How Do We Build Trust Into Contact Tracing?

JOHN DANKOSKY: This is Science Friday. I’m John Dankosky.

As some states send workers back into the world, many are asking, what’s next in the spread of coronavirus in the US? There’s still no vaccine or proven treatments, and reliable testing remains out of reach for many.

But in the meantime, there’s one intervention that has almost universal buy-in– contact tracing. That’s the process of identifying everyone a person with COVID has potentially infected, calling them all up, and then telling them to stay put so they don’t infect anyone. That’s for every single identified person with COVID.

Now, to this end the states are looking to hire human contact tracers, while companies like Apple and Google and others in tech are working on smartphone apps that could automatically notify people if they’re at risk. But what factors will determine the success of contact tracing? SciFri’s producer Christie Taylor talked to two public health experts about what makes this process so complicated.

CHRISTIE TAYLOR: Before you get too excited about an app for tracking COVID, public health researchers want to caution that there’s no true replacement for the human labor of contact tracing. It’s a difficult, labor intensive process. And different estimates point to a need for as many as 100,000 to 300,000 contact tracers in the country.

Joining me today is Dr. Crystal Watson, a senior scholar at Johns Hopkins University’s Center for Health Security and lead author of an April report from the center outlining a contact tracing plan for the country, and Dr. Louise Ivers, executive director of Massachusetts General Hospital Center for Global Health and associate professor of medicine at Harvard Medical School.

Welcome to Science Friday both of you.

LOUISE IVERS: Thank you.

CRYSTAL WATSON: Thanks very much.

CHRISTIE TAYLOR: This phrase contact tracing has been in the news a lot. It’s been part of a lot of public health discussions. At its simplest, what is it, Dr. Watson, and how does it help slow the spread of a disease?

CRYSTAL WATSON: So contact tracing is really intended to break chains of transmission among people. So the idea is that when someone becomes sick, they are quickly isolated, either in their home or in a health care facility, if that’s needed, if they need medical care. And then a public health worker will ask them about who they’ve had contact with during the time they may have been infectious with the SARS-CoV-2 virus.

Once they have identified the contacts, which could be from one to three days before they develop symptoms to up to seven days after symptoms have resolved, then those contacts are contacted themselves. Public health workers get in touch with them and ask them to stay at home and quarantine for 14 days. If they can’t stay at home and quarantine for 14 days, then they offer them another facility to quarantine from safely. But the idea is that they are then staying out of circulation in the community. And if they are indeed infected, then hopefully they won’t be passing along that infection to other people.

CHRISTIE TAYLOR: Dr. Ivers, we just talked about that definition of contact tracing. And to my ears, it sounds almost deceptively simple. You find people. You talk to them, then you talk to more people. You’ve personally done this with a lot of other outbreaks. What makes it actually complicated in real time?

LOUISE IVERS: Yeah. I think it’s complicated only by the fact that humans are complicated people. Contact tracing really, when it’s successful, is very people-centered. It requires establishing trust. It has to be an activity that people understand is confidential. You have to make it very accessible. When you talk to a person who has an infection, they are obviously going through their own health issues. And they have needs that we need to be able to address and manage and assess.

And the same is true when speaking to contacts. Contacts who’ve been exposed to an illness might be fearful. They might be concerned. As Dr. Watson said, they may need support to be able to isolate themselves– put themselves in quarantine.

They may have other worries. It may be a crisis for them. So there are many pieces. There are many human factors to it. That’s one side.

On the other side, when we look at an epidemic that’s very fast growing and that has a very high reproductive number, we have to be able to identify and very quickly isolate contacts in order to be effective. And so one of the challenges of the human endeavor around contact tracing is that it’s resource intense. You need a lot of people.

It can be slow, which is sometimes OK if the disease is slow moving. But it can also be incomplete. So everyone does not necessarily remember who the people were that they were in contact with or may have exposed. At the moment that is probably not a huge problem, because so many of us are socially distancing. But as we reenter more normal society, when we take the bus, or an airplane, or we go to public spaces, the human memory of who you’ve been around may be more incomplete. So those are some of the challenges to contact tracing in the context of this particular infectious disease outbreak.

CHRISTIE TAYLOR: Contact tracing has been used to combat outbreaks of HIV, Ebola, a lot of other infections around the world. Have we learned anything about the right and the wrong ways to go about contact tracing from these decades of experience?

LOUISE IVERS: I think what’s important to highlight is perhaps as you said earlier. Ordinary people are learning a lot about contact tracing now. It’s a new concept for many people, but it’s certainly not new to public health. And public health experts, community health workers, disease investigation specialists, people have been doing this for a very long time. And they understand its complexity.

And so certainly I think a way to not do things right would be to kind of erase that lived experience, and scientific experience, and knowledge, and try to jump over it, somehow thinking that technology could just no longer need the human factor. So I think knowing how to do contact tracing right just highlights some of the things we spoke about already, about remembering that people are involved, remembering that outbreaks are not just scientific events. They’re human events where infections are in social context. There’s a social science to this as well. And so it’s important to meet people where they’re at, to adapt your efforts to the local community, and to get at some of that kind of cultural humility.

CHRISTIE TAYLOR: So Google and Apple have actually teamed up to offer this Bluetooth-based solution. People’s smartphones will tell each other when they’ve been in contact with someone who has a confirmed or suspected case. Other people are working on smartphone apps. Crystal, your report has called for technology to be part of the conversation. Where does it actually fit into the Johns Hopkins vision at this point?

CRYSTAL WATSON: Yeah. I think it has the potential to be helpful. It has the potential to be a workforce multiplier for those large numbers of contact tracers that we need across the country.

I do think that it can help particularly in a couple of aspects. The first is identifying additional contacts that someone who’s sick with the disease may not remember they were in contact with, or may not know those people if they were out in a public setting. I think it can be helpful with that.

It can also be helpful with identifying and contacting those contacts more quickly. This virus moves very fast. The illness progresses quickly, and the virus spreads quickly. So the faster we can notify contacts and ask them to quarantine safely at home, the less spread we’re going to have. So by addressing those two issues, I do think technologies can be helpful.

CHRISTIE TAYLOR: And Louise, you’re working with the team at MIT that’s building rules or protocols around how to do this smartphone tracing without violating privacy. Why is that part so important?

LOUISE IVERS: Well, I think there’s two– so two things maybe to highlight around this work at MIT that I’ve been involved with. One is actually that the Apple and Google approach right now, which use Bluetooth energy to try to identify when phones have been in proximity to each other.

And so I think the first step is trying to really identify the efficacy of that approach. It’s not 100% certain yet that the information that would be made available by that kind of technology is in the range of being medically relevant or public health relevant. So it’s important to try to understand the confidence with which we might be able to alert people so that we don’t have too little of an alert, so that we’re detecting the right amount, and that we wouldn’t have a false alarm.

The other piece of it, though, is really, as you said, centered around privacy. It’s interesting, because from a public health perspective, public health has authorities to know confidential information in important outbreaks like this. And yes, it’s just that by establishing confidence, by being a trusted body. If we were to enable a technology that is pervasive and held in the private sector to collect large volumes of information about who we associate with, who we’re in contact with, or where we go, or other things, it certainly could be very concerning to collect all of that information in a database.

So the group at MIT is led by professor Ron Rivest, who is a cryptographer, and who really centers his ideas around the protocol of using Bluetooth, for example, around the idea that it could be done in a way that was private and protected the privacy of individuals owning phones.

CHRISTIE TAYLOR: I mean, is it possible to go too far in the wrong direction in preserving privacy? Does the Apple-Google partnership really give public health workers enough to work with? Crystal, what is your privacy/information balance ideal?

CRYSTAL WATSON: So privacy is obviously very important to all of us. My bias is slightly on the side of public health, where public health officials and contact tracers really do need some more granular data to be able to identify cases and contacts, but most importantly to support them. We’ve talked about enabling isolation and quarantine. That’s providing supplies, support for family members, whatever it takes to have those people be able to stay at home, or stay isolated, or quarantine safely. And so if public health doesn’t have information that can identify people, then they cannot provide that service, which is, I think, core to this epidemic response.

So I do think they need some information. I think it can be opt-in. This is very core to public health. They collect sensitive data routinely. And they know how to handle it. And as long as they have the trust of the community, I think this is the balance that needs to be struck, that public health needs enough information to take action and support people throughout this pandemic.

CHRISTIE TAYLOR: Trust keeps coming up in this conversation about public health and the ways in which public health efforts rely on trust. I know there was a study out of Oxford that suggested that we need something like 60% of a population to use something like a contact tracing app if it’s going to be successful and giving meaningful information. Louise, I mean, as someone who, again, has done this contact tracing in person, how do you see trust working into a more electronic version of that?

LOUISE IVERS: The trust will have to come when the public health authorities and officials find a technology that meets their needs, which I think relies on the efficacy of the technology to detect contacts and a low rate of false alarms, establishes a privacy balance that’s appropriate, and that’s integrated into their workflows. And then public health can be connected into specific applications, and then asking the citizens using the trust they have.

And again, public health authorities do this. This is what they’re trained for. This is what they have experience doing. So the integration with that is the most important piece.

CHRISTIE TAYLOR: Just a quick reminder, I’m Christie Taylor, and this is Science Friday from WNYC Studios, talking to Dr. Crystal Watson and Dr. Louise Ivers about public health, contact tracing, and the various solutions for doing so in coronavirus times.

What about testing? Does contact tracing also needs us to be testing people at a certain threshold?

CRYSTAL WATSON: In my opinion, I think we need to be testing anyone with COVID-19-like symptoms. Anything on top of that would obviously be an improvement, but I think that is the bare minimum. We need to be able to identify cases, symptomatic cases, and quarantine their contacts. Then if those contacts develop the disease, whether they’re symptomatic or asymptomatic, they’ll be quarantined at home. They won’t be out in circulation. So that helps address the asymptomatic and presymptomatic transmission issue that we have here, which is very complicating for containing this disease. So I do think that’s the bare minimum.

One of the problems we’re having with testing right now, in addition to access, is that the testing is not very timely. So by the time a positive test comes back, it may have been three, four, or five days since that person was tested. And if we wait to identify contacts, and ask them to quarantine, and tell that positive test, many of them may have already become infectious and gone out to infect others. So the timeliness of the tests is very important, too.

CHRISTIE TAYLOR: So we’re talking then about this human army of contact tracers, 100,000 to 300,000. Is it as simple as hiring them and training them? Or is there more to this picture of building a contact tracing network beyond that?

CRYSTAL WATSON: I think it’s going to be a complex and a resource intensive undertaking. We do need to hire people. We need to hire people quickly. But obviously, they need to be trained very well. We need to hire the right people who have the skills to be very compassionate and to preserve privacy. They’re good people people.

But they also have great fidelity with tracking and recording data, and making sure that that data is maintained privately. So there are key skills that we need to look for.

And then we also need to manage this workforce as well. This is a lot of people that we’re going to be hiring to do this work. And so there is going to need to be good management, good strategy, good technical support. And that’s– hopefully, we can look to the federal level to the CDC to provide some of that support.

CHRISTIE TAYLOR: Louise.

LOUISE IVERS: One of the complexities is the– in Massachusetts, for example, there are 351 boards of health who do contact tracing. So we didn’t really touch on how that adds to the nuances of it, that you have 351 different groups of people doing this in Massachusetts alone. Because in some ways, I think, when you think, “This seems simple. Why don’t we just do it?”, there are a number of challenges to the implementation of it. But that does not mean that we could not do it.

CRYSTAL WATSON: Just to respond to that, our call for contact tracing at a national level really is for support and kind of a vision for contact tracing. We were really– we really wrote our report hoping to get the support at the federal level, both in terms of funding, and guidance, and training, and technical support, but it’s definitely not realistic to have one contact tracing army that’s led at the federal level. That’s just not how our public health system works.

CHRISTIE TAYLOR: Anything else that you hope people keep in mind as we’re embarking on this effort, Louise?

LOUISE IVERS: There is no one single silver bullets to respond to this pandemic. We need comprehensive integrated approaches to outbreak management and control. And so it’s really important for any technology that comes about to realize that it’s going to have to be integrated into that comprehensive approach. And we talked about testing, and contact tracing, identifying clusters, social supports for people who are able stay at home.

So I think the biggest message is that we have to be ambitious about different approaches. We certainly can do it, but there’s no single one thing, I think, that’s going to get us there. We have to make sure that we’re trying to do things in a comprehensive way.

CHRISTIE TAYLOR: Well, thank you both so much for joining me. Dr. Crystal Watson is a senior scholar at Johns Hopkins University’s Center for Health Security, and Dr. Louise Ivers, executive director of Massachusetts General Hospital Center for Global Health.

CRYSTAL WATSON: Thank you very much.

LOUISE IVERS: Thank you.

Copyright © 2020 Science Friday Initiative. All rights reserved. Science Friday transcripts are produced on a tight deadline by 3Play Media. Fidelity to the original aired/published audio or video file might vary, and text might be updated or amended in the future. For the authoritative record of Science Friday’s programming, please visit the original aired/published recording. For terms of use and more information, visit our policies pages at http://www.sciencefriday.com/about/policies/