Coronavirus (COVID-19): Press Conference with Caroline Buckee, 03/25/20


You’re listening to a press conference from the Harvard T.H. Chan School of Public Health with Caroline Buckee, associate professor of epidemiology and associate director of the Center for Communicable Disease Dynamics. This call was recorded at 11:30 a.m. Eastern Time on Wednesday March 25.

Previous press conferences are linked at the bottom of this transcript.

CAROLINE BUCKEE: Hi. This is Caroline Buckee. I’m the associate director of the Center for Communicable Disease Dynamics and an associate professor of epidemiology. We are working hard on [INAUDIBLE] outbreak in the US, and also [INAUDIBLE]. And I think [INAUDIBLE] just open it up to any questions. My area of expertise is on epidemiology, what we can do in the short, medium, and long term. So I’m happy to answer any questions you may have.

OPERATOR: At this time, we’ll open the floor for questions.

Q: Hi. I was hoping you could talk a little bit about younger people and the infections that they’re getting. Originally, we had the impression that young people were at least somewhat immune. But then the CDC data last week suggested that they were not, at least [INAUDIBLE] if not to death. So, if you could just talk about that a little bit, that would be great.

CAROLINE BUCKEE: OK, great. Yeah, so just to be clear, they’re talking about young adults and not children. Some of the data from China suggest the attack rate seems to be fairly constant across age groups. So, people are getting infected regardless of their age. But of course, the ones that end up in hospital vary, with elderly populations being much more at risk.

I would just caution everybody to remember that, when you see some figure about the fraction of people in ICU beds that are between a certain age range, it’s important to remember how big that age range is. So, if they’re saying, you know, that 20% of ICU beds are people between the ages of 20 and 60, the denominator there is much, much bigger than the denominator in, you know, 65 plus.

So, the absolute risks in young adults remain relatively low, although not zero. So young people still can get severely ill, but I think the main takeaway there is whenever you see a figure like that, you have to try and think about what the denominator is, and what fraction of the population that actually represents.

Q: Thanks. And so young people are not getting abnormally, or getting sick out of proportion to their population, but just that they can get sick.

CAROLINE BUCKEE: They can get sick. I would point you to the best analysis of this from data which was done by Marc Lipsitch at the Center and colleagues. If you want to follow up, I can send you that article. There is still a risk for young people. And remember that the risk is going to depend both on preexisting conditions, but also what’s happening with the epidemiology. So, who is getting infected, and how many infections those severe cases are arising from.

So, because we still don’t understand how many people have no symptoms and how many people have mild symptoms, it’s still quite hard to estimate things like a case fatality rate or what fraction of infections are severe. So, there is still a lot of uncertainty about that. It’s certainly not completely benign in young adults. We have seen cases of very severe illness and death. But we don’t know what to divide that number by to get an accurate estimate.

Q: Thank you very much.

OPERATOR: We’ll take our next question in queue.

Q: Thank you for doing this. I have a two-part question following on the previous. Some universities, including here in Arizona, remain open. And so, my first question is, would a dorm environment be a likely place for the rapid spread of COVID-19, with students returning to campus from around the state and around the country after spring break? And then secondly, where states are not taking extreme stay-at-home measures, such as here in Arizona, would it be wise for schools and universities to take such steps anyway?

CAROLINE BUCKEE: So, to your first question, absolutely dorm environments are places where we expect COVID-19 to transmit efficiently because people are in close proximity. And unless they can be confined to their rooms, they’re sharing bathrooms, they’re sharing kitchens. And so, we think that there’s a lot of opportunity for transmission. So anywhere where there’s a lot of people living in close proximity, we’re at high risk of transmission. So, prison populations, university dorms, and so on.

This issue of spring break is a very serious one. And as we saw in Florida, there are many young people who are not taking any kind of social distancing precautions. And they are now traveling home. So that is a very big risk. And we need to make sure that we are careful. The problem, one of the problems is that, because surveillance is so patchy, because testing is still so weak around the country, we don’t even have a good sense of the highest risk returning travelers.

So clearly, New York is a hotspot, but Florida and other places where people are coming back from, we still don’t know how to even assess what the risks will be. But I think that states like Arizona should absolutely be very worried about their returning spring breakers, and think hard about asking them to stay home for two weeks.

To your second point, yes, in areas where there’s not strong social distancing, some – you know, states vary in terms of their policy, and sometimes it’s in relation to their testing capacity. So, if they’re not testing and they’re not seeing deaths yet in their hospitals, it’s easy to stay complacent. But you have to remember that there’s a time lag here, where transmission happening now is going to end up with hospitalizations in two and three weeks.

So yes, I think social distancing measures should be taken, and schools and colleges should be taking it upon themselves to really try hard to limit dense populations of people that could be spreading the virus. So, I would say that, after attending some spring break, a sensible policy might be to have everyone returning from spring break self-isolate for two weeks in order to avoid exactly what you’re saying, which is, you know, outbreaks caused by returning travelers.

Q: Thank you for that. Appreciate it.


OPERATOR: We’ll take our next question in queue.

Q: Hi. My question is about Imperial College research that suggests that cases might rise after restrictions are lifted. I was wondering if experts believe that now, as we see restrictions being lifted in Wuhan coming up in April, and potentially later in European countries, is it possible that we might have to have more restrictions or more confinement policies afterwards, after a period of lifting restrictions? Thank you.

CAROLINE BUCKEE: Yeah, I think that the Imperial model is correct to be concerned with the possibility of resurgence of the epidemic when we release these social distancing restrictions. In essence, we are increasing the contact rate again, to the point where the disease could spread if there are enough susceptible people in the population. A lot of the impact of releasing social distancing will depend on how many people got the virus and how many people are now immune.

So, it hinges on this issue of if there are lots and lots of asymptomatic people that don’t have any symptoms, but they have had the virus and they have gained sufficient immunity – and these are still things that we don’t know – then that means that, when we lift the restrictions, then it should be OK, because the disease can’t spread because enough people are protected. If, on the other hand, there is a lower fraction of asymptomatic people, and the infection only reached a limited amount of people, and so we don’t have that what’s called herd protection or herd immunity, then the possibility of resurgence is very real.

And just to be clear, you can have local outbreaks happening if you’ve had a community that’s cut off either geographically, because it’s remote and the infection hasn’t reached there, or within a population where you’ve isolated particular communities. So, for example, in many places, we’re isolating nursing homes. We must be very careful when we start relaxing social distancing measures that we make sure that those places remain protected as long as they need to be protected.

One of the only ways that we are going to be able to tell if we’re ready to release social distancing measures is through serological surveys. So that’s when you go and you test people’s blood. And you’re not looking for the virus, as you would with a diagnostic, but rather you’re looking for an immune response that’s specific to the virus. So, then you can say, well, this person has been infected in the past.

At the moment, we still don’t know the relationship between how many antibodies you have in your blood and if you’re actually protected. So not all antibodies are going to be protective against re-infection. So, at the moment, there are still a lot of unknowns about how much immunity people are getting, whether people with very mild symptoms or no symptoms are getting adequate protection. And so, we’re going to need to figure that out. And that’s going to be key to how much we worry about resurgence. Yeah, does that answer the question?

Q: Yes. Thank you very much.


OPERATOR: We’ll take our next question in queue.

Q: Hi. I’m wondering if there’s any information about what makes people susceptible to severe illness, besides an age and having a preexisting condition. Is there anything that a person can do, aside from trying to avoid infection– I mean, in addition to trying to avoid infection, but something you could do that would make you less likely to get really, really sick if you did become infected?

CAROLINE BUCKEE: At the moment, I don’t think that there’s a strong evidence base for that, apart from overall health, age, and comorbidities. I think those are the ones that we have very clear indications on. And you know, on a personal level, I guess staying healthy, going outside, vitamin D, using a humidifier in your house, these kinds of things can be very helpful.

There was an article written by Joe Allen at Harvard with Marc Lipsitch about some of the challenges to how you can protect yourself and loved ones if you have somebody in your household with COVID. And some of those suggestions are generally quite useful, as well. Again, I can send those, if you’d like.

Q: Yes, that would be great. Can I ask a different question? Do you think it’s inevitable right now that– you know, we’re looking at hotspots in various parts of the US. Is it inevitable that the entire country will be affected?

CAROLINE BUCKEE: I think so. If you look at where the disease is spreading, I think it’s almost everywhere, including in places like Puerto Rico and Hawaii. So, places that would be cut off. I do think that there is a possibility, like I was saying before, that very remote communities might avoid it, just by chance, with the travel restrictions. So, in terms of spatial distribution, there might be pockets of people that remain unaffected if they’re very remote. But for the most part, I can’t imagine that there are any major cities where we’re not going to have some level of an epidemic.

Q: Thank you.

OPERATOR: We’ll take our next question in queue.

Q: I wanted to follow up on the previous question. I mean, do we know enough yet to project the trajectory of the epidemic here in the United States? And do we have any idea when it is likely to peak here?

CAROLINE BUCKEE: So, this is a very challenging thing to model. And just to give you an idea, so some of the factors that go into whether you can accurately predict some of these things are things like the asymptomatic ratio, so how many people don’t have any symptoms, how many people have mild symptoms; the structure of households, so how is our society divided into different households; and then, crucially, the contact rate.

So, if you think about the basic reproduction number, which is the average number of secondary cases a new case leads to, we think that’s between 2 and 3. But right now, we’ve shut down society, to some extent, with social distancing. And we don’t really understand how that impacts the contact rate. And the contact rate scales linearly with the reproduction number. So, when you change the contact rate, you change the reproduction number. And so, at the moment, given that things are changing every day with social distancing, it’s very hard to predict what’s going to happen.

Having said that, I think it’s clear that, from a hospital preparedness perspective, they can do a simpler kind of analysis, and think about how exponential growth is going to affect them with varying rates of exponential growth, because in many places, I think that’s what’s going to happen. The model’s output have a lot of uncertain parameters. So, the short answer is predicting the peak is hard, and potentially not actionable in the short term for policymakers. Yeah.

Q: And I wondered if you could talk a little bit about your research into mobility, since that seems to factor into the social distancing equation.

CAROLINE BUCKEE: Yeah, sure. So, my work in the past, and now, we have been working with tech companies to try to – for people that have location-based services on apps and on their devices, the tech companies have access to that data. And we’ve been working for several years on the appropriate privacy protocols and aggregation protocols to make sure that individual privacy is protected, but that we can use it to inform the response here.

So, what we’re doing is the company is aggregating the data up to a spatial scale where we know that people cannot be identified. And then we’re using that to look at things like, say, on a county level, how much mobility is there within the county, on average, how far are people traveling within the county, is there a change in, say, commercial zone travel versus residential travel.

And the idea is that we are able to inform policymakers and update them so that when they put public messaging out there around social distancing, they can see if it’s working. And they can see where it’s working. And then if people start getting fatigue around social distancing, they can also see when it starts to stop working and then change course.

So, we are trying to help state and local policymakers with their decision-making by providing them access to aggregated data sets. And in particular, not just providing them with data, which I think a lot of companies are doing, but provide them with analytic support to make sense of the data, to figure out what it means, and how they can use that information to do a better job with their social distancing messaging.

Q: Very good. And are you able to say anything yet?

CAROLINE BUCKEE: Well, so from the data– we’ve been working with Facebook’s Data for Good team, and what I can say is that you can really see changes happening. In and around the Boston, Cambridge area and in Massachusetts, travel has reduced significantly over the last week. And so, it seems like people here are really taking social distancing seriously. And that’s fantastic, because it means we’re going to save lives in the future.

Q: Very good. Thank you.

This concludes the March 25 press conference.

Press conference with Marc Lipsitch, professor of epidemiology and director of the Center for Communicable Disease Dynamics (March 23, 2020)

Press conference with Marc Lipsitch, professor of epidemiology and director of the Center for Communicable Disease Dynamics (March 20, 2020)

Press conference with Barry Bloom, professor of immunology and infectious diseases and former dean of the School (March 19, 2020)