Coronavirus (COVID-19): Press Conference with Michael Mina, 04/27/20

You are listening to a press conference from the Harvard T.H. Chan School of Public Health with Michael Mina, assistant professor of epidemiology. This call was recorded at 11:30 am Eastern Time on Monday, April 27.

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


MICHAEL MINA: For those who I haven’t spoken to yet, my name is Michael Mina. I’m an assistant professor of epidemiology and immunology and infectious diseases at the School of Public Health at Harvard, and I am a molecular virology director in lab diagnostics at Brigham and Women’s Hospital. And I can take any questions.

MODERATOR: First question?

Q: Professor Mina, thank you for taking questions. You mentioned last week that the real hold up in testing involved the inability to get the supplies necessary for testing and you mentioned in particular a plastic part. And I wondered, Is that part of the nasopharyngeal swab that is typically used, and are there other missing things from the supply chain like reagents?

MICHAEL MINA: So lately the biggest hold up has in fact been this swab. So the plastic part is referring to is just the little swab that looks like a chopstick.They cost about 20 cents, 26 cents or so. And that has continued to be a somewhat of an issue in terms of being able to obtain them in high volume.

There’s also the tubes. So that swab after its after it’s removed from somebody nasopharynx, it needs to be put into a little conical tuber, like a little cylinder with some media, with some solution at the bottom of it. There’s a few different types, but just those tubes have been actually very limiting in terms of receiving them and being able to order them in time. So those have continued to plague a lot of the country in terms of limiting the type of testing and there’s certainly still issues with the actual test being available.

In some bit in terms of the turnaround time that still can be an issue in some places in the country. And other places that aren’t linked up to Quest or LabCorp, one of the big diagnostics, and don’t have their own testing capacity are still certainly having a lot of trouble if they’re going through the State laboratories for example that I think there’s still big backups across much of the United States.

Q: Thank you.

MODERATOR: Next question?

Q: Hi, Michael, I just wanted to ask you – it’s not about testing, it’s actually about Florida beaches, because some of the beaches were already open in Miami. They are still closed, but can you talk about how the virus is doing in that kind of environment and salty water and, you know, on sand surface?

MICHAEL MINA: Uh, yeah, I’m actually not extraordinarily familiar with what the virus would do necessarily in the sand. If the sand is hot enough, then the virus would die. But that’s, you know, midday when the sun is really beating down on it and it can be too hot to touch sometimes in some places, but in general, I would imagine, I wouldn’t be able to talk about the characteristics actually about, you know, how long the virus would live in sand.

But sand is just, you know, it’s essentially silica particles and so it would presumably not be neutralized by the sand, it would probably persist for, you know, hours at least especially in the evening or when it’s not getting hit by lots of sun and heat. Heat kills viruses when it’s hot enough.

In terms of salt water, I think it would probably get diluted very quickly in an ocean, and so I wouldn’t be too concerned about transmission, although we do know that there are certain viruses that are similar to measles that actually can transmit in water. So it’s not that viruses can’t transmit it all. There’s lots of viruses actually that transmit in fresh water. But I would imagine a respiratory virus would probably not be too apt to transmit although I don’t actually know the answer.

Q: Thank you. Do we still have to follow the same rules six feet in the water, do you think?

MICHAEL MINA: Well, I would say, that above the water, I think that there’s still plenty of room for transmission for people, you know, if your lower half of your body is submerged. Certainly the people interacting in the water could still very well be transmitting to each other above it in the air. And so, I do think that that would be something that should be taken into account when people are at the beach. I think beaches, just like anywhere are still very prone to allowing people to interact if not more so, and therefore are a likely place for transmission in particular because of how packed they can get.

Q: Thanks so much.

MODERATOR: Next question?

Q: Hi, thanks again for doing this. I was hoping you could talk a little bit about how long someone is contagious and how do we know. I know that like at Mass General, the hospital workers have to have two negative tests before they can come back to work. But somebody in the general public who doesn’t have access to that, how would they know when they stopped being contagious?

MICHAEL MINA: So, we can look at the dynamics of what happens to the virus inside of the body. And we see that usually people become symptomatic within the first week after infection, usually shorter than that. The amount of virus that they have in their nasopharynx tends to peak around three to five days or so after they present with symptoms and then the virus starts to get cleared from the body thereafter.

We tend to see that by about eight or 10 days after symptom onset, many people will have viral titers that go below the limit of detection or at least viral titers that are no longer culturable, meaning we can’t take a sample from them and get that sample to grow up new virus in a Petri dish. And so, most likely by about 10 days, most people should start to have cleared the virus sufficiently so that they’re no longer transmissible but we also know that the individuals are very heterogeneous meaning some people might clear it very quickly and some people may actually take quite a bit longer, two weeks if not three weeks to clear the virus.

And so, I think a good rule of thumb for this virus based on the kinetics of the virus in peoples’ nasopharynx is to wait for about two weeks after symptom onset before assuming that you’re most likely not transmissible

Q: Thanks so much.

MODERATOR: Okay. Um, next question. Despite the lack of testing several states are allowing businesses like department stores and dining restaurants to reopen this week. Is it too soon for this and if businesses are reopening now, when will we know more about what the impacts of this would be in terms of increased cases or deaths?

MICHAEL MINA: It probably won’t be immediate but if businesses are opening today, we can anticipate something that looks a lot like if the virus in the areas have been already mostly depleted, which they haven’t so much been, but in large part they’ve gotten to lower numbers. In general, once you open up, there’s going to be a delay. It’s going to be one case will turn into two cases and it will kind of grow exponentially from there. But it takes a little while to actually see that signal in the same way that it takes a little bit of time to see the signal when an epidemic first begins.

And so, I wouldn’t anticipate that, you know, this week we would see if there’s going to be increased spread again because of opening businesses. I don’t anticipate that it would be this week.

But it might be a few weeks from now that we would actually start to see little pockets. And I would suggest that cities that are opening up their restaurants and salons and regular businesses that customers can walk into frequently, I think that these businesses should probably be trying very hard to work with the Department of Health. Not just the business, but the towns to really having enhanced surveillance in those areas to watch out for this early, to have a low threshold to say the cases that we’re seeing right now are just slightly above what we saw last week, for example, and have that be a signal.

Of course, if cases continue to drop, that’s a good sign. And if that persists for many weeks, then that’s also of course a very good sign. But if cases were dropping and then they start to level off and maybe pick up even a small amount that could be early signals. But I wouldn’t anticipate that we’d start to see that for at least a few generation times of this virus, which might be a couple of weeks or more.

MODERATOR: Okay, I have a follow up from another reporter: He would also like to know is it a new world, or what is it going to be like after we are reopened?

MICHAEL MINA: Well, that’s a great question. I think that’s the million-dollar question.

What will life look like and it ties in very much with the last question, which is, you know, should we anticipate seeing increased cases now that stores are opening, businesses in some areas are going back and opening back up. We’ll learn a lot over the coming weeks and months about whether or not this is going to be a long term change.

The other option is that we have actually been doing a sufficiently good job to roll out testing and get surveillance efforts in place that maybe we would be able to catch cases before they become very large outbreaks. I don’t think that that’s really the case. I don’t think we have necessarily gotten testing and surveillance efforts to where they really need to be to make that a reality and to prevent large outbreaks from happening, especially if you look across the whole of the United States, but I do think that’s the direction we have to go.

I expect that we will see more cases though once we open back up. We know that the vast majority of people are not immune. And, you know, we might have some herd immunity built up in certain very densely populated areas, but not in others. And so, we’re going to have to be very cognizant of that fact. And even in those places, in the densely populated areas, it’s still not to the herd immunity threshold where we actually anticipate that transmission has stopped or that transmission would be unlikely. If we’re talking about 20% of people being infected in places in New York 80% of people are still susceptible and so that to me suggests that we’re still at the very beginning of dealing with this.

I think we’re going to see a lot of, sort of, starts and stops. We’re going to see institutions or businesses start up and go back to work and then we might see an outbreak happen in an office building, for example, and maybe have to shut down. There was just a nice paper, unfortunately I’m forgetting the name of the folks who it was by, but they did some nice modeling work to suggest that one avenue forward could be to have something like four or five days on at work and then 10 days off, so that if somebody actually were to get infected at some point during those five days on, they would then have their most infectious period while they were off. That would essentially allow these sort of cycles of being back at work to sort of mitigate big transmission events or at least prevent large outbreaks, while also still potentially keeping some number of people infected in order to start building up herd immunity over a long period of time.

So, you know, that’s kind of an extreme but creative example of how work might go back. I don’t think in this country that would happen, but it suggests the type of creativity we might have to have.

MODERATOR: Another quick follow up, you seem to be saying that we are not ready to reopen. What is your biggest fear as states reopen anyway?

MICHAEL MINA: That major outbreaks start to happen again. I think we’re still going to have nursing homes, you know, get major outbreaks, you know, occur in nursing homes as States continue to open, we can anticipate that we will see outbreaks. Even if they are among young healthy people.

Those outbreaks aren’t going to, and those people don’t end up in the hospital as much, that transmission, I haven’t seen the types of protective, sort of, barriers or whatever it might be, and protective efforts be put into place to really protect the vulnerable in our population, elderly, people in nursing homes, senior centers, things like that where the virus can transmit. I’ve seen everyone trying to do what they can, but no real effort or I should say no real structural changes that would potentially protect these people in the future, and so my concern is that will end up right back where we have been. With major cities, having healthcare systems that get overrun quickly because of major outbreaks that happen in, you know, one nursing home, you know, today, a different nursing home or center tomorrow. Or just across the population in apartment buildings, you know, places where this virus will transmit quickly. I’m afraid that we’ll just end up repeating the past.

MODERATOR: Okay, next question?

Q: Hi. yeah. My question was, I was wondering if you could break down the concerns about the existing serology tests that, you know, various labs have been working with, what are some of the concerns about their accuracy and like, given any limitations what do they actually tell us about immunity at this point.

MICHAEL MINA: So, there’s two pieces to that. One is the technical aspects of the test and one is the biological aspects of interpreting the test.

So, the technical aspects are we want tests that are accurate, we want to tests that are not going to tell somebody that they’re positive and have been exposed when they haven’t and we want tests that are not going to tell somebody that they haven’t been exposed and they have. And both of those have different potential problems.

If we’re using these tests for serological surveillance, which is maybe something new to a lot of people, but a lot of people have now heard of it over the last couple of weeks. This is a type of surveillance in public health, where we use, instead of going out and asking people questionnaires, “have you been sick,” we take a sample of their blood and we use the record of immunity that’s captured in everyone’s blood to be able to get an estimate of how many of those people have been sick or, sorry, have been infected in the past with the virus.

And if those percentage points are off at all, meaning if the specificity of the test is 98%, that might sound good. Most people would say, oh 98%. It’s pretty good. It’s good grade on a test and it’s been great. You know, it’s a good number to have out of 100.

But when you’re talking about measuring the prevalence in a population, that maybe has a true prevalence 2 or 3%, then a 2% reduction in specificity can go a long way to give you erroneous results. So, if you have a test that is 2% off and you roll it out in a population that has never seen the virus at all, you might say that 2% of the population has been infected.

In the United States, that would be around 6 million people. So when you add it up in these big serial problem studies that’s a lot. And part of the problem is we don’t actually do this test on every single person. In public health, if we want to measure the prevalence in a population we don’t normally go and try to capture every single person. We find representative samples. So, we go to people who we believe collectively represent the population as a whole and we test them and then we extrapolate out from those numbers on what the actual problems in that particular community or region would have actually been. So, a few percentage points off in terms of specificity can be detrimental to that.

Sensitivity, meaning how many people, if the sensitivity is not 100% you might miss people who are positive, that the test is not detecting. That isn’t so horrible. If you know what the sensitivity is, you have decent estimates, you can you can start to make up for that on population prevalence studies, you can assume some level of have missed cases. And in some ways it just allows you to have a slightly more conservative estimate. Of course, if it’s huge then that’s a problem. But if it’s a percentage point or two that won’t totally destroy your estimates.

But what it will do is that then sensitivity will be particularly bad for clinical diagnostics and for example for using it to test a patient in the hospital. If we miss a patient who may have been exposed and maybe is still infectious and we start using serological tests for diagnostics, which you can potentially do, I don’t recommend it but the FDA did approve test for diagnostics recently, then it can actually be quite bad because then you might have a patient who you think is negative, but who might actually be infectious and transmitting in the hospital.

And so, we want to be careful about how we’re using serology in particular when it, especially in the healthcare setting, when sensitivity might not be 100% we probably want to be more conservative rather than less in that setting.

But those are mainly issues for these point of care instruments. We’re going to see a lot of new tests that come out that are in the laboratory space that are really, sort of, high fidelity tests. Many of these are based on a technology called ELISA, and they can be quantitative, they can be very accurate, but they require, more or less, a central laboratory, a lot of times, to do it.

And so, we’ll see those in the hospital. And we’ll see Roche, and Abbott and Diasorin and all of these companies are coming out with their own version of this serological test. And generally, those will be pretty good coming from these companies.

We’re also going to see homegrown tests. We’re developing a few different tests here at Harvard to sort of make in a high throughput fashion so we can actually do surveillance efforts. You know, hopefully partnering with communities or governments and things like that. And so, it is possible to make high fidelity, high accuracy tests. It’s just, putting them on those plastic cartridges can sometimes cause problems.

The interpretation is a whole different issue. And that’s where I think actually having quantitative testing is very important. And so, what I mean by that is that the test itself gives you a value, similar to how people are used to getting a cholesterol value or a sugar level. This is an antibody level and that antibody level would, what we look for, what we call thresholds of protection and that is a value above which that we correlate, from an epidemiological approach, we can correlate with protection. We have to do the studies and they’re getting started and some are ongoing already.

But if you measure a number of people, let’s say healthcare workers, and you have a diversity of antibody levels at the beginning of this study, you can then ask how many people go on to get a second infection. Ideally, nobody will get a second infection if they have antibodies, but that probably isn’t going to be the case. We will anticipate that some people will get reinfected. Hopefully, those people will have low antibody levels and we’ll find that as long as people have above a certain level, let’s say 100, you know, whatever the units might be, if you have 100 or above, this is a made up number right now, then maybe you are protected from reinfection. And if you’re 100 or below, then maybe you will still be susceptible with some likelihood of getting reinfected if you’re exposed.

So, we need to find those thresholds of protection through rigorous studies right now. And then we need to do quantitative analyses on people if we want to hope to be able to give them, sort of, an immunity passport, if you will. So, until we do a lot more research and until the availability of very quantitative tests becomes much more available, I suppose, I think that we will have a hard time really being able to interpret peoples’ antibody tests. And unfortunately, a lot of the tests that have been introduced, both these plastic cartridge ones as well as the commercial ones, they give just a plus or minus. And so that’s why many of my colleagues and myself and people across the world are in parallel developing very quantitative tests.

MODERATOR: Great, thank you. Next question?

Q: So, I mean, you’ve been touching on this. And I know there’s been a lot of conversations around this. But you know, I write a lot about one concern. Workers at their respective industries expect to get a call back to work, restaurants, you know, cruise lines, if that even applies sometime this year. Retail workers, whatever. And you’ve definitely been touching on this throughout this conversation, but like macro level, what are you know X number of things from a public health perspective that need to happen before people can like en masse exit their homes to do their work on the job.

MICHAEL MINA: It’s a, you know, there’s no textbook written for this oddly, you’d think that there should be. So, these are the discussions, many of us have been having on an ongoing basis with policy makers and amongst ourselves. So, what does need to be put in place. And I think there needs to be very rapid monitoring and whether that is purely test based, purely symptom based, if it’s app based. I think we are in a place in time when we can actually leverage and sort of combine all of these different measures, to try to be able to detect early on if an outbreak is happening. It might be things that are as significant as having testing on a on a daily basis in some workplaces.

And so, that becomes a big logistical hurdle, but can you use essentially like pooled sampling techniques, where everyone swabs themselves and sort of pools a sample and one test gets around to see if anyone in that group of 10 people, for example, has an infection at the time, and if so, then maybe you ask all 10 of those people to stay home until you figure out, you know, are any of them going to become additionally infectious.

That would be sort of an extreme example of what type of system we have to set up before people go back to work, but also not a completely unreasonable idea, especially if these very rapid point-of-care viral tests become available that might actually be something that’s feasible. And these are things that might look like a little toothpick that you know you touch to the roof of your mouth or something it can light up if it’s positive after 10 minutes. You know, something along those lines, we might end up seeing those in the near future and lots of people are working on this very cheap, ready-to-use, point-of-care virus tests.

But then the other programs that really need to be set up, we need to have contact tracing in an efficient and manageable way and I think that contact trace in the way that it works now, when you have an army of people go out and try to track down every single person that was infected and try to track down all of their contacts, that can work very well for viruses that aren’t widespread, but once a virus becomes widespread, it becomes very, very difficult, expensive, resource intensive to actually be able to do that and to consider doing it across the country, it just wouldn’t happen in a way that’s really effective, at least, you know, there will still be outbreaks across the country.

So that’s where I think, you know, we all have these little devices in our pockets all of the time, these phones and those could actually be leveraged, I think, in a way to really work with governments and public health agencies and Departments of Public Health to utilize that information to accelerate contact tracing. If somebody gets an infection, you know, maybe we have a society where everyone’s at least willing to participate.

You know, I think there will be a lot of privacy issues, but the hypothetical situation would be that everyone would participate and if somebody becomes infected, then it’s very quick to say who were called all of their contacts been over the last five days and send a text message to all them to stay home or order a test and you have sort of a rapid deployment of tests, and you ask everyone to test themselves four days later, for example, at a time when you anticipate that their viral load would have been become detectable. But in the meantime, they stay home.

I think that those are the kinds of really adaptable, quick to roll out, sort of, processes we need if we really want to do a good smart job at how to protect ourselves in the future. I think we’re a long way from a lot of that, though.

MODERATOR: Do you have a follow up question?

Q: Yeah, yeah. So it’s – so like, again, macro-level it’s testing and contract tracing would sort of like be the tools that we would use.

MICHAEL MINA: Yeah, I think so. And monitoring, you know, using these other types of monitoring systems. You know, I am in favor of like trying to work with the Verizons or Googles or whoever, it needs to be to be able to leverage the huge amount of data that we all create every day, both location-wise and otherwise, to try to see if we can actually use new, more sophisticated, mathematical, data-driven tools to detect things in the absence of testing. I think we also need to really scale up the ability to do tests and the logistics surrounding the tests. I don’t know if any place in the country really that has true high throughput testing capabilities at the moment. If we’re going to rely on serology as one of the tools that we use moving forward, I think we need to figure out how to do that at very high throughput, for example, and I think that we’re a long way off from actually having the laboratories that are sufficiently set up to test potentially hundreds of thousands of these antibody tests in a good quantitative way. So, I think there needs to be infrastructure advancements as well.

MODERATOR: Another question from zoom chat: Do we definitively know that people who recover from COVID-19 have immunity from reinfection. If yes, for how long does this immunity last. And what else do we know and don’t know about this.

MICHAEL MINA: We don’t definitively know that at all. We anticipate that the virus will certainly serve to act much like other acute respiratory viruses. We know that there is some immunity that’s essentially conferred from an infection and usually antibodies and the serology type of test, that’s what they look for, can be a good proxy of that. But we will need to do the test and the studies to determine if people who do have antibodies, if they become exposed again, will they be able to be reinfected.

And so, some of those studies will, for example, utilize healthcare workers. Healthcare workers might be a higher risk than the regular population to be exposed and also might be willing to participate in these kinds of studies where we might have to be testing them on a weekly basis, for example. So, the studies that are being produced and set up right now will help us to understand whether or not people are actually protected.

We do anticipate that some protection will happen now. Probably most people will have some level of protection for at least a few months if not a few years. It could be that protection wanes over, or decreases over the course of six months after infection. And the next year, people might be likely to become reinfected again. We just don’t know. My feeling, which is not based on data for this virus, but it’s based on data from other viruses, that people will most likely have some level of protection that will degrade over time, potentially even over a year. And they’ll be able to see get reinfected. But that reinfection might be less severe than the first infection they had. And so that’s a good thing.

The bad thing there is that if people can be reinfected the next year, than in the absence of a vaccine, then we would anticipate that that vulnerable people who have not yet been infected will still, or anyone, will still be able to acquire it from those individuals getting it a second or third time as the years go on. Even if those people aren’t themselves particularly symptomatic, they can still transmit it, potentially, so that’s what we’re concerned about. And so, probably immunity will be partial.

MODERATOR: Okay, thank you. Next question?

Q: Hi again. I have a question about the timeframe for the rollout of serological tests and who would get them first. I presume it would be healthcare workers and first responders, since it would be important to know whether they might have some level of immunity. But how long before a typical office worker could get tested for, you know, potential past exposure and how, would you, I mean, I’m asking you to guess here, of course, because there’s many, many factors that would feed into this. But I’m interested in what the timeframe might be as compared to, say, the development of a vaccine.

MICHAEL MINA: Oh, much quicker. And so, we’re trying to just, I can speak from my own limited little bubble of experience, and so we’re trying to set, we have two different efforts that are ongoing at the moment. One is to develop, working actually with industry partners and philanthropy, is to develop a way to rapidly deploy these test kits that will allow people to at home, give themselves a small finger prick and put a drop of their blood onto a piece of filter paper and then mail that back in to get a test performed in a very quantitative way. So that’s the test kit, you know, those we hope to start deploying very soon and some places around the country have actually started doing it sort of more ad hoc packaging of their own test kits and mailing them out to people to do early surveillance and that’s been great.

I think the way to make it sustainable and available is to both, you know, have it be, to a certain extent, direct to consumer or work with businesses directly to offer it as a service, and that is I think one way to make it actually happen. I don’t foresee the NIH funding, like we can’t. I think this is where industry and academics like really generally, there’s a tension there and academics, usually to a certain extent frown upon industry and vice versa. But I think that this epidemic is raising an acute awareness that there needs to be stronger partnership and the way that business works is sometimes essential to actually get things like this done. So, we’ve seen industry start stepping in to help you know with these logistical things just sending out kits and receiving the back.

And then needs to be the testing facilities to actually perform the tests and again, that’s something we’re trying to build now. Some sort of facility to actually process serological tests and that can also link back into the Department of Health, for example. So, we would want to let the Department of Health and know peoples’ serological status as a public health tool even if it, you know, and that will be on every State’s DPH to figure out what exactly that looks like.

But that would be very, very crucial to have that sort of infrastructure built and to be able to actually provide these tests and results back out and I would say most testing facilities are not really set up for that at the moment because most biological tests in this country for healthcare have been set up through healthcare facilities, hospitals and the throughput that is anticipated, even at the big laboratories like Quest and LabCorp is they are still set up to receive tests that would come from hospital networks and clinics and things like that and not the kind of throughput that would be from the everyday person in an office who wants to get tested. So, that’s where I think this new infrastructure needs to be built, and we are also working to try to build like a high throughput testing facility, similar to what we did at the road for PCR a couple of months ago. And that’s now been sort of moving very well. We’re trying to do the same thing now in a different location for antibody testing and we hope that in the next month or so, sort of high throughput testing will become available.

I think similar efforts might be happening elsewhere. I’m not sure, but the time scale your initial question was what’s the time scale. And I think it’s much quicker. It’s imminent, it’s within the next month or so. We should anticipate seeing good tests become available in centralized laboratories, even if they’re specialized centralized laboratories to really make it happen.

MODERATOR: Next question?

Q: Yeah, thanks. I have a question about PCR testing. So, as we know that that’s something that we still need to scale up but there are also a lot of labs out there with unused capacity. And some of that has to do with the supply chain, but I’m also hearing that that also might have to do with either testing criteria or some sort of administrative issues. So, I’m wondering, what are some of the other barriers that you see just sort of tapping into some of this unused capacity and sort of how we can get around that?

MICHAEL MINA: You know, a lot of it is logistical. How do you get a test from point A to point B, a sample. And there are certainly still, the supply chain issues certainly are still happening there is excess capacity. In general, I mean, there was a tremendous focus early on in this epidemic in the US that testing capacity itself was limited but that’s not normally the case. As I think everyone knows that it’s not normal that we’re limited by tests and what we’re seeing now is more of the norm. We’re limited by access to testing, sort of procedures, how do people get tests performed. Everyone right now has to go and see a healthcare provider. So, this generally under-samples impoverished individuals who are low-resource, low-income individuals who can’t just hop in their car and, you know, and go to a drive thru, make an appointment. You know, that just the networks and the information flow is not always there across the Community in a way it should be.

So, I think that there needs to be just much greater access to the information of how to get a test, it needs to be simplified, it needs to have no cost, no strings attached. Those are the types of things we really need to do to make testing, effective and worthwhile. We have also a lot of people who are undocumented. People are very scared to get a test and many are not, I’m sure, because that requires them to engage with some sort of bigger system that they might be afraid of at this point in our nation’s history. And so, I think that, you know, undocumented individuals or people for any reason who have some fear of being recognized in any way by our government, you know, whether it’s right or wrong in terms of whether their assumptions of what the actions might be, I think that there needs to be a lot of messaging and very clear guidance that, you know, not should not be a barrier, that those two things can be kept separate. It’s not clear to me that they can be separate, but I think that’s also a major barrier right now for at least a decent fraction of our population from getting tested.

The other bits are just more, it’s just more logistics. I’m on a committee in Massachusetts to talk about just how do we load-balance the system, how do we actually get samples from, you know, Western Mass to Eastern Mass where there might be more testing. What kind of courier services can do that, do we need sort of a FedEx hub model or do we need sort of a parallelized, distributed model. I would say that Massachusetts is probably one of the leading states in terms of having these conversations, but the logistics surrounding this have been big hurdles and, in particular, it’s the reporting back and forth and having all the software interface. Because we’re, we are not a sort of universal health care system. We don’t already have a way in place to have centralized access to information about people. You know, every institution or clinic needs to be interfaced separately to whatever laboratory, they might be interfacing with. And so it creates really large hurdles.

Anyone else?

Q: Yep. Hey, thanks. Um, this isn’t about testing, but maybe you can field it. Um, do you have any sense of which therapies might be kind of promising, whether it’s Remdesivir or plasma. What’s the latest occurred on that end.

MICHAEL MINA: So, I think Remdesivir is still a leading candidate. I think there’s some monoclonal antibodies that we will probably start to see really try to, I personally think that monoclonals are a good way to go, our body knows how to handle them, the toxicity will hopefully be low with those types of things. But they’re still early stages. There were just two new monoclonals in a Nature Medicine Report recently, or Nature, I forget.

Remdesivir I think is a good option. I think it makes sense, you know, biologically it’s a true antiviral versus something like hydroxychloroquine and, you know, I’m excited to see some of the early results that will come out of it. Of course, there was some information leaked maybe a week or two ago that you know it’s very hard to interpret what the, what the information actually was. But I think we’ll start to see testing results, you know, or study results come out hopefully in the near future.

I think hydroxychloroquine, I, you know, I’ve never been a particular fan of it. If it were going to have an effect, it would be probably have an immunomodulatory and not an antiviral effect. And immunomodulators, are you know, they’re not the kinds of drugs that you want to use too early because you could end up causing, you know if you’re if you’re trying to push for an anti-inflammatory response to get around some of the excess inflammation that’s associated with this virus, you don’t want to do that too early for example, as a prophylactic because that could actually cause a lot of the virus to grow more rapidly so those things are kind of double-edged swords and I don’t necessarily know that they would work at all in the first place.

So I think Remdesivir is still the main one that I’m banking on. There’s been a few others and I’ve had trouble keeping up with all the different trials that are ongoing, but my anticipation is that we will see positive effects from that. But I don’t have any first-hand knowledge of it.

MODERATOR: Great. Next question.

Q: Hi, thank you very much. Um, I wanted to follow up on what you were saying before about contact tracing and obviously it’s not possible to do at the current moment because you just couldn’t hire that many. But if you were to have some benefit from technology, phones or otherwise, what metrics should states, especially the ones that are opening up now use to determine how many contact tracers they need. Is it is it based on their state population based on the number of confirmed cases. How, how should States think about how many they need.

MICHAEL MINA: So certainly, if you’re in an urban, that will have to be dynamic. It will have to do with how many people, what’s the prevalence and that’s why the moment the prevalence gets too high, it becomes really untenable.

If you have one person infected, then maybe you only need a few contact tracers to go out and find all of their contacts. But obviously, one person is not where we’re at right now. If you have hundreds of people who are infected, then you probably need, you know, depending on what the setting looks like, you might need a few people for, not necessarily a few people for every person but, you know, you could think of having a near one to one with like every person that gets infected there is somebody who has to go and track down all of their contacts over the course of a day, and that can take a very long time.

If it’s a rural place, you know, on the one hand, you might have fewer contacts and fewer opportunities for spread, but it also might be that those individuals live much further away, and it just physically takes longer to figure out all those contacts and track them down. If, on the other hand you’re trying to do contact tracing in a very densely populated area of a city, you might see multiple cases happen pretty quickly, and each of those might have, you know, 10 or 15 contacts in a given day or more. And you might have to go and try to track down all of those and probably a virus like this will outpace the contact tracers always, you know, in a setting like that.

And so, that’s where I think that it’s, I don’t want to say it’s a futile effort, but it’s a difficult effort when you have any sort of percentage points that are in the population that are really becoming positive, I think, and so it becomes very, very difficult to do, you know, in the type of setting, we’re at now where we actually have, you know, still thousands of people who are infected, at least in any given state, we’re having, you know, we still have large numbers of people who are becoming infected every day.

And to really be able to track those down, there’s going to be a delay. There’s already delay from the testing time and this virus just moves so swiftly through populations that it becomes really, really a difficult thing to be able to keep up with it. And, you know, if it was something like HIV, it would be a very different story. HIV transmits much more slowly, just because of its transmission routes into potentially fewer people. Something like an acute respiratory virus that can very swiftly transmit in a matter of days to, you know, an exponentially increasing number of people, it just as hard to keep up. And we see it happening already where we get, you know, I know from my own personal experience from testing that I’ve done, I’ve had contact tracers get in touch with me and asked me for information on patient X or Y, who was diagnosed, you know, 10 days ago or 11 days ago with this virus in a facility that already had a large number of people who were also infected. So at that point, it really does beg the question, what’s the point of doing that type of contact tracing with such a delay. Those people have already cleared their virus and gone on to infect other people. So if you’re if you’re behind the curve than then it’s almost, in a way, it’s not, you have to get ahead of the curve with contact tracing to make it really usable and that will depend on the setting and the prevalence in terms of the absolute number of people needed.

Q: But I’m just to be clear, when you talk about the prevalence sort of overwhelming contact tracers, any area of the country that’s still seeing even a couple hundred cases a day, which is almost every area of the country, it would be, it would be too much, given the delay in testing time and, you know how, how many contacts that person has once the economies are reopened. That’s what we’re talking about right?

MICHAEL MINA: Right, that’s correct. And then that’s where I think contact tracing is absolutely essential, but once you get prevalence way down. If you can get it down to really be feel like you have a chance at capturing most of the epidemics and outbreaks, I think that that’s where it can be very useful. But brute force, going out and, you know, sending thousands of people out to contact trace people who were infected, you know, a week ago has diminishing returns for sure.

Q: Thank you.

MODERATOR: Okay, next question?

Q: Thank you. So if I understood you correctly, a few questions ago you said that you expect that within a month, we will have high throughput serological testing here, so the natural questions that follow are who’s we, who’s doing that, and what kind of capacity do you expect to have. And just kind of briefly, can you distill why that will matter.

MICHAEL MINA: Sure. So, when I say we, I’m saying that more in the royal We, I think.

We, as a country will start to see different groups pop up with testing facilities that can perform tests on, you know, many in a day. And that might be thousands, you know, instead of a lot of the clinics and hospitals will have their own capacity. But again, that’s going to be focused primarily on patients in those places. And so, I think within a month we’ll see some places become available. You know, start offering testing to a more general public, but maybe in low thousands, for example. And then I think we’ll, in maybe two months, we’ll start to see if it becomes, you know, if it continues to be necessary, I think that there will probably be some testing facilities open up that can maybe do much larger numbers than that, you know tens of thousands potentially and I know that we, me personally, and some others are sort of brainstorming, how to make that work technologically at the moment. And so, we’re looking at the horizon and asking is there going to be sufficient testing available, and I don’t really think that, in absence of some extraordinary efforts to really get super high throughput testing, I don’t think we’ll have the type of testing that public health really will demand, you know, but we’re trying to change that.

And what was the second part of your question?

Q: And why does it matter to be able to do that level of testing.

MICHAEL MINA: Well, I think, I think antibody testing, you know, so we’ve been so focused on viral testing, which is very important, but at the same time, I’ll give you a concrete example: say we’re going and we are sending, you know, people in to test everyone for a virus in a nursing home, and we get we get all the tests back from that nursing home to find out the 10% of the nursing home is infected. Assuming that we can’t do this every single day or week for every nursing home around the population and nursing homes are just one example, we won’t know how to interpret that 10% of people in the nursing home. Does that mean that 10% of people are infected currently and the epidemic is on an upswing and we will expect to see, you know, next week that 15 or 20% of people will be infected? Or is it on the downswing and actually that most of the people in that nursing home, for example, have already been infected?

And that’s where serology can really help us and I like serology because it can also give us not just a not just a history of infection without knowing anything about how long ago it was, but by measuring both IGG and IGM, IGM is a more early marker, or it’s a marker of more recent infection. If people, we can look at the ratios of the two numbers for example, and we can say, okay, this person has IGG to IGM ratio of x, and we might be able to start inferring that they were actually, that they have only recently recovered from their infection. So, it gives us a lot of additional information above and beyond what just virus testing will provide.

It will also let us know, in the same vein, it will let us know where we see immunity gaps, where people have not been exposed at all and maybe those are places that we might want to have enhanced surveillance, for example, if they are completely susceptible, but still at risk of maybe if they’re in a nursing home or a senior center. And maybe other places will say, oh, you know, this has already blown through this whole place, infected 80% of those people and we need to keep an eye on that. But, you know, we can at least assume that there won’t be a major outbreak there again at the moment because the virus has already gone through and we expect at least some community to help protect, at least for the short term.

So that’s where I think doing wide scale surveillance with antibodies, it ends up being more informative because you don’t have to capture people right in the moment in time that they’re infected and that might, for a lot of people only be eight days or so. And so, if you don’t catch them, you don’t know anything about, were they infected last week, or will they get infected next week and antibodies, at least help you understand where that person is in the course of the epidemic.

Q: And super quick follow up. Is there a precedent or a model for the scale of the technology that you would need to apply to do it at the level you need?

MICHAEL MINA: There isn’t a model at the moment because it’s never been needed before on such a scale, I think. And that’s something we’re working with. Various people in the Boston community, Cambridge and Boston community to try to make a model for and I’m hoping that it will succeed because you know I think that to really process, like we’re still talking about how we don’t have enough capacity for PCR testing. I think if we can get serological testing to a good place where it can really be done in large numbers, I think it can be extraordinarily helpful to understand and contain this virus. And so, and I think if we can do that the demand will actually end up being, you know quite a bit higher than PCR testing because we’ll find that it’s actually very, very informative for public health in a way that PCR testing might not always be.

So, I think the potential demand will be very high. And who’s going to do that, I don’t think we can count on the federal government or, you know, or state governments to really be able to develop robotic systems that are going to be able to process, you know, tens or hundreds of thousands of specimens a day. And so, there needs to be a new model built to really be able to do that if we want to bank on serological surveillance as a mainstay of outbreak control into the future.

Q: Got it, thank you.

MODERATOR: Okay, it looks like that’s our last question for today. Dr. Mina, do you have any final words, you’d like to say before we end the call?

MICHAEL MINA: No, I don’t think so. I wanted to put a plug in for something I’m doing. I don’t actually know the day. So hopefully, like maybe I’ll tweet it out or something if people follow me on Twitter, but there’s a famous chef with the same name as me, Michael Mina, and we’re going to try to get together and do a Facebook Live or Instagram Live kind of event to talk about things like restaurants and you know what this virus means for restaurants reopening and closing and things like that. It might be this Friday or Thursday. I’m not positive when it is, but I’ll try to let people know if it’s happening, might be of interest to follow.

This concludes the April 27 press conference.

Michael Mina, assistant professor of epidemiology (April 24, 2020)

Bill Hanage, associate professor of epidemiology (April 22, 2020)

Richard Serino, distinguished visiting fellow at Harvard’s National Preparedness Leadership Initiative, former Federal Emergency Management Agency administrator, and former Chief of Boston EMS (April 20, 2020)