A New Strategy in the Fight Against COVID-19

We're going to be very aggressively focused on early case recognition, isolation, and contact tracing. Contact tracing is the process of identifying people who've come into contact with an infected person and alerting them so that they can get tested and isolate themselves to slow the spread of the disease. This is a job that has traditionally been done by public health workers in several countries around the world, including Singapore, South Korea.

Much contact tracing was done digitally with the help of apps the people downloaded onto their phones. To learn more about contact tracing, how it works, what it's good for, and what the challenges are facing contact tracing. We're joined by doctor Louise Iverse. She's the executive director of the Massachusetts General Center for Global Health at the Massachusetts General

Hospital in Boston. She's a doctor who specializes in infectious disease and IT outbreaks, and she's been thinking very deeply about that question. Louise, thank you for being here. I wonder if you could start by just explaining to us contact tracing one oh one, what is it? And then we'll move to what's it for. So contact tracing is basically identifying and then isolating people who have been exposed

to an infectious disease. And the idea is that by isolating those people who been exposed, you can interrupt the chains of transmission and in good quality contact tracing. Explain to me what the mechanics are. First we discover that person X call M Noah has been exposed to the virus and is infectious. Then what happens. So the first step is identifying what we call an index case. So an index case is a person who is confirmed to have the illness in this situation COVID nineteen. So let's

say you have a positive COVID nineteen test. Then that triggers a notification to the public health authorities. So then what they do is call you up and speak with you and try to motivationally interview you to remember where you've been, who you've been in contact with. For COVID nineteen, we think especially about being with someone within about six feet of distance physically for about ten or fifteen minutes more.

And we make a list of all your contacts, and then we set about communicating with those people who are contacts and giving them particular pieces of advice. You might offer contacts a test. You certainly want to know if any contact is already having symptoms, because they might actually become a case. Then if you have symptoms and a positive test, you become a case. You definitely want those

folks to quarantine. And you know, many people are not in a position to easily self isolate, so many people talk about supported isolation or supported quarantine. You know, where you investigate in speaking with the contact what their barriers are to being able to stay home for two weeks. So that's kind of the process. And why does that

not very quickly spiral to enormous numbers. I mean, I can understand that under social distancing it might not spiral that much because I'm not sure I've actually been in contact that much with anybody except from my kids. But if I had been out and about because there wasn't social distancing, there would presumably be many people, and then that would lead to many other people and the tree would grow very, very rapidly. So how is this doable

in a non socially distanced environment. I mean, you just perfectly described the pandemic, you know, because that's exactly why there's exponential spread, is because every person potentially infects other people. So you're totally right. Contact tracing is a big task. It quickly becomes very many people, and you are essentially trying to chase ahead of the speed of the virus and at least take out of circulation enough people so that they don't go on to contribute to that exponential spread.

So it is a very very big task. It is certainly normally in public health what we try to do at the beginning of epidemics, and so there's certainly debate at the moment in Massachusetts, say about how helpful it is to be trying to do contact tracing. I do think that the human capacity to do it gets quickly stretched. If we had over two thousand positive cases and new positive cases in Massachusetts yesterday and the same amount the day before you quickly get into the number of contacts.

It's very, very many. And this is why I personally became interested in the idea that technology should be able to somehow help us in trying to deal with this huge task. In a place like Massachusetts is a good example where we both are, where there's already what you epidemiologists call community spread. What would the scale of contact tracing have to be actually to use it to beat

the virus down? Has that ever been done? I mean, obviously, contact tracing has been used successfully as it was in South Korea to help control the virus, but only a few people have it. But are there any examples in the world where contact tracing has been used to knock

back community spread? I think you could look at Ebola virus disease as a example of a very frightening illness with a much higher mortality rate that we're seeing for COVID nineteen, where contact tracing has essentially been a major contributor to ending the outbreak. Certainly in the twenty fourteen twenty fifteen outbreak in West Africa before there was a vaccine. They're very different diseases, so I'm not trying to say

that they are transmitted the same way. But the question for COVID nineteen is if we don't try to contain the virus, what is our alternative. We don't have a treatment, we don't have a prevention, a vaccine. I think from the perspective of public health folks, if we don't try to do contact racing, we're just kind of sitting back to say, well, it escaped and there's nothing we can do. We just have to wait. So I think when you look at the numbers in the United States and certainly Massachusetts,

it just seem overwhelming. But basically, the two features around contact tracing that are important are you have to be fast, because you need to identify the contacts and put them in isolation quickly, and then you have to have good quality testing. So if you have an accurate test that's widespread, and you have contact tracing that's fast, then you could be successful. You know, Massachusetts just hired or is in the process of hiring a thousand people to work on

contact tracing, and I think that's a big start. I think even if we don't really believe that it's possible right now at the peak of our epidemics, we definitely need to have contact tracing set up and established and robust for when we do all start coming back into circulation and moving around, when we get to the bottom of our plateau, because we're going to have other waves of illness and we need to be able to blunt those much more aggressively than we did with this first wave.

If you have a thousand contact tracers up and doing their jobs at their phones, and you have two thousand cases in a day, does that ratio work? I mean, how long does it take one contact tracer to trace all of the contacts of a person? So I think the first job is really the case investigation, so they speak to the case. I've spoken with some people who are working on this now and they say that first call can be forty minutes, forty five minutes. Speaking to

contacts can be shorter. Some contacts don't answer the phone. Some contacts immediately say okay, no problem. Others may take a little more time. Others may be in difficulties, you know, they're worried about their job. They need to be referred to social services. It's not a fast task, so that I grant you, It is not fast. And how does one find people whose phone numbers one doesn't know, or whose name is one doesn't know. Do you just have to sort of give up on that person you try

to use some reasonable detective efforts to find them. Yeah, I think you know. Again. Contact tracing historically is a pretty nuanced activity, and the people who are trained to do it in normal public health circumstances are like investigators. They're like disease detectives. They tried to understand what's going on. They tried to look at how many contacts became sick, and so if many of the contacts became sick, you

might need to do a deeper investigation. It's nuanced, and we are in the middle of this massive crisis in some ways, you know, trying to adapt what's a thoughtful and specialized and very human interaction into perhaps something that can happen faster. But yes, one may not know all of the people that you were in contact with. If you took a bus or a commuter rail, you may have been in the same car as someone for twenty minutes.

You don't know who they are. So there are certainly weaknesses in the capacity at this level of scale to do it. If we look at say South Korea, who had a very successful containment effort. They used many components to their contact tracing activities. They looked at closed circuit television, they looked at credit card transactions. They had a lot of components to their contact tracing, and they had large numbers of people doing it. So I think on principle,

it's clear that it works. And I think the question that you're coming back to me with is really is this feasible at the scale that we're at. And my response to that is, we have to make it feasible, because in the absence of a vaccine, in the absence of a treatment, if we don't try to expand our contact tracing, I don't really see other solutions for us for the eyebreak. We'll be back in just a moment.

You mentioned the obvious appeal of so called digital contact tracing or automated depending on what terminology you like better, as opposed to the manual or what I would call the human form. And South Korea has gotten attention for what it's done. Israel has also gotten attention, positive and negative, for the fact that it's internal intelligence services. Just told everybody one day, by the way we've been tracking where you all are on your GPS phone data this entire time,

and now we're going to put that to COVID. I actually know somebody who, as far as she knows, she was just in her house and she got a text message from the Internal Security Services in Israel saying you've been close to somebody who has the virus, and you know, she tried to reach some human being to say no, I wasn't you know, maybe someone stood outside my housin smoked a cigarette, but she wasn't able to reach anybody, so she self isolated. So that's a very immediately intrusive

form of this. My questions about this are really fond of two categories. One is can it work? That's question one, and then question two will be can it be ethical? So I think we see that it can work because it has worked. So it worked in South Korea, some component of it worked in Wuhan, and actually, from a public health perspective, if we could identify every case and identify every person who'd been in contact with that case and isolate them, we would be able to stop transmission.

So theoretically it can work, and in practice we've seen places being able to implement it, especially at the beginning of their eyebreaks. So I do think it's feasible. The question is, in terms of technology, what can work Bluetooth. I'm learning a lot about. It's very interesting because our phones, if they have Bluetooth in it, are emitting Bluetooth signals

all the time. And the idea that your phone itself could identify other phones that you have been around and hold that information for some period of time and then notify you if and only if another phone gets flagged as being infected. It's an interesting opportunity to use a piece of technology that already exists, that many of us

already have. So I think the idea of using Bluetooth proximity to try to help discover contacts is a very interesting and intriguing idea, And I think we have to figure out how it can be used, how accurate it is, how many false alarms it might give, how well it is at picking up whether you've been within six feet or if that's an error, If it can tell if

there's a wall between your six feed like. There are many questions about it, but I do think it's intriguing and it's very laborious to do all the manual tracing yourself. So I think technology should be able to help us. Here's what seems to me to be the biggest challenge there in order to do the human contact tracing, the

manual contact tracing. You said, what triggers the whole thing is that the public health authorities are informed that someone has had a positive test, and the people who are doing the contact tracing are indirectly working for the government.

And if you're going to connect up the digital data to actual human tracing in order to give people the support and advice that you need, at some point you're going to have to connect the digital data that's been gathered to the real human being who is connected to the government. And I imagine there are some barriers we could put there to maybe keep certain aspects of the

data from the government. But I think for the ordinary person going to think, well, look if my phone tells somebody everyone I've seen and what that means where I've been, that effectively the government is going to know that. How does one begin to think about making people feel safe and secure under those circumstances in the normal process. COVID nineteen is a notifiable illness Therefore, anyone who runs a test a laboratory, who runs the test, they must notify

the public health authorities of a positive result. So already there's a kind of trust given to the state about a medical condition. I think many people don't realize that that exists. Already. There are notifiable diseases. They've always been notifiable diseases that the public health authorities are told about

with your name, not just that there is a kise. Yes, effilment has this problem, and that is what normally triggers case investigators and contact tracers for other illnesses, although of course they're much more snow moving in general, so it happens at a much smaller volume. So no one has really heard of contact tracing before. So if in a

perfect I'm calling it a human system. But of course the people who do this for a living keep reminding me, like we use software, it's not an unaided human process exactly. But if you had a perfect process, there would be this huge notebook or database that already did have all your contacts and all your information, and we already allow that to happen in the public interest because of notifiable infections.

The interesting thing is that the human ability to do that is flawed, and so that massive database is not really there. And also in Massachusetts, for example, there are three hundred and fifty one boards of health. Those are municipal boards of health essentially county and municipal authorities exactly, And so there's some decentralization of the process, there's some

kind of enactment of local authorities to do things. So there's not one massive, big data set of where everyone has been and who they've been in contact with, but

in theory there could be. And yet when we propose to do that through technology, it does increase I think both the power of that creation of a data set, which makes us concerned rightly so, and I think it also would potentially increase the scale of the data that's available, and it would congregate it in a single place potentially, and it makes people concerned understandably about having a data

set that others could access or use. So two people who I've been listening very carefully to and learning a lot from in the discussion about digital tools, especially bluetooth tools, have been a professor Ron Revest and Dannie Weisner, both at mit Ron actually created this group which creates a protocol a Bluetooth Proximity protocol CULPACKED that would ostensibly always have the information remaining on the phone and really only

allow anonymized information into a centralized cloud. What Apple and Google have proposed to do with their operating systems on iPhones and androids to as far as I can see, is following the recommendations that are in the PACKED protocol that Ron has led, and that will maintain a degree

of privacy for individuals. That seems robust technological solutions, I'm sure are part of the answer here, but they will also have to be a policy component to it as well, especially if we need to ultimately connect the data on people's phones about their contact to human tracing. The state will have to make a policy decision about that in light of the full set of public values and public norms and constitutional rules in it. That's going to be

a tricky moment. Yeah, no single component of addressing the pandemic will work on its own. It just won't. Outbreaks are complex social phenomena as well as scientific events, so we have to have a comprehensive, integrated approach that acknowledges that humans don't make rational choices, that the virus is

brand new and we're still learning about it. That we have to test more, that we have to connect testing to contact tracing, that some people can't self isolate without support, that some people a particularly vulnerable boat to being infected, and to the solutions we're proposed here, like the contact racing, and that we have to have a smart approach that can take care of the most vulnerable people, take care of sick people, help other people from getting infected, and

allow us to start going back into society. Again, no one thing is a silver bullet, and it really requires strong leadership that has good scientific understanding and that embraces the social component of disease as well as just the

medical component of disease. Sometimes when I'm listening to the debates about should we quote unquote reopen things, it starts with some people saying no, too dangerous, other people saying we must do it, and then comes a sort of reasonable voice saying, well, we can reopen once we have sufficient testing and sufficient contact tracing and supported isolation, and in a kind of Goldilocks way that seems to be,

you know, the reasonable middle answer. One takeaway from me of our conversation is that it actually may not be that simple, and that the presence of robust testing and robust contact tracing where there's community spread doesn't guarantee successful control of the epidemic. Am I hearing you correctly in

that regard? I think you know so. I'm an infectious diseases doctor at Mass Journal Hospital in Boston, and as a doctor in the hospital, it's been really terrifying over the last few weeks to see our cases going up and up and up. So I think that some of the hesitancies about reopening, so to speak, are that we fear that reopening in a trickle wouldn't really work and

that we would quickly become overwhelmed. And I think the tentativeness over the medical side's capacity to care for sick people is really understandably a large part of the hesitancy in terms of opening up. I do think if we have really widespread testing, but we have to scale up testing really a lot, a lot more than we're doing, and if we have robust contact tracing in place, we could start reopening. The fear is that that brings us into a certain amounte of uncertainty, and none of us

like uncertainty. We're really fearful that our hospitals will be overwhelmed and that people will die unnecessarily. And that's a scary thought. But at some point, you know, my opinion is that we have to make a plan. I think. I think the biggest challenge right now is that I don't really see the plan. I haven't seen the plan. You have you seen the plan? The plan, you know, and if you haven't seen it and I haven't seen it, there is no plan. I mean, I think, Yeah, there's

a lot of magical thinking. There's a lot of thought about, well, maybe we could hold out until there's a vaccine, but there may not be a vaccine. There's a lot of thought of we have to hold out until the therapies get much better, but we don't have any guarantee that the therapies will work. And then the fallback is when people say herd immunity in some generic sense, as though they knew exactly what that meant. And that's obviously a scenario in which you will be very difficult to avoid

overwhelming the hospital. So I agree with you. So I have in my global health career, spent a lot of time responding to eyebreaks. You know, none like this but in disaster settings, you know, you have to move quickly and make decisions in the context of uncertainty. You know, in the US we are not as used to doing that because we have always really got the resources to do the modeling and get the science and wait until

it's perfect. And I think we have to move ahead with making decisions that are imperfect using the best data that we have. But we do have to use the data that we have. I mean, we have to move forward. So you know, technology and other ideas they may seem ambitious, contact tracing might seem oh, it's not feasible, it's too many, But the way to be successful is just to take the next step forward and believe it's possible and keep

building it as you're doing it. I think that's what we have to think about, or else, you know, we're just going to be struck. Thank you so much. This is tremendously clarifying, and there are bits of it that are inspiring and make one think there is a way forward, and then there are bits of it that cause one to feel a bit more panic. And I really appreciate it your honestly about it. Thank you all right, thank

you Here on deep background. We've spoken to several guests who focused on the mechan that we can use gradually to emerge from the conditions of social isolation that we're presently in. Testing has been a consistent theme that we've heard about again and again and again, more and better testing.

Speaking to doctor Ivers makes it clear that the testing is primarily valuable, however, to the extent that it is then used to lead to contact tracing, and the idea is that contact tracing itself can help manage the disease. This may be the only long run strategy that can be used in the absence of a vaccine and in the absence of better therapies. But contact tracing on its own is not a magic bullet. You have to reach

enough people and that's a challenge. It's also very difficult to make contact tracing work in a scenario where as we have, the disease has spread into the entire community and we're not just tracking down a handful of names. Contact tracing is going to be central to our public discussion in the weeks and months ahead, its successes, its limitations,

and the challenges that it faces. We will keep a close eye on this subject going forward and come back to you with more about contact tracing as this story develops. Until I speak to you next time, Be careful, be safe, and be well. Deep Background is brought to you by Pushkin Industries. Our producer is Lydia gene Coott, with research help from zooe Wynn. Mastering is by Jason Gambrell and

Martin Gonzalez. Our showrunner is Sophie mckibbon. Our theme music is composed by Luis GERA special thanks to the Pushkin Brass, Malcolm Gladwell, Jacob Weisberg, and Mia Lobel. I'm Noah Feldman. I also write a regular column from Bloomberg Opinion, which you can find at Bloomberg dot com slash Feldman. To discover Bloomberg's original slate of podcasts, go to Bloomberg dot com slash Podcasts. You can follow me on Twitter at Noah R. Feldman. This is Deep Background.