths a day from COVID are we happy with? Are we happy if we can keep it out of the top 5 deaths and it's just somewhere below five on the list of things that kill Americans? Do we want to keep it out of the top 10 deaths? Do we want to do better than that?

Chris Hayes: And when are we willing to do that, right? I mean, that's the other thing, right? Because there's a cost benefit trade-off here, and I think this is really where the rubber hits the road in this discussion. Ron Johnson, the senator from Wisconsin said this thing very early in the pandemic, where he said, “There's trade-offs, and like, we could put the speed limit at 40 miles an hour or 30 miles an hour, and a lot less people would die on the road. But like, we don't do that.”

Now, this was a ludicrous thing to say at the moment because car crashes aren't contagious, and they don't have exponential growth. So like, obviously, if car crashes on a given set of roads in Wisconsin, were contagious and exponentially grew such that all cars in Wisconsin were totaled in a few weeks. Like, no one would be like, “Well, we can't do anything, right?” So the whole point is that like that logic doesn't work when you're dealing with exponential growth, outbreak, pandemic spread like we were.

Ellie Murray: Right.

Chris Hayes: But once you have a bunch of means and immunity wall treatments, like a whole bunch, a whole suite of things you can throw at the problem, then you start to get more into this question of the tolerable level, right?

Ellie Murray: Right, exactly. And so, I mean, just with the car crash example, the counter argument might be in situations where it's like a heavy snowstorm, where one crash can easily lead to another crash as cars piled up on each other, we slow down.

Chris Hayes: Right. And we keep people off the road.

Ellie Murray: We keep people off the road. We slow down. We reduce, right?

Chris Hayes: Right.

Ellie Murray: Because that's a situation where actually you can have locally exponential growth contact.

Chris Hayes: Contagious. Right.

Ellie Murray: And so we do something there.

Chris Hayes: Right. That's exactly right.

Ellie Murray: So once we have exponential growth, this is where it really becomes important to think about what are we going to do? When are we going to stop it? Because it's a lot easier to stop it early than it is to stop it late when something is growing exponentially, right? Like, the difference between stopping it at 200 cases or 400, 2,000 cases, or 4,000, the amount of resources you need as it grows is also exponential.

Chris Hayes: Right. But then it seems to me like we're not in that place anymore with COVID in the U.S., right? Like, we have outbreaks where I'm sure like there's exponential growth among a small population, right, the people at this birthday party. But we're not seeing these like crazy, I mean, maybe we are, do we still see sort of like, I guess we do, right?

Ellie Murray: Yeah. I mean, basically, every surge is an epidemic growth curve --

Chris Hayes: Exponential, correct. Right, yeah.

Ellie Murray: -- exponential up, exponential-ish down. Some of them come down a little bit slower because we stopped doing precautions, and then they kind of take off a little bit more. But, yeah, every time we see a surge, that's the exponential growth happening on a population level.

Chris Hayes: And that's why it is really different than others. I mean, that's what's so weird about where we're at right now because, like, other stuff doesn't really do that, except with seasonality. But this is not doing it with seasonality.

Ellie Murray: Right. So I mean, it's a little bit complicated because when we look at something like the flu, we see the seasonality. And there is pretty good evidence that humidity levels do have something to do with our susceptibility as humans to the flu virus. So in certain humidity conditions, we are more susceptible, and that increases our risk of catching it.

But the flu also mutates at such a speed that about every seven, eight months, if you encountered the flu again, it would be a completely different flu as far as your body is concerned. It may not be totally, totally different so that you get really sick, but you could still catch it again. And so we see that combined with the weather kind of creates the seasonality.

With COVID, it's not clear how like environmental conditions are driving it. But it is the case that every big surge we've seen has been a different variant or sub-variant, right?

Chris Hayes: Right.

Ellie Murray: So it seems like whether or not there are environmental things that increase or decrease our personal susceptibility, the virus is changing quickly enough, or there's enough diversity out there in the virus pool itself, that every four, five, six months, we're encountering a new variant, and we're getting sick again.

Chris Hayes: That's right. That's what's driving it right now, right? It's mutating, and the mutation is escaping. I mean, everyone, I think, now knows people have gotten it two, three, sometimes four times. I've only had it once. But again, there's a profound kind of like ethical question about the value of life, and particularly the value of life of people that are marginalized or immunocompromised, right? And like, what society, as a whole, should be doing, thinking about the value of their life and keeping them safe. So that's this, like, really profound social and ethical question.

Then there's a kind of like an adjacent macro policy question about sort of trade-offs, right? Like, you can shut down all public buildings, but then that has other effects. There are negative effects to people not circulating. And kids not going to school was a perfect example. But then there's just like, if I go to the first person, which is like what will I tolerate?

This is the way I think about it. I guess I could deal if it's just going to be the deal that like we all get COVID like once a year. And it's kind of a bummer, not severe, but it's sucks. But I can't tolerate that like three times a year. And I sure as hell can't tolerate it if like each time creates some significant risk of some long-term problems.

Ellie Murray: Right.

Chris Hayes: And I'm just not sure, like it sort of seems like maybe that's the world we're stumbling into.

Ellie Murray: Yeah. And I think that sort of both of these things are kind of unknown. So the early variants that we saw in the first three or four waves, my understanding, and I'm not a viral geneticist so this is probably something to follow up. But my understanding is that those mutations all arose roughly at the same time. And so, those weren't the virus mutating from wave to wave. It was just like there was a bunch of versions out there, and we kind of caught them sequentially.

Chris Hayes: Yeah. And some outcompeted others, right?

Ellie Murray: Right. Whereas Omicron seems to be kind of like, first, we had one version of Omicron, and then we're getting the next up variant and the next up variant.

Chris Hayes: Yeah.

Ellie Murray: And that looks like maybe Omicron is out competing the rest of the variants, and then itself kind of evolving in ways that keep getting us sick. And so maybe Omicron will become kind of stable enough that we don't really see these surges. That would be a best case scenario probably. It's also possible that some other mutation is out there, that's worse, that just hasn't infected the right people to take off, or that Omicron could develop into something worse.

Chris Hayes: Well, I don't want to hear that.

Ellie Murray: And the more people that we have that are infected, the bigger pool of potential options there are for the virus. So that's the first unknown.

Chris Hayes: Right.

Ellie Murray: And then the second unknown is, as you say, we don't know what this virus does to you 10 years down the line, five years down the line. No one on Earth has had it that long.

Chris Hayes: Right. Yeah. I mean, I want to sort of put aside the long COVID question, because I think that actually deserves its own podcast because that's a very complex and fraught area of real medical and scientific disagreement, partly because we don't have the data to make definitive statements.

Ellie Murray: And I think that there's issues there. There's long COVID. And there's also issues like if you're someone who, in developing COVID, then developed a heart condition.

Chris Hayes: Right, exactly.

Ellie Murray: And now, you have just this long-term illness that is sort of a sequela of COVID, which may or may not be considered long COVID when we kind of firmly decide on a decision of that. But there are, clearly, problems that happen as a result of COVID.

Chris Hayes: More of our conversation after this quick break.


Chris Hayes: Let's talk about endemic, like going from term pandemic to endemic and what that term means, because I think a lot of people have talked about it and this was the thing that caught my eye because you wrote about this in the thread. Like, the idea that it would go from epidemic to endemic, and that's going to be the sort of denouement of our COVID story. What does endemic mean in an epidemiological context?

Ellie Murray: So endemic in public health, kind of generally means just that a disease is occurring in a kind of expected way, in a population, in a given time. So we can think about it as that sort of controlled, not an outbreak definition. For epidemiologists who do modeling, in the modeling context that has a much more specific definition, that is sort of the long run average value of the number of secondary cases is below a certain threshold, then it's endemic.

But that definition has really only been applied in the modeling context to say, because if you want to study the properties of a model when it's occurring normally at expected levels, you need to be able to figure out when your model is running at that expected level. In practice, in public health, it really just has that sort of vague like not in an outbreak or pandemic meaning.

Chris Hayes: Right. Like not doing the big epi curves.

Ellie Murray: Right. So we generally think of seasonal flu, we probably would generally classify that as endemic because it's predictable and we have a rough idea of how much flu to expect every winter in every area. But we still see pandemic flu happening on top of, or in addition to seasonal flu, where we have some new strain that shows up and spreads really rapidly.

Chris Hayes: Right. Like the swine flu, or the avian flu, or whatever.

Ellie Murray: Right. Exactly. So another thing is like if we had endemic COVID, could we still also have pandemic COVID from a different strain, variant? Maybe. I don't know. It's not clear. So far, we've only had like one main variant circulating at a time. It doesn't necessarily mean that's how it has to always be. It's not the case for the flu. The flu, I mean, we always have several variants circulating at a time for the flu.

The other piece about the term endemic, though, is that when you really look into the history of how that word has actually been used, it has this sort of other kind of implicit meaning, that endemic is a word that people have used when they want to say, “This disease is not a problem for people like me. It's only a problem for people like them. And those are people I don't really care about.” And typically, it might be someone talking about endemic cholera in some colonial town or something like that, where it's like those people over there, in this country that I think is lesser than me, those people who are someone I think of a second class, they have the disease. They need to worry about it. I don't need to worry about it.

Chris Hayes: That's really interesting. So endemic, like in the cholera context, is like it's an acceptable level of cholera for them.

Ellie Murray: Exactly.

Chris Hayes: Cholera is just going to be around for them, for that population. And so, cholera there is endemic. Whereas, like later, right, in public health, we have a zero cholera approach.

Ellie Murray: Right.

Chris Hayes: But the general approach is like zero cholera, no cholera is tolerable. We don't say like, “Well, there's going to be some cholera and you should probably take some precautions like boil your water, or maybe filter it or whatever. But, like, we just have a zero cholera approach.

Ellie Murray: Right. Exactly. So if we sort of socially look down on people who took cruise ships, we might say there's endemic norovirus on cruise ships, and like those people get norovirus. The rest of us don't need to worry about it. And that's what we've kind of been seeing with the way that endemic has been used for COVID, too, is that we're seeing things, like people saying, like, “Oh, well, it's really only the unvaccinated that have to worry about COVID right now. So it's endemic. Those bad unvaccinated people are getting COVID. The good vaccinated people are not getting COVID. So we don't need to worry about it.”

And I mean, first of all, that's not what the data says. Lots of vaccinated people are getting COVID. They're just not having quite as severe outcomes.

Chris Hayes: Right.

Ellie Murray: And also, the unvaccinated people, they're not all unvaccinated because they're like bad people.

Chris Hayes: Right.

Ellie Murray: There's no moral component to being vaccinated or not. There's lots and lots of reasons. There's many reasons as there are unvaccinated people.

Chris Hayes: I would not say there's no moral component for some people in refusing vaccination.

Ellie Murray: But I think if they were just making that decision for themselves, it doesn't tell me that they're a good or bad person. I might say that there's a moral component about them, trying to convince other people not to get vaccinated for something, but the vaccine has been shown to be very effective, that I might then start to think that they're bad people.

Chris Hayes: Right.

Ellie Murray: But if people just decide for themselves, there are lots and lots of reasons why people don't get vaccinated.

Chris Hayes: Sure.

Ellie Murray: And some of them are very extreme. But a lot of them are like, “Well, the clinic is only open between 9:00 and 5:00, and I have to work from 9:00 to 5:00.

Chris Hayes: Sure. Right. Although, on the vaccination question, there has been a little bit of horseshoe theory around COVID and vaccines I've noticed recently, which is like the most sort of extreme take COVID seriously. It's really dangerous and bad, and we need to sustain NPIs indefinitely, at a very high level; end up coming around and sort of touching the anti-vaxxers, which is like vaccines are kind of the high side the point. You're never going to vaccinate your way out of it, yada, yada.

And like, it's true that high level of vaccinations is probably necessary, but not totally sufficient to suppressing at a level that you would really like, like, say, South Korea. But it's also true that vaccinations are really big part of it, and we have much lower levels of vaccination than other pure countries. We're basically in where Russia is. And that's a real problem. Like, it's clearly much, much, much, much worse here than it would be if we had higher levels of vaccination.

Ellie Murray: Right. And I think one of the things we're seeing is like this real polarization in the way people talk about cover precautions, that, like, so many people just have their like one favorite precaution. And they're like, “This is the only thing you should be doing. Everything else is useless.” And that's completely the wrong way to think about it.

Like, yes, the vaccines work really, really well. They work way better than we had any right to hope for in the beginning. However, they don't completely protect you from infection. They don't completely protect you from severe disease and death. They don't completely protect you from transmitting it to other people. So alone, vaccines are not sufficient. And you can show mathematically that alone, vaccines will not stop the pandemic. They will not get rid of COVID. They may not even be enough to get COVID to be always controlled.

Chris Hayes: Correct.

Ellie Murray: So we need something else. We need other things. And promoting masks, or promoting occasional closures, or distancing, or whatever is not anti-vax. But what we see is that a lot of people have in their mind that there's only one precaution you can take. And so, we end up with these weird things like people who are proponents of upgrading ventilation systems in buildings, but are also dissuading people from wearing masks. And it's like, well, if the ventilation is not upgraded, you should be wearing a mask.

Chris Hayes: Right.

Ellie Murray: They're just two sides of the same coin.

Chris Hayes: We'll be right back after we take this quick break.


Chris Hayes: Part of this, too, is a sociological fact about just people's exhaustion, right? I mean, like one of the things that you see is and you see this in the data on, for instance, boosters. Like, there's tons of seniors who took the vax and are not boosted. Obviously, these are not ideologically anti-vaxxers. I mean, maybe they had some conversion. That's a possibility. But my guess is that the vast majority, they're just kind of like, “It seems fine now,” or I don't know. It's kind of complicated.

And so there is a lot of low hanging fruit, right? Like, our senior booster level should be twice. It's like 35%, I forget, I'm not going to pull a number off the top of my head, right. But that's like the most obvious thing in the world. Like, every senior should get boosted, and there's just millions who are not yet, right?

Ellie Murray: Yes.

Chris Hayes: And that's driving really a lot of preventable death, like a ton of preventable death.

Ellie Murray: Right. And I think what we're seeing here is that there's fatigue, but I don't know that it's necessarily fatigue with doing the things. It's fatigue with having to think about what things are you supposed to be doing. And that kind of falls under the idea of there's just this really big executive function burden that's been put on us in this pandemic --

Chris Hayes: Yes.

Ellie Murray: -- where you have to figure out should you do this event? Should you do that other event? Should you do a rapid test? Should you wear a mask? What kind of masks should you wear? Should you shake hands? Should you hug? Should you stand six feet away? Do you need to be indoors or outdoors? Is it okay to drink? Is it okay to eat?

Chris Hayes: Right.

Ellie Murray: And then, on top of that, the vaccines were rolled out in such a way that it was like this week, Group X, Y, and Z is approved to get a vaccine or a booster. And then the next week, it's like now this very specific group is approved. And now, this very specific group is approved. And so, I am absolutely not an anti-vaxxer. But I didn't get my booster until my work sent me an email saying, “We're going to have a booster clinic, come and get it.”

Chris Hayes: Right.

Ellie Murray: Because I was teaching several classes, and I had no time to keep up with when I became eligible or not. And the classes I was teaching were outbreak investigation and infectious disease upbeat. Like, I was doing all COVID all the time and I couldn't keep up with when I needed a booster.

Chris Hayes: Right.

Ellie Murray: And so, how can we expect people who maybe don't even have the Internet to know?

Chris Hayes: Right. Totally. Yeah. I'm not like trying to blame the people themselves. I mean, here's the fundamental tension that I think we're sort of talking about, which is like the broad social political feeling that this has gone from an uncontrolled thing to a controlled thing, right? With the fact that it's still kind of uncontrolled, and that's really what it comes down to is everyone is kind of past their limit of what, here in the U.S. context and I know other countries have done a much better job. And the way they've done a better job.

I think it's a complicated mix of like public health experiences that they had previously like in South Korea, where they had mares, which was a real disaster there and it changed the way they did a lot of public health, to other places that are, frankly, higher trust societies, which are doing a better job like in parts of Scandinavia. Like, there's a whole complicated reason. But the reality presently now in the exceptional United States of America is that the actual public opinion on this is that people are over it in the aggregate.

Ellie Murray: Yeah. I mean, that's definitely what I'm seeing. People think it's over. And I think that's also the message that we're getting from all of our leadership, that we've been getting for months, right? When the vaccines first rolled out, Walensky said, “Okay, you can take off your masks.”

Chris Hayes: Yeah. But just to push back on that, I think the causality goes the other way. I think the leadership, particularly public health leadership, Biden administration, were doing more than people wanted longer. I think that there was exhaustion and demand exogenous that came from people. I mean, keep in mind, people in huge swaths of the country threw their masks off in April of 2020. Like, this was not a country that was like gung-ho on public health measures from the jump.

Ellie Murray: Yeah. So I think that there's a number of things here. Yeah, a lot of people just didn't even put on masks. Right. And then we saw very conflicting information from different groups. We had bad tests. We had bad data. We had information about like, “It's over. You can do whatever.” So, when we look at surveys of people, asking people, whether they think we should be doing more, or if they want to wear masks, or if they think masks should be mandated, we actually see a lot of support for those ideas.

Chris Hayes: Yes.

Ellie Murray: And I think there's also this confusion. So I mean, the public health system in the U.S. is not a system. It's a bunch of local --

Chris Hayes: Right. There’s just a million --

Ellie Murray: -- health departments and then some state departments. And then the CDC and the administration, often they're completely separate hierarchy. And so, it's not I think that any of the administration have been doing too much. It's only possible for them to do so much, so that everything is like different rules for different neighborhoods half the time, and no one knows what to do.

Chris Hayes: Yeah. And I want to just sort of slightly revise what I was saying before because I think you raised a good point. Like, when I say public opinion, I don't mean like the polling, right. I think that like, in the aggregate, there's a certain level of just mental exhaustion that just came from the pandemic and having to change things. I think there remained a desire for simple clear guidelines, if that was always wear a mask or whatever. And then that became harder to maintain, partly because things got more complicated and because of countervailing pressures.

And so what you ended up with was this patchwork that then became a little bit of like no one liked it. Like, it's like you've got this just incredibly list of executive function problems. So if you're not saying, like, “Always wear a mask in every indoor space,” right, which is a clear thing. But that's a tough thing to maintain because that became extremely politically toxic for a certain very powerful portion of the political spectrum.

So then you say, “Well, it's sort of up to you.” Now, you no longer have any clarity, right? So everyone now is just in this, “You're on your own sort of world.” And we're all muddling through, as again, like we basically live essentially normal lives, except we just keep having outbreaks.

Ellie Murray: Right. And I think that this is exactly it, that there's been this sort of, “Well, figure it out.” And a lot of that came from the fact that because this was new, and everyone is so connected to information as it comes out, that scientists were having to unpack what was happening, and what this virus was, and how it transmitted, and what we needed to do. It was being unpacked like in real time on social media half the time. So people were seeing like the inside of the sausage-making, on figuring out what to do.

And then that seemed like, “Oh, scientists disagree.” And it's like, well, of course, people disagree with something that we've never seen before, right? Like, the story of eight blind men touching an elephant, and one thinks it's a tree and whatever. And it's like, you got to talk to each other to figure out that it's an elephant. And eventually we did that and we came up with some principles. But at that point, it was like, “Oh, everybody disagrees. There's so many conflicting messages.”

Chris Hayes: Right.

Ellie Murray: And everyone had kind of latched on to the piece that they thought it was, and you have this like, “No, it's definitely a tree camp.” And then, “No, it's definitely a snake camp.”

Chris Hayes: Right.

Ellie Murray: And nobody is recognizing that it's an elephant.

Chris Hayes: Right. I mean, the question of what to do now, to me, I mean, when I say we're all living our normal lives, is that true? Like, you have millions of immunocompromised people who are not and can't go to birthday parties, and can't go to Broadway shows, and who are looking at a society that is basically like tough.

At the same time, I think it's very hard and essentially politically impossible to reimpose any large scale, non-pharmaceutical interventions other than maybe mask wearing in public, big public places, in places that are amenable to that. So New York City subway, airports and airplanes, but I mean, that's gone too. Thanks to the right-wing court. So I guess the question is like what is this now? And is this the status quo until it mutates into something that's going to send us all back into lockdown again, which oh, my God, or what?

Ellie Murray: Right.

Chris Hayes: What's the end?

Ellie Murray: I mean, that's the million-dollar question. And part of the reason that is not answerable is that what we do, going forward, can change the answer. So it's absolutely within the realm of possibility, that if we continue to do what we're doing, basically, most people back to normal, lots and lots of virus out there. It's allowed to mutate. We're not really tracking cases. We're not doing very much virus sequence surveillance in the U.S. It's entirely plausible that this could mutate into something really quite a lot worse.

And I've seen protein RNA model simulations that show it evolving into something, with a 10% or more death rate. And I mean, that's an extreme. It's not to say that that won’t happen. But the more viruses out there, the more chance there is that something like that could occur. If we reduce the amount of virus out there, then something like that becomes less and less likely.

But I think the other thing is like, we're in this place where a lot of people think it's over. And they think that means it's not there, and it's never coming back. Whereas if this is it, and we're at “endemic COVID,” quote-unquote, in the best case use of the word and not the, “it's a problem for them” use of the word, what endemic COVID actually looks like, our best guess right now is that we're going to have a surge in cases when schools will go back in the fall. We're going to have a bigger surge after Thanksgiving. We're going to have an even bigger surge after Christmas and New Year's.

And any school that goes back to in-person in the first two weeks of January is going to see school closures because of outbreaks, probably continuing to propagate through the spring, intermittently until it's sort of slowly dies off. And then next summer, we can expect a summer peak because we're seeing a summer fall and a winter peak pretty regularly over the last couple of years. And then the following fall, a peak as school goes back, and then a peak as the holidays come, and then on and on and on. And that means three surges a year is our best case forecast, I mean, our best forecast right now.

Chris Hayes: Wait, but what is it about this virus that makes that the case? Like, what I don't get is, obviously, we're producing antibodies. Like, it's not a naive population anymore. Everyone had it, basically. I mean, not everyone, but the vast majority of people have had it. People had multiple plans. They also have a lot of vaccination immunity. Like, why would it not be the case that that immunity confers, over time, sufficient protection to not have this continued three peaks a year, or is it just that it keeps mutating enough to produce that?

Ellie Murray: So the immune system is so complicated. And so, I mean, there are so many components to it. But the kind of simple answer is that any given pathogen, we could have something where one exposure to it gives you perfect lifelong immunity, which we call sterilizing immunity, which I think is very confusing term for people. But it means that your system sterilizes out the pathogen.

Chris Hayes: The chickenpox version?

Ellie Murray: Right. Although some people can get chickenpox occasionally, multiple times, but it's generally pretty good. Generally, you get it once. You're not going to have it again.

Chris Hayes: Right.

Ellie Murray: Or, on the other side, you can have basically no immunity at all. And it seems like norovirus is probably closer to that end of the spectrum. I've definitely heard stories of people where the kid brings home norovirus from school and it circles through the family, and that eventually circles back to the first one that had it. And it's like an awful loop. So those are all within the realm of possibility, right?

And so the question is where on that spectrum does COVID fall? And part of that is about how good the antibodies our body produces are at identifying this is COVID and I'm going to stop it. And there is some literature out there that suggests that maybe, even though our antibodies are recognizing spike protein, the immune cells responsible with killing those spike proteins aren't actually doing a great job of that. That would move us towards the norovirus end of the spectrum.

The other question is, how often does the virus change? And particularly, how often does the virus change? Those pieces of it, which are the ones that our immune system best recognizes. And this is I think where COVID is tricky because the spike protein, which our vaccines are based on, which our immune system seems to do a pretty good job of recognizing, is actually one of the most variable parts, it looks like, in terms of the structure there. And it's pretty susceptible to changing in ways that make it not quite fit.

And immunity works like a lock and a key, right? So if I came along and I like messed with your door lock, with I stuck like a nail file in there, metal file in there and like shave some things off in your door lock. How long do you think I'd have to do that before your key didn't work?

Chris Hayes: Right.

Ellie Murray: Right? And that's the question, right? How many people have to mess with your door lock before your key doesn't work anymore? Or how many people have to mess with your key? And basically, what we're saying is like the more viruses out there, the more people there are going around like messing with people's locks. The more likely it is the locks are going to fail.

Chris Hayes: Well, so then the question is like, so if you were made dictator of the United States, with plenary power to just dictate all healthcare policies, like what would be the ideal solution for this stage of COVID?

Ellie Murray: So I think what we need is, and I think this is not entirely a scientific question because as you said, people are exhausted.

Chris Hayes: No, not. No.

Ellie Murray: And so that's what makes it so tricky. Like, I can tell you all the things I think, scientifically, you should do. But I think we have told people that this is a “Use your own information, judgment, decision-making processes to make decisions about what risks you're going to take.” And people like, “Nope, no.” How many people even knew the word pandemic before this? How many people knew what a virus was?

The whole ivermectin debacle, I saw so many people saying, “Well, ivermectin should work because it kills parasites and COVID is a parasite.” It's like no, a parasite is like a little creature that lives inside you. A virus is not a parasite. They're not the same thing. Like, we just have basically like zero public health literacy.

Chris Hayes: Yeah.

Ellie Murray: And so it's like we're telling a bunch of 2 year olds to make their own decisions about what to eat for dinner. People are not going to make good choices.

Chris Hayes: Right.

Ellie Murray: And so we need to give people the tools to make some decisions. We need to tell them about what kinds of things are going to cause them to be at more risk or less risk, what kinds of choices can they make. And then we also need to vary the risks that people are exposed to. So for example, government buildings should have maybe an hour every morning that requires masks, so people who are immunocompromised can still get government services, or maybe one day a week, or something.

Transit, whether masks are required on a bus or not should probably be at least up to the bus driver because if that bus driver is living with someone who's high risk, they shouldn't be forced to work in an environment where they're going to get COVID.

Chris Hayes: Yeah.

Ellie Murray: And so, what we need I think at this point, because we're clearly very firmly committed to the “everyone making their own choices” is that we need a whole range of different risk levels that people can then choose from, and we need to help people understand how to choose from them. And instead, what we're seeing is either everything is really stringent. Masks are required everywhere, things are shut down, or everything is completely open and no one is wearing masks. And people who wear masks are getting dirty looks. And everyone just like pretend it's over. And neither of those extremes is going to be workable.

Chris Hayes: Yeah. And we did a whole podcast with Dr. Linsey Marr about the sort of ventilation part of this, which is a huge part of the story. And some places, I think, have gotten better about that. I think people have been putting in HEPA filters and stuff.

And it was interesting, we bought a ticket for Broadway. Basically, our whole family got it. And I was like, “All right, we're using these antibodies.” Like, we're going to see a Broadway show. I think we're a week out from COVID. Like, our antibodies are very strong right now. Let's go use them. But, obviously, Broadway is a really interesting case because I love the theater, A. B, it’s a huge economic engine for New York City. Shutting it down was brutal. It was also brutal for a lot of actors and dancers, and all these people, right?

So you want to have live theater. Like, you want to be able to have live theater. Live theater requires a bunch of strangers, a thousand of them getting together. And again, I don't have the ability to discern whether this is BS, or not basically, about the air filtration system. Now, this does seem to me a place where like in a Broadway theater, they're making a lot of money on Wicked. Like, they can afford to do what they need in a building like that to really ventilate the hell out of it.

And that just struck me as like, okay, this seems like, I guess, a best case way to go about this, right? Like, we're going to have people come to Broadway. We're putting in ventilation. Again, I know there are different theaters that have done different levels. So I'm not saying what my judgment is about Wicked versus anyone else. It just seems like, in an abstract sense, it struck me as we were preparing to go to this. Like, here's a model for something that seems a little more straightforward and tenable than a lot of the other things I've seen.

Ellie Murray: Ventilation is potentially our best option according to some people. And I don't know how great we can do the ventilation everywhere. And it's certainly going to cost a lot. And it's probably a social engineering project and equivalent to switching to sewage pipes and indoor plumbing.

Chris Hayes: Yeah.

Ellie Murray: So it's not a short-term solution. But how do you know when you walk into a store, or a restaurant, or a Broadway theater, that this place has even assessed its ventilation, let alone upgraded it?

Chris Hayes: Yeah.

Ellie Murray: And so, you can't make a decision. And also, there's no incentive for a business owner or building owner to upgrade their ventilation, if there's no way for you to tell that their building is any different. And so, you don't know what to do with that information. You don't know how to assess if their ventilation is good enough. And until we have public education on what that looks like, what should the ventilation be like? What are we aiming for? Is it there? These kind of pieces of information communicated to people will take away so much of that burden of trying to figure out what to do.

And we can think about it like the way in New York City, we have these like ratings on restaurants, right? If you're like really worried about getting food poisoning, you're not going to go to like a D restaurant. You're going to look for an A restaurant.

Chris Hayes: Yeah.

Ellie Murray: And that tells you that a certain set of criteria has been met recently, and that this place has been found to have good health hygiene. What does that look like for COVID? How do we communicate that?

Chris Hayes: Yeah.

Ellie Murray: What should we consider there? I think those are things that should be on the top of people's list. And then also communicating about when we can expect the potential for COVID season to happen. Does it make sense to have every school in a district go back on the same week, or should we stagger them a little bit so that we don't have this huge mass of kids coming into the infection pool? Should we say, okay, after the Christmas holidays, we're going to do a 14-day quarantine. And schools can only start in-person January 14th or later.

Chris Hayes: Yeah.

Ellie Murray: That could potentially even get rid of the norovirus season for all we know.

Chris Hayes: Totally.

Ellie Murray: Like, we've never tried these things.

Chris Hayes: Right. What if a holiday Christmas vacation is two weeks instead of one, right? And that's just a new thing. We just make part of what, like that we know there's seasonality.

Ellie Murray: Or even if we don't want to shift the school year, maybe the first two years are remote. And that, as a teacher, you can prepare for that.

Chris Hayes: First two weeks is remote.

Ellie Murray: Pardon?

Chris Hayes: First two weeks is remote.

Ellie Murray: First two weeks. Yeah.

Chris Hayes: You said the first two years, and I lost my mind.

Ellie Murray: Sorry. The first two weeks. Yes. Maybe the first two weeks is remote. Those kinds of things, we need to get creative with solutions. We need to try them out and we need to assess what happens. So we need to test COVID. We need to record those rapid test results.

Chris Hayes: Right.

Ellie Murray: We need to know if our solutions work.

Chris Hayes: Well, then to that point, this is also where I do think there's certain ways that we could make these changes if they felt stable and baked in. So it's like, if it's just the fact that this is now the case, the first two weeks back are remote.

Ellie Murray: Right.

Chris Hayes: And we just know that and parents can plan for it, like that is a more tolerable thing to do than like, “Oh, no.”

Ellie Murray: Right.

Chris Hayes: And that's why we keep having like, “Oh, no, we're having another outbreak. It's like, well, no kidding, we're having another outbreak.”

Ellie Murray: Right.

Chris Hayes: Which is the holidays, everybody was indoors with each other, inside in the cold weather. Traveling across country, yes, of course, we're having another outbreak. So it's like, if we do things that have a predictability and regularity to them, and they aren't the full, like we're canceling school, that is just a much more tolerable sort of way of dealing with the, quote, “new normal.”

Ellie Murray: Right. And I would say that kind of looping back to what is the definition of a pandemic? A pandemic is when at any moment, we could be in another crisis.

Chris Hayes: Right.

Ellie Murray: And we're still there.

Ellie Murray: That at any moment, it could be like, “Oh, no, we got to do this again.” Whereas if we get to a point where we have some predictable things, they turn on and off, unpredictable, even if we're saying mass mandates or not, if they're tied to a specific case level, or a specific wastewater virus level, then I can look at the curve and see how much I individually think, okay, are we going to cross that threshold or not? But everything is just as like case by case, individual by individual decision-making process, that no one knows what's coming tomorrow. No one knows what's coming a week from now. And there's just so much uncertainty. It's not good for people. It's not sustainable.

Chris Hayes: Dr. Ellie Murray is an epidemiology assistant professor at Boston University School of Public Health. She runs a Causal Decision Lab there, that's in the department, along with her team. She's a social media editor at the American Journal of Epidemiology. She co-hosts a podcast called “Casual Inference.” Dr. Murray, that was really, really illuminating. Thank you so much.

Ellie Murray: It's great to be here. Thank you for having me.

Chris Hayes: Once again, great thanks to Dr. Ellie Murray. I should note that since we recorded this episode, that she and her colleagues actually published some findings on masking in school, something that there's been lots of study on and a lot of, I think, inconclusive data in the aggregate. It's been hard to sort of get control trials and sort of zoom in on the causal mechanism. But they looked at Massachusetts, which allowed school districts to sort of optionally rescind mask requirements. All the two districts Boston Proper and Chelsea in the Greater Boston area ultimately removed the mass requirements.

And her research looked at what resulted. They found that school districts that removed mask policies had nearly 12,000 cases of COVID and students and staff over the next 15 weeks of school year. She found the effect was largest for staff, and that from February to June 2022, over 40% of the COVID cases among them were attributable to the change in policy. Dr. Murray and her team considered the effect of mandatory isolation and estimate nearly 20,000 excess missed school days in children who attended schools in the districts of study.

As always, we love to hear from you. You can tweet us with the hashtag @WITHpod. Let us know your thoughts on this phase of the pandemic and what's on your mind. Email Be sure to follow us on the ticky-tocky by searching for WITHpod on TikTok.

“Why Is This Happening?” is presented by MSNBC and NBC News, produced by Doni Holloway and Brendan O'Melia, engineered by Bob Mallory, and featuring music by Eddie Cooper. You can see more of our work, including links to things we mentioned here, by going to

Tweet us with the hashtag #WITHpod, email Follow us on TikTok by searching for WITHpod. “Why Is This Happening?” is presented by MSNBC and NBC News, produced by Doni Holloway and Brendan O'Melia, engineered by Bob Mallory and features music by Eddie Cooper. You can see more of our work, including links to things we mentioned here, by going to