Coronavirus (COVID-19): Press Conference with Marc Lipsitch, 04/7/20

You’re listening to a press conference from the Harvard T.H. Chan School of Public Health featuring Marc Lipsitch, professor of epidemiology and director of the Center for Communicable Disease Dyanmics. This call was recorded at 11:30 am Eastern Time on Tuesday, April 7.

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


MARC LIPSITCH: Welcome everyone. Thanks for joining. I don’t have too much to say by way of introduction; it seems like the big questions at the moment, as I see them, focus around the extent to which social distancing and other countermeasures are working in this country. Which I can say a little bit about, although I think mostly the jury is still out.

I think there are a number of questions, which I don’t necessarily have a whole lot to add, about the impact in the rest of the world. I’ve been more focused in the US – my attention in last few days, last few weeks, in fact, but that’s clearly a crucial issue.

And I think the question of how to emerge from this crisis, which I wrote about now last week in STAT and the Boston Globe, arguing that we sort of have grabbed a life raft and are trying to figure out what to do to immediately save lives, but the prospect of dry land in the sense of a final end to the pandemic still remains quite far away if indeed most of the population is still uninfected as appears to be the case. But that also remains to be seen.

And that that we need to figure out both short term and long term plans. The last few days, a number of people have proposed in various forums what those plans might look like. And so, I’m happy to talk about those issues as well.

MODERATOR: We have a question.

Q:  Hey, can you hear me? Marc, good to talk to you again. I know we spoke recently. I just want to ask you a very broad question, and just get your thoughts on this. We’re obviously looking at data here at the Florida level. I wanted to ask you, kind of, from a national perspective, what states are doing the best job and maybe the worst job to as far as collecting reliable, accurate information about kind of the scope of the infections in their, you know, jurisdictions?

MARC LIPSITCH: I’m not sure I have the level of detailed knowledge to make a comment on that. Yeah, I don’t, I don’t think I want to get into individual states, despite having –

Q: Sure.

MARC LIPSITCH: – made some comments on your own state in the past. But I think, yeah, I don’t, I don’t want to start making league tables, when I don’t have really full information.

Q: No problem. Could you tell me kind of what information would be ideal for states to collect or what the most meaningful information at this point might be?

MARC LIPSITCH Yeah. I mean, I think as a several week-long term, which feels like long term, a several week from now vision, the importance of serological information is extremely high as a way of figuring out where we are in the epidemic, from as random a sample of people as can be accomplished, and there are various ways of doing that.

One way that Massachusetts is thinking about is with newborn heal sticks, which have antibodies acquired from the mother and therefore can give a sense of infection rates in the mother. I think that’s a promising way.

I think other types of random or pseudo-random samples of the population could be could be valuable as well for serology in order to figure out whether what I said, that a minority or relatively small minority of the population has been infected to date is really true, because I think it’s only based on the best evidence we have so far about viral shedding of people who are virus positive and infection and there are a lot of ways to miss that.

So, I think serologic studies will be really essential and figuring out where we are in the epidemic in each place and will help to give a retrospective sense of sort of how well we controlled transmission as well.

In terms of information that’s available now, I think we’ve all been talking for weeks now about the need for more PCR-based virus testing. And that is certainly an important need for a whole variety of purposes, for surveillance.

It’s challenging to interpret, even if we had very good virus testing and tons of tons of resources to do it. The biggest challenge is finding again a sort of relatively random sample of people to test in order to figure out what proportion of the population is currently infected.

In Iceland, for example, they’ve done it by just inviting people, first sort of through bulk messaging.

Which gets a somewhat random sample of people, but probably mostly maybe over represents people who feel a little bit sick. And then through direct SMS text messages to individual people saying, would you please come and get yourself tested for surveillance purposes.

There are a lot of other approaches to doing that, all of which have their limitations and strengths, but together can help to provide a picture that’s more comprehensive than just who manages to get a positive test through the healthcare system for diagnosis because that’s going to be very much weighted towards the sicker people and towards those who are in areas that have enough testing capacity.

And then the third layer of surveillance is syndrome surveillance, where emergency departments or primary care practices or other types of health care delivery areas report the number of the number and proportion of COVID-like illnesses to a central system. And those are aggregated

And as we’ve discussed, I think, before, the relevance of even influenza-like illness, but maybe COVID- like illness more specifically, to counting cases is becoming increasingly accurate because the other causes of those kinds of symptoms are becoming less common as the season wears on and flu and colds go away.

So, I think syndrome surveillance has an important role to play. So those are the sort of three pillars, as I see it.

Q: Thank you.

MODERATOR: Okay, next question.

Q: My apologies if you written recently about this and I just didn’t see it, it’s really hard to keep up with everything. Do you have a thought about seasonality at this point? There seems to be this emerging notion, you know, on the part of some of the members of the White House Task Force that there is going to be seasonal waning or low and that we’ll get a break over the summer. Do you have thoughts about that?

MARC LIPSITCH: Yeah, I think it remains to be seen. And I don’t think anybody can reliably predict what will happen with seasonality.

My own prediction at the moment is that, and this is based on analyses we’ve done of the seasonal Corona viruses that are currently undergoing peer review or we’re revising to I hope finish peer review soon, but that are described in a preprint from several weeks ago.

So, our analysis is that transmission may well get slower over the summer. But that’s still consistent with increasing number of cases, it would just be a reduced reproduction number. And on top of that, social distancing and potentially case-based interventions as the number of cases, well, just on top of the seasonal changes, social distancing could help bring the reproduction number below one, so synergistically yes, but not seasonal changes alone.

And I think that’s consistent with what we’ve seen from other parts of the world where they are having ongoing transmission, despite either being in the southern hemisphere, like Australia and southern Africa, that are outside the tropics, or being in the tropics and having summer in many ways, summer-like weather but still having transmission.

So, I think it would be somewhere between optimistic and wildly optimistic to think that transmission would go negative, that the reproduction number would go below one from summer weather alone. But it would probably be sensible to think that it might decline a little bit.

Q: Which preprint server was the paper on and did you see Arnold Montose piece over the weekend in JID? I’ll send it to you. He and some people did a study on COVID kinds of viruses and seasonality. And they say there’s a marked pattern of seasonality.

MARC LIPSITCH: Yeah well that’s clearly true. The issue – and that’s what our preprint and hopefully soon publication will also show, does also show and what the CDC has shown in the past – but the issue is that pandemic, as I’ve written in a different piece on our website, a blog post on our website about seasonality a few weeks ago – the decline of infections at the end of the wintertime in normal seasonal viruses, like flu and seasonal corona viruses, is a collaboration between the seasonality and human immunity.

So, as we build up herd immunity because there’s just been a big epidemic, we’ll also see the seasonal ability of transmission go down. And those two factors together can bring down the reproduction number below one. But as with flu, with pandemic flu when you have lots of susceptible hosts, you don’t have that second factor of herd immunity boosting the impact of seasonality.

Q: Great. Thanks very much and thanks for doing this.


MODERATOR: Next question.

Q: I also thank you for doing this. Really appreciate it. Do you have any thoughts on the – I know you said you’re not focusing on international right now, but we’re seeing a lot of scenes from Asia, in particular, about trucks driving around and spring disinfectant, guys walking around with backpacks spraying stuff. Is the disinfecting effort worthwhile or theater do you think? 

MARC LIPSITCH: That’s a really good question. I think there’s a very broad question about what is theater and what is important and it’s going to be really hard to tell because nobody is doing this as a formal experiment.

Everyone is throwing a mix of interventions at the problem, as they should. I mean, this is understandable, but it’s going to be a challenge and probably interesting methodological work for a bunch of us in the future to figure out, try to tease this apart.

But, I think the role of the outdoors in general is open to question. You know, different countries have, for example, restricted outdoor activities to varying degrees, even different states within the US, I think.

And you know, it’s clear that ultraviolet light kills viruses, it’s clear that ventilation dilutes the risk of acquiring a virus from being near someone just from airflow. And on the other hand, people sometimes have very close contact with each other outdoors and do not socially distance when they’re outdoors and even, you know, touch each other or get very close to each other. And maybe that is that that kind of outdoor crowding is dangerous.

So, I don’t really know what the answer is. I mean, I’d personally be surprised if disinfecting outdoor spaces has any impact, but that’s more of a hunch based on first principles than a really scientifically-informed view.

MODERATOR: Okay, next question. Go ahead.

Q:  My question is about the timeline – you kind of laid this out at the top, but I just wanted to ask, what do we kind of need to start to see to potentially consider lifting distancing restrictions? I know that there’s been the AI report that came out that kind of laid out some of those thoughts and I framework, but I’m curious if you agree with that. And if you think that that’s possible by the end of this month, as the President has repeatedly suggested.

MARC LIPSITCH: Thanks. I think the AI report is a good start and I don’t remember exactly what their four criteria were but they were they were sensible when I read them.

One has to do with hospitals having significant open capacity. Another is to do with declines in various indicators over a 14 day period. You know, I think those are both very important points and I think it’s unlikely that that will be true, that even those two will be true for large parts of the country in two weeks.

And some of them might be true because some part of the country hasn’t gotten to that point yet, but I would not yet – well yeah, I guess if the declining incidents would probably not be true in a place that was still on the upswing but low so I think there are going to be a lot of places that are – if not everywhere in the US – that are still not meeting those criteria in the end of this month.

I mean, after all, New York is building field hospitals. So, if they’re anticipating that level of increase, then the idea that they’ll be back to below capacity seems a little bit hard to imagine

So, I think there may be individual places that are starting to see that those kinds of results but maybe not very many.

It’s an important question. I haven’t formulated my own list, but I think that that’s a good starting place and it requires a lot more thought. And also, I think it also requires – I can’t remember if this was one of their criteria – but it also requires that the place that lifts the restrictions be prepared to deal with the consequences of that, in the sense of having case-based detection, case-detection capability and contact tracing capability and those kinds of things.

And given the, you know, that there’s still a huge swab shortage, among other things in the United States, or at least in my state, I just don’t see the idea that most places are going to have plenty of capacity to do testing, much less the other parts of contact tracing.

Q: Appreciate it. Thank you.

MODERATOR: Next question.

Q: Thank you for your time. My question actually is very similar to last one. Would you like to predict when will be the peak in the US?

MARC LIPSITCH: No. In a word, no. I mean, in part because the uncertainties that I talked about biologically before that we – I think in a couple of weeks, there will be enough serologic data, if things go well, to begin to see whether our estimates, of how our estimates of known cases relate to the estimates of total cases that have occurred.

There are some reasons why that could go wrong. But I believe that in a few weeks, we will have a sense of the scale of herd immunity in the population and that will help to remove one of the key uncertainties. But the other uncertainty is the economic and behavioral and political uncertainty, and I’ve never been one to try to make predictions on those fronts, but I think the peak will be, of course, much sooner if we continue restrictions, but it may not, but depending on what we find from the serology, I think the most likely guesses is that it will not be the last week, it will be the first peak.

And that’s sort of what the life raft analogy we made last week was meant to say, that unless there are way more infections in the population than anybody has realized, then we have a long way to go post letting up on restrictions.

Q: How about New York state?

MARC LIPSITCH: Yeah, I don’t even want to predict for there. I think it’s – yeah, I’m going to keep my powder dry on predictions for the moment.

Q: Okay, thank you.

MODERATOR: Next question.

Q: Thank you for doing this as everyone else said. You can always count on me to ask the local question. So are you following the Massachusetts numbers. Some people think there are some inklings of a bit of flattening. Are you seeing that at all? And it sounds also like the Governor is using projections that are lower than yours at least were. So how would you describe the state picture?

And then one side question just out of pure curiosity. I’m hearing from several people that they swear they were shuttered away for three weeks and even so, they’re coming down with the virus. Could it be that the incubation period is longer than we think? Thank you.

MARC LIPSITCH: Yeah. You’re making me feel guilty. There are so many kinds of data that I’m not seeing, including some very important kinds. So, I don’t want to speak about data that I haven’t seen yet, but I think you’ve given me an agenda for the afternoon. Not that I don’t have three other agendas for the afternoon.

So, I won’t comment on that. I think the in terms of the infections in the people who say they’ve been shut it away. I think there are incubation periods, probably for every disease, that are longer than, than the typical – that just means that there’s a distribution.

And it probably depends on the dose that you’re exposed to, perhaps, and on the route of exposure and on your own biology and all sorts of other things. So, you know, I’ve tried to emphasize that whether something is possible and whether something is common are two different questions, and

public health policy that’s based on a 14 day incubation period can still be justified if there are a few percent who have a longer one because there’s no maximum, you know, we’ll never know the maximum of the distribution, only the upper 90 or 95%, or some cut off like that.

So, I, you know, when people say that my questions would be, what does shut away in their house mean, does that mean really never left it? Does that mean they weren’t exposed to a family member who was infected, which they probably couldn’t be because the family member might have been asymptomatic.

So, I think we’ll probably never know for individual cases but there are certainly ways that you could have someone who literally had never not left the house and not knowingly been exposed to a sick person but still didn’t have a long incubation period. It was just some other type of exposure.

Q: Thank you.

MODERATOR: Next question.

Q: Hi, yes. I have two questions actually. First one. Do you have any information on how is the spread doing within American prisons and then did have any recommendations on how to prevent that spread in the prisons?

MARC LIPSITCH: Yeah, I mean, I’ve just read stories about it. I haven’t studied it formally but prisons have all the, I mean, as you know from tuberculosis in Russian prisons, prisons are great places to spread infections, respiratory infections. There’s crowding, there’s poor ventilation, there’s usually shared facilities of all sorts, so it’s only to be expected that presence will be hard hit by this infection and certainly there are stories about happening.

I don’t really see ways of reducing that other than improving the ventilation and reducing the number of prisoners and their density and their sharing of cells and all that kind of thing. So, it’s the same kinds of measures we would take somewhere else, it’s just the purpose of prisons is to do quite the opposite of what we need to do for social distancing, so it’s tricky.

And, you know, in the United States, we certainly have a lot more people in prison than need to be in prison. So, there is one very simple solution which is to partially empty them but short of that fairly obvious point, I think it’s going to be a problem in prisons everywhere.

Q: Thank you so much. And second question is, well, I saw Miami Herald. Maybe they asked already but what is the situation in Florida, because everybody was expecting Florida to be one of the hottest spots because there are obviously a lot of, you know, older people here and tourists, but seems like it’s doing pretty good. It’s not at least in like the first fifth states. Can you comment on for the situation?

MARC LIPSITCH: Yeah, I think Florida is there clearly important local hotspots, in Florida around Miami and surrounding areas. And it seems like other parts of the state are not as hard hit but southern Florida seems to be quite hard hit, I think.

My own suspicion, although from talking to reporters in Florida, I’m not sure if the suspicion is correct, but my suspicion is that there’s, as in other places, a lot of unnoticed transmission because of limited testing and certainly the governor has been strongly minimizing the extent of the problem in Florida, so I’m curious whether the seemingly low levels in some parts of the state are real or from lack of testing.

Q: Thank you so much.

MODERATOR: Next question.

Q: Professor Lipsitch, thank you. Hello. Thank you for this talk. I’d like to ask you, according to your opinion, do former patients develop immunity, or they might get sick again? And also, if you have any fear that the virus mutations might reload the threats in the near future?

MARC LIPSITCH: So, it’s not my opinion that we don’t know, it’s just a fact that we don’t know what immunity is like to this coronavirus in terms of its effectiveness and whether it’s completely protective, partially protective, or not very protective, and how long it lasts. There just aren’t studies yet. That is a very large question on everyone’s mind for all sorts of different reasons. It would affect how soon the epidemic might come to an end, it would affect whether we can test people for antibodies and send them back to work, etc.

So, there are a lot of studies being designed at the moment to try to figure that out. We need serologic tests before we can do that and we need specifically quantitative serologic tests that give you a level of antibodies, not just a yes/no answer.

So, the clear answer at this point is we don’t know. The indication from other corona viruses is that previous infection is partially protective even a year later.

I don’t know if there have been studies of very short term, less than a year, but there are been challenge studies one year later, where individuals who had previously been deliberately challenged with a coronavirus not so closely related to this one but somewhat related to this one, were re-challenged year later and they had very mild infections, if any symptoms at all. And they also had very attenuated shedding of virus. So, that’s not complete immunity, but it might actually be the best possible thing if you sort of keep getting boosted but don’t get sick and don’t shed very much virus.

We also know from SARS, the original SARS, the SARS in 2003, that it seems like immunity was a bit longer lived than that, maybe several years. And so, all we can do is reason from analogies.

In terms of the question about mutation, the virus mutates. That’s why you can build trees of its descent and trace that it moved from Washington to some other place and the Iceland study has also looked at geographic spread. The way you do that is by seeing which strains share mutations.

So that’s not in question. The question is whether those mutations will lead to escape from our immune systems or escape from a vaccine, if there is one.

And on that I’m guided by Trevor Bedford at the Fred Hutch Cancer Research Institute, who’s been studying this question and his opinion is that those changes will be not very fast and not very likely, so that we should have it – it shouldn’t immediately escape immunity.

We know with flu that potentially can, with the flu viruses can escape immunity. As far as I’m aware, that is not the main challenge in maintaining immunity with coronavirus. Rather it’s just the virus, I mean, that the immune response itself is not that effective

So, I think it all remains to be seen. But that’s not at the top of my list of worries at this point.

Q: Thank you very much, professor.

MODERATOR: Next question.

Q: Hi Professor Lipsitch, thank you so much for taking the time to do this today. I wanted to ask, the Kaiser Family Foundation today published a survey showing that most Americans feel the CDC or the federal government should be responsible for leading the response to coronavirus, but the same poll found that most Americans think their state governments are leading the response. And it just kind of struck me that the CDC seems to be a little absent. And I don’t know if you’ll comment on that, but if you would comment, hopefully, on whether it’s important to have a federal agency like the CDC leading the response and, if so, why? I don’t know if this hampers exchange of high-quality information about jurisdictions?

MARC LIPSITCH: Yeah, I think it’s absolutely essential. And I have colleagues who work in state and local public health who have expressed frustration, to put it politely, that they can’t get clear information from the CDC about decisions they need to make.

And, you know, there are many decisions that need to be made locally, but the data and the information about whether a particular intervention is effective, or how to structure it, or how to do surveillance, all of that is best provided from a central authority because the decision may rest on local criteria but the data are often things, as you point out that have to be shared between jurisdictions, because not everyone can study every question.

So, central information is essential and is not being, in my experience, provided effectively from the federal government on certain issues. I think it is on some others, but there’s definitely frustration about the information flow.

There’s also frustration about the flow of goods and the stockpile and the ability to get hold of everything from testing materials to PPE, to other items, ventilators that that are needed. And there’s been a lot pointed out already about the fact that if we leave it to states bidding against each other, that’s a massive market inefficiency where all the surplus goes to the companies and all the burden is borne by the taxpayers. And that’s another area where federal leadership could be important and has been missing.

So, I think, on at least those two issues, and then thirdly, and probably actually most importantly on strategy and communication that the CDC traditionally has been very communicative during, for example, the 2009 pandemic, having daily briefings from public health professionals.

And that is not happening in this pandemic and it’s a much worse pandemic, so it’s even more needed. So, I’m not sure that the CDC is fully able to make its own rules at this point. In fact, I’m pretty sure they’re not, but I think it’s sorely missed, the sort of confidence and sense that somebody understands at a high level, what’s going on, that is projected by this top CDC career official making those statements every day.

So, I’m sorry to I think that lack is harming the response.

MODERATOR: Do you have any final thoughts you’d like to share with everyone?

MARC LIPSITCH: No, well, I guess I might share the thought, which I’m in the process of writing up with some colleagues that some of the more invasive types of solutions that are being proposed to get us out of the distancing and into a period of contact tracing and the like – I find very alarming the article in this morning’s New York Times advocating involuntary out of home quarantine. It struck me as really unjustified on a public health grounds and also quite worrisome on grounds of human rights and the amount of suffering, it would cause.

And as I say few of us are on the process of writing about that, but I think I would just point out that the justifications used in that article by people I admire and respect don’t hold up, in my opinion. Specifically, they cite the cases of Korea and Singapore to justify involuntary out of home quarantine. And Korea and Singapore have used involuntary quarantine but mainly in the home, which is a very different thing.

They assert that the home is a hotbed of transmission, which I don’t – I’m not sure what evidence they’re citing. But my perception is that in cases where there’s extreme social distancing, the home becomes the only locus of transmission because other opportunities for transmission are reduced. But that doesn’t mean that it’s an intense problem, rather that it’s what’s left of the problem.

So, I will elaborate on that more later, but I just wanted to point that out.

This concludes the April 7 press conference.

Benjamin Sommers, professor of health policy and economics (April 6, 2020)

Michael Mina, assistant professor of epidemiology (April 3, 2020)

Joseph Allen, assistant professor of exposure assessment and director of the Healthy Buildings Program (April 2, 2020)