Coronavirus (COVID-19): Press Conference with Michael Mina, 06/22/20


You’re listening to a press conference from the Harvard T.H. Chan School of Public Health with Michael Mina, assistant professor of epidemiology and a faculty member in the Center for Communicable Disease Dynamics. This call was recorded at 11:00 am Eastern Time on Monday, June 22.

MODERATOR: Dr. Mina, you have any opening remarks?

MICHAEL MINA: No, I’ll take questions.

MODERATOR: All right. First question.

Q: Hi, can you hear me? Thanks for doing this. I just want to ask it looks like in Europe, a lot of even the countries there that were really hard hit have seemed to have gotten a lot better. Their new cases per day have really fallen. But in the United States, you know, we still have a lot of new cases per day and it’s even maybe going back up now in the U.S.. So I wonder just kind of why you think that divergence happened? Is it just that Europe had stricter lockdowns than we did and stayed in the lockdown longer? Or is there some other explanation?

MICHAEL MINA: I think that a lot of Europe did get cases down much lower than the United States did in many of our states. And so we’re now seeing rising cases, particularly in the states, that really never got cases to a level that I would consider sort of sufficient for really opening back up. For example, in Massachusetts, we would like to see cases get down to about two per hundred thousand before we really have sort of, as an ideal location and as we’re opening back up.

And so I think that that’s essentially what we’re seeing is people sort of preempting this and and opening up maybe too early in the United States far before they maybe should if the goal is to to prevent cases from spreading in the communities. We are very heterogeneous as a country and across our states. Perhaps we’re actually more heterogeneous than across countries in Europe. And we’re seeing different states taking into consideration and balancing the costs versus benefits of staying closed from an infectious disease versus economic perspective very differently. And I think that’s really what’s driving the cases in the US we’re seeing in general. I think to a certain extent, well, not even to a certain extent, quite literally, where I think we’re seeing apathy about caring about this virus in many states.

Q: Yeah. That makes sense. Do you think Europe is gonna be able to, I mean, to stay low? Do they need to be able to have, you know, that really good contact tracing and testing in place? I mean, do you feel like they have that or is the natural thing that now that they have let up their lockdown’s very expected to go back up again?

MICHAEL MINA: I think it’s a great question. Different countries are certain have different levels of surveillance in place. And the same with the different states here do. And I think we’re seeing warning signs from across the world at the moment that there are potentially second waves that this virus, as places have opened up in Beijing, for example. Places that seem to have had cases very well under control are now seeing increasing cases again. This speaks to the transmissibility of this virus, just how difficult it is to control and contain without pretty serious surveillance efforts in place as places open up, whether it’s in Europe, Asia or the United States or elsewhere. I think we will have a very difficult time stopping continued spread. And this is all in the context, too, of it being the summer time and being, you know, a lull for seasonal viruses. And so it is, again, very concerning for the fall.

Q: Thanks.

MODERATOR: Next question.

Q: I recall last week you mentioned masking and you, if I remember right, you said something to the effect of if we had universal masking, we might be able to contain it just with that one step. I wonder if there’s any research going on about masking specifically whether you do feel that way and just kind of as we open up and talking about rising cases, whether universal masking might be an answer.

MICHAEL MINA: Yeah, I think that there’s research being done. It’s it’s pretty difficult research to do it to a certain extent because you essentially have to be making some assumptions on what transmission would be. You have to either have really controlled groups, which I would say that having a randomized control study of mask wearing probably is not appropriate. We wouldn’t want to ask a group of people to not wear masks. So this makes it difficult to get a very accurate estimates of the true efficacy of masks. We can use observational studies where people are not wearing masks and try to understand what the contacts were and what the transmission levels looked like.

But some more sort of laboratory based studies do suggest that mask wearing should really cut down significantly on spread. And I do think that alone, if everyone – let’s take an extreme example. If everybody wore masks all the time whenever they were around other people. I do think that that alone could have the potential to cut R0 below one and keep this virus largely under control. It won’t cause the virus to go extinct. It won’t cause the epidemics to completely burn out. But it will it would probably greatly reduce transmission.

That said, I think what we’re seeing is – now these are just media pictures and things like that, but when I look across the pictures from various public events lately, it’s disheartening to see the numbers of individuals not wearing masks at some events or some parties and things like that and so I worry that the populace might not take to wearing masks diligently.

Q: I don’t know if you want to comment on the politicization of mask wearing.

MICHAEL MINA: It’s something that should never be politicized, it’s clearly become a political issue. And that’s just, it’s just a shame. I mean, I don’t mind just saying bluntly, I think that it’s being politicized because our president has generally criticized the need for masks for months now, including not wearing them and, you know, in hospitals – or maybe it was the vice president, I forget which one, but you know that this is something that will have true life and death ramifications, in particular when you’re when you’re putting thousands of people together. If the general theme is to not wear masks, I just think it’s extraordinarily disappointing and dangerous.

Q: Thank you.

MODERATOR: Next question.

Q: Hi, thanks for the availability. So there was a Nature study out last week that showed that a sizable number of people are not mounting an antibody response. It was 20 percent. I’ve talked to a number of sort of really sick patients and doctors who say, you know, they or their patients tested positive for COVID in March or April through a PCR test or based on symptoms, and they’ve had multiple negative antibody tests. They’re using some of the more reliable tests. So what do you think is going on there? I mean, are some people not producing antibodies? Are they just clearing them up quickly? And is there any connection between these people who are still so sick like three months later and not producing antibodies, does that make them more susceptible to that?

MICHAEL MINA: Well, it’s a great question. So in general, we know that there is a sizable fraction of people who don’t produce antibodies to any given pathogen at the time, even measles. We see against a measles vaccine, which is one of the most efficacious vaccines we’ve ever developed, people don’t always have good antibody responses, including people who have gotten measles. So with that in mind, this is actually not extraordinarily uncommon. We’re just looking for it now.

In general, the world has not been looking for antibodies against pathogens during the first time that somebody would see them. When we look at birth cohort, so little children are a good example of, are a good use case because they represent individuals who are getting infected for the first time to a pathogen. So in their personal body, every infection early in life is kind of like a novel pathogen. And we know for them that we see huge bursts of antibodies after infection. And then in a large fraction of children, they wane very quickly. And what we know is that antibodies, every time you get infected, your B cells become a plasmablasts, they pump out lots of antibodies and they generally they exist in the periphery, in your bloodstream. And then they they start to atrophy or apoptos and go away. And so you kind of have this big burst of antibodies and contracts in a very small fraction of those antibody. Those antibody producing cells will migrate into the bone marrow. It’s the bone marrow that really keeps the long lived antibody response for years.

And so what we know is that it’s very likely that you need repeated exposures to drive good, strong antibody responses that are lasting. This is the same – this isn’t surprising for this virus. I think maybe some people are surprised because it hasn’t been well explored. But if we think of this, just like we think of vaccines or other infections, we know that we have to boost people with vaccines multiple times and other viruses that we’re used to testing antibodies for, by the time somebody is an adult, they might have seen some of these antibodies, some of these viruses, 50 or 100 times or more. And so at that point, they might have had built up really good, strong antibody responses over a lifetime.

So I think what we’re seeing is probably a reflection of normal biology, where when people get infected, once they have a very robust, strong expansion of their plasma cells, those cells produce the antibodies and then they undergo the expected contraction. And a small fraction of them will persist to keep creating antibodies. And we have to maybe boost that multiple times in order to get a very durable, high level antibody response that persists.

This, of course, could be problematic for vaccines to a novel virus. But we also have to recognize that antibody responses, especially the responses that we are that we’re measuring in peacetime, are months after an infection, for example, are not the whole story. If you have a large contraction of the cells that are producing the antibodies, maybe your antibody levels do decrease substantially, maybe even two below detectable levels for some people. But it doesn’t mean that the immune cells that we’re trained to recognize to produce those antibodies initially are not still sitting there quiescent and waiting to expand again.

And what I mean by that is when somebody gets reexport goes to a pathogen, they get an animistic response, a secondary response that allows the cells that are already pre-formed, even if they’re in very small numbers, to expand rapidly from two cells to a billion cells. And so it doesn’t mean that you won’t be protected. You’ve already introduced to your body that pathogen.

You’ve already introduced that immune memory. And so the antibodies that are sitting around in a quiescent state are not necessarily reflective of your body’s ability to protect even just – that’s saying nothing about the T cell side of things.

Q: Thanks. And is there any evidence that there’s a connection between these people who are still sick like three months later, I think that they refer to themselves as long-haulers, them not producing antibodies? Is there something about being sick longer that results in that or no?

MICHAEL MINA: So I don’t think we have enough data at this point to really pin it on that. It could be. We know, for example, in all these measles, again, that’s what I asked and study the most, that T cells are really the necessary cellular population to drive measles out of a body, you know, to really clear the infection fully, even if somebody has a good antibody response.

If you have no t cell response, you can have trouble clearing it. So the same thing might be happening here. I wouldn’t pin it necessarily on just a low antibody response detectable in either neutralizing tests or ELISAs. It could, but I think we need a lot more science to understand it better.

MODERATOR: Next question.

Q: Hi, thanks for taking my question. I’m wondering if you can speak sort of generally to the state of serological testing both across the US and around the world. What are kind of the ranges of percentages of populations that have been infected? And are there any clear associations between sort of the the rigorousness of a government’s response and the proportion of cases that they have?

MICHAEL MINA: Well, we can look at Sweden as a good example, and that was a country that right or wrong, decided to try to take a different approach. It generally hasn’t really worked out well, but they’ve decided not to really close down their country and the economy as our neighboring countries did. And we saw a greater, greater population numbers get infected and many more hospitalizations and deaths than might be expected based on the rates in the neighboring countries. Surprisingly, the serious positivity wasn’t immensely, you know, it wasn’t orders of magnitude higher than the neighboring countries. And that, of course, is worrying.

It suggests that it just takes a few extra percentage points of people getting infected to really overwhelm the system, a health care system, and and kill a lot of people. And that’s essentially what we’ve seen in the United States as well. We have had – New York is a bit of an exception. Well, it’s a large exception in the US. But if you look in Massachusetts, we didn’t have our hospitals overrun, but we definitely had pressure. There was a lot of pressure on the hospitals and it was pushing the ceiling a bit. And despite that, we still ended up with probably less than five percent overall seropositivity in our state. So that suggests that if you were to add, if you were to double that and go to 10 percent seropositivity, you’re still not anywhere near herd immunity. But we would have potentially really thrown our health care system over the edge had that happened.

And so that’s just to say – so, I started going down that road because Sweden was one example that didn’t have the same kind of closures and they actually didn’t have an extraordinarily large number of cases. I think in general, if we look across the country, across the world, I think we’d be very surprised if we saw many places that are over 10 percent or so that are significantly over 10 percent. Most of the world I bet is still below 10 percent seropositive. I do think that closing did help. But maybe, you know, it helps to keep the outbreaks under control. Again, this virus hasn’t spread for very long still; it’s only been months, not years. And so the fact that we even have 10 percent with closure suggests that this can spread very, very widely. So I think we can anticipate that the closures have had a significant role in reducing spread.

And we have some research right now looking at the effectiveness of non pharmaceutical interventions, like closing down. And we have seen a decreasing R, meaning the transmission of the virus has generally reduced, has sort of gone down as states have closed down more and more. And of course, now we’re seeing other states are opening up, we’re seeing spread and the effective R increase again. So I think this is pretty convincing evidence overall that closures do have a beneficial effect to slow spread, that these responses being pushed by governments in general have been useful to stop spread. Of course, it always needs to be balanced with the economic hardships that might come from it, but from an infectious disease, public health perspective, I think they are effective.

Q: And so given that reopening is sort of happening to varying degrees across the country and overall, many parts of the country seem to still not be ready in terms of contact tracing and testing systems, I guess. What does the fact that only five to 10 percent of the population is infected or has been infected sort of mean for the future of the epidemic or the pandemic over the next few months and into the fall?

MICHAEL MINA: I would say that it doesn’t bode well. We I can’t imagine, at least I don’t think that our country any way. It’s hard for me to speak to other countries that I don’t know the social structure as well, but in our country, the U.S., I think will probably not make the decision to close down again in the same way that it did. The hope during that closure was to really buy time to build these these surveillance systems, build testing capacity. I think, by and large, that has happened to an extent more so in some places than others.

I don’t think any place has the type of surveillance capacity that I believe is truly needed. I think that should include serology during sort of, you know, serology as baseline surveillance on an ongoing basis. It’s cheap and easy to collect and gives you a whole history and trajectory of the epidemic. So it’s a very powerful public health tool. And then virology as a surveillance mode during in the midst of an outbreak, when you’re trying to control an outbreak in a setting, in an institution and business, whatever might be virology becomes very, very important as well. So I think having, we generally haven’t seen the systems. There’s very few examples that I can think of where there’s actually capacity to do that at scale and this type of scale we might need come the fall to control epidemics when and if and when they start.

So I think, you know, we continue to not put the money where we really need to, I think, to make these things work. We continue to sort of neglect public health. And public health has been neglected for many decades in this country and we still are neglecting it, oddly. All the laboratories are still privatized, for-profit laboratories for the most part. And there hasn’t been a tremendous effort to really build up infrastructure in the way that we need to develop a true surveillance system that can actually stop outbreaks.

So I think that come the fall, we are still going to be in a pretty bad position. And it might be worse because in general, society isn’t going to want to close down again. And frankly, I don’t think society can close down to the same extent without enormous subsidies and trillions of dollars of appropriations from Congress to go out to the general public to keep their storefronts open and food on the table. So that all being said, I think we’ve shown that we have very little capacity to save and protect the elderly and vulnerable even during periods of shutdown. And I think none of it bodes well for the fall.

Q: Thank you.

MODERATOR: Next question.

Q: What are the odds if you step out your front door into the open air, leaving your house or your apartment building, what are your odds of getting a serious infection walking down the street? Put it in terms of numbers per thousand. One per thousand?

MICHAEL MINA: You mean per thousand walks out of my front door? I would say – that’s a very difficult question.

Q: That’s why I’m asking it.

MICHAEL MINA: Well, I think where I live, there are very few people walking. I have a great quiet road here, and I think I could probably walk out probably 10000 times and not get an infection. If I was in New York City, I might say, you know, on Broadway or something, I might say something very different.

Q: I’ve got a follow up question. So I have an office up here in Lowell where I do my science writing and my science editing and I look out the window and Lowell is not the best educated town. And I see a lot of really old people, they look to be about 60 to 65, carrying their groceries home from the Market Basket by hand because they don’t have cars and they’re wearing these huge black masks. And from what I understand, the odds of actually getting infected in the open air are very slim. And I want to go down and talk to these people and say, give yourself a break. Your odds of actually being hurt walking home from the supermarket are close to zero. Would you do that to them if you watch these old people walking with their groceries home while wearing masks? Would you intervene and say, really, don’t do that?

MICHAEL MINA: No. No, I think masks are are not harmful in general, I think that they are useful without being, you know, it’s not like a medicine or a you know, it’s something that can have severe side effects. I think that in general, you know, if somebody is if it’s know, extraordinarily hot out and somebody is in the middle of an open park with nobody around them, I would certainly say, you know, you can take off your mask, relax for a little bit. But certainly if you’re walking back from Market Basket, it’s not just for protecting that individual. I think it’s also, there are a lot of individuals who nobody knows where that individual has been and they don’t know where the other individuals that are walking past them have been. And so I think, you know, as long as you’re in a populated area and walking next to people or across from people, I do think that you may as well wear a mask at this point in time, in particular if you are older. Well, no, I would say just anyone. But if you’re worried about your own safety, certainly individuals who are older should maybe be paying attention even more.

Q: One final question, then I’ll let him go. What do you think of the psychological effect of a society in which almost everybody in the open air is masked and cannot see each other’s faces and cannot have each other’s faces seen? Is anybody doing research on that, sir?

MICHAEL MINA: So, it gets out of my comfort zone scientifically but I’ve thought about it. And I think it’s interesting because there are places in this world that have more of a history of of wearing masks on a regular basis, East Asia in particular. And. But it’s certainly something that warrants study and, you know, if this is going to be a long term thing of wearing masks, you know, for years and, you know, I don’t really see that as the case. But it could have psychological effects. I mean, our facial expressions are extraordinarily important to how we how we socialize and the type of signals we get. And in particular, we’re not used to wearing masks in society that we’re used to interpreting. So I think it’s important research to be done. It’s very interesting question.

Q: You don’t know who’s doing it, though? You don’t know anybody who is doing that?

MICHAEL MINA: I don’t know anyone working on the psychology of mask wearing, no, but it it really is fascinating. I hadn’t thought about it too much in that in that sense. And it’s really an interesting question.

Q: Well done. Thank you.

MODERATOR: Next question.

Q: Hey, Michael. Just a question on the data. I just wonder how reliable any of these data are on the prevalence. And by that I mean like the Hopkins data are completely meaningless because you’re not adjusted for testing levels, right? And I say that as a Hopkins grad. And also like, look at Florida, you’re just completely non random samples, right? They just take whatever comes in. That’s the lamp post effect from statistics, right? You lose your keys on the street, but you look under the lamp in the park for them because that’s where the light is.

And then also the Time series is not non random in the same way twice. It would be weird if it were. And just one final thing on Florida. They have this footnote. If a person is positive and negative in the same day, only the positive is counted. So, you know, you can multiply that across all the states. As journalists or policy people or market people, like what data should we look at to really understand the prevalence? I don’t think Florida is meaningful at all for the reasons I just stated. And also the Hopkins data. I don’t know. What do you think?

MICHAEL MINA: You know, I think that viral data, viral testing data is from a true prevalence, historical prevalence perspective, it’s extraordinarily difficult to make any sense of it. We know that testing has been off. We can look at, you know, as we start to get better ideas of case fatality, of infection fatality rates in the elderly, for example, or across the populace, we could start using hospitalizations and case fatality to sort of back calculate the expected number of true infections that might have led to that, to what we can detect. And this is sort of ascertainment bias and sampling bias and and reporting bias all wrapped up with this epidemic. And I do think it’s become very, very difficult. I think the only real way to understand prevalence and historical prevalence and epidemic trajectory is to do very high quality representative sampling for serological testing.

And of course, as the first question today or second question pointed out, that might also not be perfect, but at least we can adjust for biological features if we know, for example, that eight percent of people who get infected. This is more random number made up. Eight percent of people who get infected don’t develop an antibody responses that are detectable, we can account for that. But I do think we need to be paying very close attention to how we’re doing the sampling, what type of representative sampling we’re doing and using using these tools to really try to make the best inference on prevalence and epidemic trajectories that we can actually devise risk maps and understand where what populations are at greatest risk, which aren’t, you know, which ones are showing up in the hospital at greater numbers because there’s actually in more cases versus because the co-morbidities in that community are higher. So these all speak to the need to do well representative sampling. Very similar to how we understand that political campaigns need to sample representatively if they want to get an unbiased opinion.

Q: All right. OK. And do you know of a website that’s actually doing that? Like a real random, ongoing random study that we could actually look at to understand what’s actually happening or it just doesn’t exist? And then just one other quick question.

MICHAEL MINA: Yeah. So it doesn’t exist at the moment. We are starting – we just had a press release last week that we’re starting a study to do just this. And I’m a P.I. on it and I’m partnering with some economists and a clinical trials company called TrialSpark. And we’re all partnering to bring these different facets in – logistics, economic, sort of political theory, to create these representative samples and then me on the immunology and epidemiology side to do just that. But it’s expensive to build them. And we’re just getting it started now.

Q: It’s going to be proprietary or public?

MICHAEL MINA: No, once we start getting the data in, it will be public.

Q: Okay, that’s great. And just, Michael, just real quick and I’m sorry, Nicole, I just had one of the question. You mentioned something I think I’ve heard you say that what really matters is the T cell level for immunity. So the question is, why aren’t we then testing for T cell and not antibody? Wouldn’t that make a lot of sense?

MICHAEL MINA: Just to be clear from earlier on, I was saying that that is essentially something that can matter. It doesn’t necessarily indicate more or less. What we’ve seen in measles is that T cells are really what’s required to finally clear the virus from somebody. But I would say that both are important. Antibodies are just a lot of a lot easier and cheaper to get. If you want to test the popular or T cells, you have to do a full phlebotomy. It’s very expensive per person, whereas antibodies very cheap.

Q: OK. Got it. Thanks very much. Thanks for your patience. By the way, please put those data up as soon as you can’t, because I’d love to be able to follow that. Thank you. Thanks, everybody for your patience.

MODERATOR: Next question.

Q: Hi, guys. Thank you very much for doing this, appreciate it. Doctor, I know you’ve talked about this a little bit at the beginning and throughout, but can you just comment generally on the overall rise in cases in the US and where you see the issues and what you see as the main driver behind this?

MICHAEL MINA: I think the main driver is apathy towards the virus and and there’s a whole, there are a lot of people who did not necessarily see the damage done by the virus when it was here – I mean, it never left – in their immediate environment and eventually have just gone to the point where they’re saying it doesn’t really matter very much. And this is, I’ve been worried about this the whole time. I remember saying to some some senators earlier on, back in March or April, they called me up and said, you know, what should what can our offices do at the national level? Things like that. And one of the things I said was to use their platforms and their voices to make it abundantly clear to the populace that the absence of virus in their community doesn’t mean that this is not a dangerous virus. It doesn’t mean that the virus is potentially spreading there, that the absence, when they close everything down, the absence of the virus means that they’re succeeding at keeping it at bay.

I don’t think that message has gotten well across to people, in particular in certain places in this country. And I unfortunately think that now as a result, we see a lot of people saying, oh, you know, the virus, you know, there’s been a lot of media about it but the virus isn’t important. And that is maybe leading a lot of people to, you know, people just get tired of social distancing, rightly so. And I think that we’re seeing people sort of swing the pendulum. They’re going out in full force and going to restaurants in some places, not wearing masks, sort of rebelling against the idea even, which I think, unfortunately, comes back to a different question where it has been a little bit politicized or a lot politicized.

And so I think that that’s a major reason. A lot of people, you know, unless you’re working in the hospitals or you have a family member who is sick, you’re not going to see the ramifications of this virus. And that is always the plight of public health. The more you do to protect the public, the more they don’t realize that you exist protecting the public. And that includes just policies. So if you do a lot to social distance everyone and put really clear policies in place that will stop transmission, then people will start to think that it was an overblown reaction and they’ll swing the pendulum the other way. It’s sort of normal behavior. We see it with vaccine refusals now and many other items in public health. And it’s a common predicament that we get ourselves in. The more successful you are, the more people don’t recognize how important policies are.

Q: I think you mentioned this earlier, but just what is your fear for the fall then, given the situation as it exists now?

MICHAEL MINA: My fear is that we’re going to have uncontrolled epidemics that explode and overwhelm health systems, frankly. My fear is that we can have a lot of states working very diligently to get cases to very low levels to a point where people might feel safe traveling again with all the precautions taken, but that there’s going to be states, there’s going to be such large outbreaks and in many states that no airport will be safe to travel from. You know, those are my fears. I just had – my brother called me yesterday and asked, should he plan to fly to my dad’s house in a few months. And the short answer was, well, the area that he might be flying from might have very low transmission rates, but he will have to transfer through an airport that undoubtedly will have a mixture of people from many states, in particular, potentially from the states that are taking this less seriously and where cases are growing. They might be the people more apt to be flying.

And so I think that, you know, the problem with infectious diseases is they neglect borders. And so, you know, as long as outbreaks are spreading anywhere, that nobody is really safe. And in particular, within a given country, they spread very easily because it’s even more limited borders and things like that. So that’s my concern, is that the more we see cases, the longer it will be before life can sort of go back to some semblance of normal. And I worry that that tens of thousands or hundreds of thousands of people will continue dying.

Q: Thank you very much for that. Appreciate it.

MODERATOR: Next question.

Q: Hey, Dr. Mina, thank you for taking questions. I wanted to ask you about a testing technique that’s getting a bit of attention. Scott Gottlieb and a few others started highlighting it as we move into the fall. Pooling test results. So the idea that, you know, you could take samples from 50 employees and sort of mix them all together. And if you get a positive sign then and only then do you have to do individual testing. How ready is that technique? Are there any technical or, you know, statistical challenges with it? Would just like to get your your thoughts on, you know, if that’s a really solid approach that we’re going to be seeing.

MICHAEL MINA: Yeah, so we published a paper on this pretty recently. Or it’s in an archive. It’s in medRxiv right now and it’s under review at a journal. We’re looking at this statistically, evaluating the role of pooling, what’s the benefit? What’s the cost? What are the logistical challenges? It’s a very powerful technique. I suggest that as many laboratories that can do it think about doing it. There are some – but there are logistical challenges that will cause many or most or essentially practically all labs from doing it.

The number one thing that people get concerned about, which I actually think is not as important, is that as you start to pool samples together, you dilute them out and you might lose sensitivity. The reality is, though, as a screening mechanism, if it means that you can get through many, many more tests, then that subtle loss in sensitivity, which might actually be negligible, depending on how many samples you’re pulling. If you’re just pooling 10 samples, the loss in sensitivity from an infectious disease outbreak control perspective will be negligible for the most part. And so I think that it can be extremely powerful to save money, to save resources, to save time.

There are logistical hurdles where if you get a positive pool, if the laboratory only receives a tube that was pre-pooled, then you have to essentially, you get that positive and then you have to call all 20 of those people up and tell them all to stay in their houses, that one of them might be infected or, you know, and and then you have people who are very anxious who need to get re swabbed somehow safely and one of them will be positive, for example. The other option is if you send all of the samples to a laboratory and they do the pooling, but then that pooling process uses an immense amount of labor and resources to put everything together, keep track of it, and then if you get a positive pool, you need to de-convolute it all. And that all slows a high throughput lab down tremendously. So there are serious logistical hurdles.

I think when resources are the limiting factor, that these approaches should absolutely be considered and in general in the United States, for our resources are a limiting factor. Time is the limiting factor. Speed is the limiting factor. And so we need to take all this into account. But it can be very, very powerful, in particular for institutions now. We’re seeing universities and businesses trying to figure out – this is a different question, whether or not I agree with this approach – but many places are now considering testing everyone three times or everyone every three days, for example. And that builds up to millions and millions of dollars very quickly. And if you can pull and bring out the hundreds of thousands of dollars, that would be much better for most institutions.

We’re also working with ministries of health in Africa, in various countries to essentially bring these same ideas to them. After we published our preprint, we got a number of interested ministers of health asking, you know, saying essentially we have X many tests and 10 times the number of people we need to test every day in our country. How do we do this? And so we’re seeing it already being rolled out in under-resourced countries.

Q: Great. And just one follow up to the serological testing. I know you’re a little more confident or bullish on their value or usefulness. When do you think we will be able to use these in a practical sense, to make decisions about returning people to school or to work or, you know, just to have some confidence of their immunity? Is it something that you think will be ready by the fall or when?

MICHAEL MINA: So is your question about individual level immune protection?

Q: Right. Yes.

MICHAEL MINA: Yes. So I am – in many ways, I think to be able to take a measurement from somebody and say you’re protected at this level, it’s going to be is going to remain difficult. I think what we can say soon, if not today, is that in general, people are not getting infected twice. And if they do, you know, it’s very rare. I was recently on a phone call with somebody who has been infected twice and both times very severely, but probably she had an immune deficiency or something along those lines. And so in general, I think we can start to say if somebody has been symptomatic with this virus once, we see that they’re not getting it again and it’s behaving like other viruses, but we don’t know how long it will last. We don’t have a number of antibodies to pin on somebody at this point. But I do think the data is now coming in fairly quickly. Some of it’s true follow up and we’re looking at protection, sort of relative risk of getting an infection over time. If you’ve already been infected ones versus not based on stratification of antibody numbers.

So I think probably, you know, in the next few months we’ll continue having a better and better idea. There won’t be a silver bullet where we say, aha, here’s the number, but we will continue edging closer to that. I think we can use serology, though, even without understanding that individual level sort of response is needed.

I think if we can at least take a step back and say we believe that people who have been infected are largely at least more protected than other people who haven’t been infected, then all of a sudden we can use this even today. We can use this as a powerful tool as we start to think, you know, especially as more and more people get infected and the seroprevalence increases, we’ll be able to use those numbers to our advantage to start grouping people in ways that make the most sense to stop or at least mitigate potential outbreaks. If you have a whole bunch of people who are otherwise evenly or able to do a given job, maybe you mix people who are susceptible with people who have already been infected on both sides so that both of these communities or groups of people, you don’t have any one group that is fully susceptible, for example. You try to, like, build your herd immunity in a very smart way. So I think that we can start using that for nursing homes. How you necessarily, you know, all nurses, if you consider that the nursing staff is the same or all equally able to deal with patients, maybe you want to prioritize people who have already been infected to work with the susceptible and some ratio because they would be less likely to bring in a new infection in, for example. So I think that there’s ways that we can use this information today and we’ll continue building up more information into the future.

Q: Thank you.

MODERATOR: Next question.

Q: Yeah, hi again, thanks. I have a question on children. I was curious if we’re any closer to understanding what children’s role in transmission is with COVID. And do you think we need to better understand that in order to know how to proceed with school re-openings in August, September?

MICHAEL MINA: So we are trying to understand it quickly as possible. It’s a hard thing to understand and to get at very quickly, especially because it’s summertime. A lot of places have been shut down beforehand. We know that there is at least one report – I forgot where I read it, that when schools opened in France in one area, there were an increased number of cases. I think in general we have seen that. We’ve seen that children can can grow this virus to very high titers, very similar with similar distribution to viral titers as adults. So therefore, we believe that children can certainly transmit the virus. The question we have is what is their relative risk of transmitting it versus an adult?

But, you know, even if their individual level risk is lower, children grouped together in such higher numbers that they can become a much greater reservoir for infections and sort of an overall force of transmission could become very high from children. So it’s something that is extremely concerning to epidemiologists, myself included. You know, I think about – it touches on everything, not just the parents who might get sick if spread to happens within a school, but the school bus drivers, the principals and teachers in the school, the community at large, the kids who are looked after by their grandparents. You know, there’s a tremendous number of risks that can occur if children turn out to be pretty high vectors of transmission. And, you know, even if they’re not the primary mode or the primary population transmitting, there’s a good likelihood that they might transmit substantially. And some of our mathematical modeling would suggest that they likely do have some role. And it’s going to be a big challenge to keep kids from transmitting.

Q: Knowing that, do you think we should not be opening schools in the traditional manner come August?

MICHAEL MINA: I think we should be considering how to do it safely. I think there are a lot of things that we can do to safeguard. Teachers – I think it’s one of the most vexing problems that we have right now. Children and, you know, will they be complying with masks, especially the youngest ones? But the point is, if we can’t get them back to school, we can’t get parents back to work and we can’t function appropriately as a society. And we know school, whether it’s school or daycare, either way, we need to get the children away from the parents to allow parents to work and to allow children to have normal social development. I don’t have a good answer at the moment for what we should do. I think that we should take this and consider it as one of the most serious questions that needs to be answered right now.

MODERATOR: If I could chime in real quick, we’re going to have Dr. Joseph Allen on the call tomorrow. And he has written up a paper or a working paper on reopening schools safely. So he will be able to answer a lot more questions about that tomorrow.

Q: Thank you.

MODERATOR: Next question.

Q: Hi, Dr. Mina. I’ll try to be brief and quick. There are some people who are believing that COVID-19 deaths are being undercounted, either inflated to generate hospital revenue or otherwise for political purposes and or ignoring underlying health conditions that might be a cause. So and then there’s an investigations by researchers and journalists that show otherwise. Curious of your thoughts on how this is being counted, if it’s being counted properly.

MICHAEL MINA: I don’t think cases are being inflated at all. I think, if anything, the true numbers of infections are underreported. This is not a political. This is a real virus, it’s affecting people in every country. It’s not some some contrived media issue. We see it very similar things happening everywhere in the world right now with devastating effects and anyone who thinks otherwise, you know, I think really has to get out of the social media bubbles.

Q: Quick, quick follow. I have I just heard something called the epidemiological assessment of causality called the counterfactual. Could you explain that briefly?

MICHAEL MINA: Well, so the counterfactual is essentially a tool or – it takes different meanings in different areas, but essentially what it means the counterfactual is something that did not factually happen, but that we believe would have happened if we did something different. So if that was vaccines, we add we add vaccines in a populous and cases go down, but before we added the vaccine, cases were horizontally distributed over time, then the counterfactual would be taking that horizontal line and continuing it into the future and being able to use that as a comparator against the factual thing that happened over that same period of time when the vaccines drove cases down. In the case of COVID-19, the counterfactual would be what would have happened if we did not close down society, you know, in March.

And we might have mathematical projections that would show cases going much higher had we not close things down. That would be a counterfactual. We use it to calculate what was the efficacy or a benefit. It’s one tool that we could use to calculate sort of the effect of policy decisions.

Q: Real quick, if someone has diabetes or heart disease and they die with COVID-19, do they die of COVID-19?

MICHAEL MINA: At this point, I would say that if somebody had been living with heart disease and diabetes and they die and they have COVID-19 in their system, or SARS-CoV-2 in their in their system, I would suggest that, yes, they died of SARS-CoV-2, of COVID-19 disease. However, it was probably exacerbated by underlying conditions.

Q: Thank you, appreciate that.

MODERATOR: Next question.

Q: Hi, good morning and thanks for taking our questions this morning. Here in Illinois and Chicago, Cook County and the state at large, they’re just really getting started with contracts to start doing contact tracing, even though the governor, the mayor and so forth had said how important that will be to reopening and really just targeting high risk areas because they’re nowhere near this 30 per 100,000 people number. They’re trying to get close to that. Are you concerned at all? I don’t know how much different it is in other states. I don’t think maybe Massachusetts is ahead. But are you concerned in general with the levels of contact tracing that are going on? How important is that to preventing future flare ups?

MICHAEL MINA: So I think contact tracing is very important. If you get cases low enough, it can be extremely important to be able to monitor, to stop outbreaks as they’re getting started. And so it’s very, very important to have those and to have the system set up. But you first need to get cases low enough so that you can stay on top of the outbreaks when they’re starting. If you throw contact tracing into the middle of a raging fire or a big outbreak, you’re probably not going to catch up unless you have immense resources and people to do that. But this virus spreads very quickly. So I think that it’s very important but it has its greatest impact when you get cases low enough so that you can really be tackling one small fire at a time.

Otherwise, it’s, you know, it’s very similar to, it’s similar to a forest fire. If you have a forest fire in the end. If you have small fires burning, you can go in and have somebody go and put out one fire. And if they see a sparks fly over there, that one person puts it out again and again. But the moment it turns into a true forest fire that one person is, even a team of people are going to do very little. You need to essentially blanket the whole place with helicopters and planes dropping massive amounts of water and chemicals. And that’s akin to us closing down. You have to eventually take a big hammer and try to stamp it out rather than doing this very targeted contact tracing approach.

Q: That answers my question. Thank you.

MODERATOR: Do we have time for one more? Or are we out of time? It’s up to you.

MICHAEL MINA: I’ll do one very quick one.

MODERATOR: OK.

Q: Hey, I know you hate to do this, but I’m gonna ask you anyway. And I also know it’s unknowable. But what are the odds of a major outbreak in the fall? Put some odds on that, please, on a scale of one to 100. Thank you.

MICHAEL MINA: At this rate, 85 percent?

Q: Wow, that’s high. OK. Thanks very much.

MICHAEL MINA: Well, I’ll clarify. I mean, we already have a major outbreak going on. So it would be a bit amazing if by the fall we stop it. I hope.

Q: If we didn’t have one. Yeah, exactly. And weather goes against us, is out of favor at that point. OK. Thanks for that. Great call. Thank you. Bye bye, everyone.

MICHAEL MINA: Thanks.

This concludes the June 22 press conference.