San Diego County’s Contact Tracing Struggles Explained
Tuesday, September 29, 2020
Photo by Elaine Thompson AP
Contact tracing was supposed to be one of the primary ways we could flatten the COVID-19 curve and reopen the economy. That hasn't really happened in San Diego County, despite the county employing hundreds of contact tracers and investing heavily in the program.
As a refresher, here's how contact tracing is supposed to work: When someone tests positive for COVID-19, a contact tracer gets in touch with that person's immediate family and everyone else he or she has been in close contact with. If one of those contacts tests positive, a contact tracer starts the process over again with that person's contacts.
The county has hired more than 500 contact tracers who are supposed to spend their days calling people who've tested positive for COVID-19, and then calling all of the people the sick person has been in contact with. Those people are told to get tested and stay home for 14 days.
But despite these plans, the county’s program seems to come up short. COVID continues to spread through our region, and contact tracers are only contacting on average 2.2 people for each person who tests positive. A new report from the Chicano Federation also finds that Latinx residents of San Diego County are hesitant to share information with contact tracers.
KPBS talked to UCSD epidemiologist Rebecca Fielding-Miller about why the county's contact tracing program has so many challenges. Her answers to our questions have been edited for length and clarity.
Q: We've heard how contact tracing is supposed to work. What can you tell us about how it's actually working?
A: I think the most important thing to know about contact tracing is it's really, really hard to do. You have to get somebody who has never heard of you and might have a few reasons not to trust you. You have to call them up out of the blue, potentially give them some pretty bad news, ask them a whole lot of personal questions and then get them to tell you everybody they've been around for the last 14 days. And then you have to get ahold of all of those people and do the same thing. So that's just hard any way you slice it. So we're doing better, but it's hard to be perfect.
Q: Why are we so bad at this? Is it underfunding or the distrust of government?
A: It's a big ask for you to pick up the phone for a number you've never seen before to talk to somebody who says, hey, I'm from the county. And first of all, I have some bad news. Second of all, I have a lot of personal questions. And third, can you please remember everybody you've been around for the last 14 days. There is an art to it and an art to doing it well. And if you think of it as a matter of scale, if you're following up on 100 cases a day and you're missing 3 percent, that's three people give or take. If you're following up on 200 cases and you miss 3 percent, that's six people. And so the more cases that have to be investigated today, the more people are going to fall through the cracks.
Q: Are there any model programs in the US who are doing it well? And if so, what are they doing that we should be learning from?
A: My understanding is Boston is doing pretty well, and the state of Massachusetts. And I think in part that's because they've contracted with Partners in Health, which is an organization that has a lot of work working in low resource settings, making sure that they're doing work that's trusted by the communities, being really aware of sort of local conditions on the ground.
We can look back to the Ebola outbreak where contact tracing was the primary strategy that we had for stopping that outbreak. And it turns out that contact tracers that are from local trusted community organizations are going to be a lot more successful than contact tracers that are from more centralized government entities. It's going to take longer, but you're going to get more people. And I think striking that balance of working with trusted community partners, but also having sort of vetted, centralized, well-trained folks is the important place to look.
Q: So we have some numbers from the county about contact tracing, interviews that I want to go over. They say they complete 76 percent of their interviews, which means when they're calling people who are have tested positive for COVID and asking who their contacts have been. But they say on average, they only get 2.2 contacts per positive case. So can you explain a little bit more what this means and what that says to you?
A: If you think about it, the folks who are at home all the time who leave to go grocery shopping, those folks are much less likely to be infected with COVID. It's the folks who are being exposed a lot. Let's say you work at a restaurant or you work in a pharmacy or a bar and you're coming in contact with a lot of people, your odds of being exposed are a lot higher. So it is probably pretty likely that the people who are sort of in this pool of folks the county is trying to reach, those people have more contacts than the average person in San Diego County, which is to say 2.2 is probably an undercount because imagine you wait tables, you might be able to say, well, here is the name and phone number of the three people who are on my shift.
But are you going to be able to give the name and phone number of everybody whose table you waited that day? Are you going to know the name and phone number of the person who was washing dishes in the back of the kitchen? So it's quite likely that you are getting the name and phone number of two people whose name and phone number of these contacts know, but there's probably a lot going unsaid there.
I think it's really important to remember that the potential issues with both who is at risk of getting sick and who is at risk of falling through the cracks for contact tracing, that's not random. So the people who are at risk of getting sick, people who are more likely to work service jobs, people who are less likely to have paid sick leave, people who are less likely to be able to isolate people who are more likely to live in multigenerational homes.
Those are the same people who, for example, might be undocumented, who might be somebody whose name you won't remember because they were your Lyft driver. Those are the same people who are not going to be caught by this contact tracing safety net.
And so what you really have is not sort of 4 percent of people in San Diego County chosen at random will get sick and half of those at random will fall through the cracks. You have this chunk of people are going to get sick and then those same sort of people are going to fall through the cracks. And you have this sort of cascade on the people who are most vulnerable over and over and over again.
Q: So it seems like we're all so concerned with the community outbreaks that the county announces every day. They say, you know, bars, restaurants, offices and so on. But then they say that that only accounts for 5 percent of all local cases. So how could that be?
A: So let's say there are two people who get sick in a gym. Two staff members in a gym get sick, they go home and four of their family members get sick. That would technically not be an outbreak since those are connected households and because it's pretty likely, since the majority of transmission is within households, that community outbreaks with people who don't live together would still be in the minority.
Q: Isn't it also if we're not effectively tracking who someone has been in contact with, then we're not finding out about more of these kinds of communities spread incidents?
A: Yes, so it's also quite likely, given the way that contact tracing works, that we're also just not catching the size of an outbreak. So imagine, for example, that to go back to the restaurant example, somebody in the kitchen is undocumented and you might not want to share their name and phone number.
Imagine you are an undergrad at SDSU or one of our many other institutions of higher learning and you were at a party where there was underage drinking, you are almost certainly not going to name people who were drinking because you don't want to get them in trouble or you don't know who they are.
Q: Contact tracing was supposed to be one of our big ways out of lockdown by keeping cases contained but is a poor contact tracing program actually worse than having none at all. So in other words, if county officials say only 5 percent of cases come from outbreaks based on a subpar contact tracing program, aren't we giving people a false sense of security?
No, I think it's really important to have that contact tracing system up and running for a couple of reasons. One is because there is an art to it and an art to getting people to share this information, we want the people in those jobs to be practiced and good at what they're doing. I think contact tracing sort of works like you can think of it like fighting a wildfire, for example. So when you have kind of a giant uncontained wildfire, which is sort of when case numbers are really high, you want to do everything you can to suppress that wildfire. And that's what lockdowns are for, asking people to shelter in place, this type of thing.
But you also want to make sure that there's not flare-ups. So when we are able to catch a potential flare-up in the community, whether it's at a restaurant or a child care setting or a gym, we can at least stop that from getting worse. And as we manage to bring the total case numbers down, as we manage to suppress the bigger wildfire situation, then contact tracing becomes even more important because we can stop those flare-ups before they start and the numbers get higher.
Listen to the Podcast Episode
Santa Ana winds are sweeping through parts of San Diego County as temperatures heat up amid low humidity, significantly raising the risk of wildfires throughout the region. Plus, Former aide to City Council President Georgette Gómez, Kelvin Barrios, is ending his campaign for her seat. Also, before they battled major coronavirus outbreaks, Imperial County’s skilled nursing facilities already had histories ... Read more →
Aired: September 29, 2020 | Transcript+ Subscribe to this podcast
To view PDF documents, Download Acrobat Reader.