Contact Tracing Is Having a Trust Crisis
Why won’t Americans who test positive for Covid-19 share the names of those they’ve been in contact with? Blame fear, confusion — and politics.
Most of the people Nasrin Sultana calls each week don’t answer on the first try. The ones who do are often skeptical when they pick up the phone.
Some think Sultana is trying to steal their identity. Even after she insists that the only personal information she needs is their name and date of birth, not their Social Security number or their bank account info, they’re wary of divulging.
Sultana is not a scammer. She’s a volunteer contact tracer for Florida’s Larkin Community Hospital. On top of her clinical assistant and coronavirus research work, she spends 20 hours a week calling people who have tested positive for Covid-19 at the South Miami hospital or have come into close contact with someone who has. She wants to ask them a series of questions about their health, their living situations, and who they’ve recently had prolonged contact with.
But of the three or four people on her list to contact every day, she says she typically reaches only one of them; based on a survey of a subset of her fellow contact tracing specialists, the group reaches an average of 40-52% a day total. The others she texts, emails, or leaves a voicemail through Larkin’s Covid Fighter App. There are plenty of people to call: Florida has passed 80,000 cases this month.
Every person she doesn’t reach, or who doesn’t believe what she’s saying, may be spreading coronavirus to someone else. “We can’t force them to participate or give us the information,” she said. “It’s voluntary.”
It’s also critical to quelling infections, public health experts say. After the blunt measures of lockdowns are lifted — a step experts say should only happen after case counts have steadily declined for three weeks and less than 10% of daily tests come back positive — contact tracers target individuals for isolation, sniff out the sources of infection clusters, and intervene before they become full-blown outbreaks.
That’s how it’s worked in countries like South Korea and Taiwan, at least. In the U.S., contact tracing is up against a lot: the national tracing workforce is experiencing massive shortfalls, and the funding required to hire and train has not matched the need. Spotty access to coronavirus testing and week-long waits for results mean that by the time a patient knows they have the virus, it may be too late to stop potential contacts from multiplying. Many states rushed to lift lockdowns prematurely, leading to a fresh surge in cases to track. All these limitations are being compounded by the fact that contact tracers like Sultana are finding it harder than expected to get people to answer their calls, share the names of their close contacts or comply with their instructions.
Public health is called public health for a reason, says Emily Gurley, an infectious disease expert at the Johns Hopkins School of Public Health: Strategies like contact tracing can only be effective if the community cooperates.
However, there’s no one metric by which to judge a successful contact tracing effort, says Gurley. “You don’t have to be perfect to have an impact,” she said. “And just because cases are still going up doesn’t mean you’re not having an impact, as well.”
If a community tests only a few people, but reaches 100% of respondents, they may be less effective at stopping transmission than a community that identifies all the local cases through testing but only reaches 40% of them by phone. And response rates don’t reflect the efficacy of each interview. Picking up the phone is one thing; giving out a list of close contacts is another.
But contact tracers do need all the responses they can get, and, crucially, as many honest, comprehensive interviews as possible. To safely lift lockdowns, the COVID-Local guide recommends cities ensure they’re reaching 75% of each positive patient’s close contacts within a day of testing.
Right now, Miami isn’t the only one struggling to reach that goal. Despite national polling in May that showed the majority of Americans would cooperate with contact tracers, by August, the reality has shaken out differently. Miami Beach Mayor Dan Gelber told CNN’s Anderson Cooper in July that the Miami-Dade county health department was only reaching an average of 18% of positive cases. In D.C., meanwhile, the Washington Post reported that tracers reached about 60% on their first tries. States and cities that started ramping up tracing earlier are showing improvement. In Louisiana, contact tracers have upped response rates from 48% in early June to nearly 70% in July, NOLA.com reported — but 73% of those they reach won’t share their close contacts. Though New York City’s tracing program has been criticized for a chaotic rollout, the New York Times reported that the city’s Test and Trace Corps has interviewed 64% of the city’s nearly 20,000 positive cases in June and July. Less promising is the proportion of people who have gone on to share their close contacts, however: 35% in June, and 42% by July.
“I’ll admit that it’s hard for me to understand why people are so resistant,” said Ed Salsberg, the lead research scientist at the George Washington University School of Public Health and Health Services. He had hoped that “most people would want to know they’ve been exposed, that they would want to know if they were at risk personally, if they were at risk of infecting their family and friends.”
Like mask-wearing, which became fiercely politicized despite evidence that it significantly reduces transmission of disease, attitudes towards contact tracing have been muddied by misinformation and mixed messaging.
“The one thing we hear that’s the most distressing from our viewpoint is that contact tracers report abusive behavior towards them,” said Steve Waters, whose company, Contrace, helps connect qualified contact tracers with health departments to work in. “They’re being told Covid doesn’t exist, it’s a conspiracy, it’s government mind control, or whatever.”
This resistance varies by region and by state, Waters says, but it tends to line up with political differences. “A large proportion of the population views not participating in contact tracing as aligned with their political tribe,” he said on a panel at the Disease Prevention and Control Summit last month.
But contact tracing pushback isn’t just about politics. Even as coronavirus disproportionately impacts communities of color, trust in the federal government is lowest among Black people, and reports show that some Latinx communities are wary of getting tested for the virus and more fearful of seeking medical help because of fears of immigration enforcement.
Guilt, grief or fear can also make people clam up. Mayor Tony Perry of Middletown, New Jersey, told local reporters that teens who have attended parties where coronavirus was spread have evaded contact tracers’ questions, worried they’d be perceived as “ratting” on their friends; even parents hesitated to be honest, worried that their children would get in trouble for underage drinking. Dr. Val Arkoosh of Montgomery, Pennsylvania’s Board of Commissioners told a local CBS affiliate that 20% of people did not cooperate with tracers because they were afraid “they may be blamed for potentially exposing someone else. Some are worried their employer might take action against them.”
In times of crisis, human memory can be fallible. “Many patients are emotionally overwhelmed, and they don’t remember,” said Andrea Crisanti, a professor of molecular parasitology at Imperial College London, at the Disease Prevention and Control Summit panel. “Maybe they don’t remember, [or] they don’t want to remember. … Although they want to collaborate, sometimes they really cannot.”
And until testing improves, Gurley says the biggest barrier to contact tracing will be diagnosing people in a time frame that’s useful. In Houston, Texas, KTRK-TV’s 13 Investigates found that in one week in July, tracers were only able to interview 38% of positive cases, far below their goal of 85%. More troubling, however, was the fact that within the previous two weeks, 36% of cases “weren’t traced at all because the results were too old.” That’s down slightly from mid-July, when more than half of cases were too dated to be useful.
Already, one promising contact tracing tool in the U.S. has been stymied by privacy fears. In April, Apple and Google introduced a plan to facilitate contact tracing by allowing people who tested positive to send out an anonymous ping via the Bluetooth connection that most smartphones have, to alert anyone who had come into close contact with their device in the previous few days. Despite privacy and anonymity assurances, surveys showed that people were wary of their location being linked to their health information.
Forthcoming research from Sarah Kreps, a professor at Cornell University and an expert in international politics and technology, helps explain that resistance. She and two co-authors, Baobao Zhang and Nina McMurry, surveyed about 2,000 people on their level of support for various coronavirus monitoring policies. They found that when it comes to expanding traditional contact tracing measures, Republicans were less supportive than Democrats. But when respondents were asked about their willingness to download contact tracing apps, there was no significant partisan difference. In the paper, currently released as a pre-print, the team wrote that this seems to suggest that “the issue of contact tracing is not necessarily ideological or has not yet been politicized, creating opportunities for bipartisan elite cues that mobilize constituents to opt-in.” But the lack of politicization masked the reality that both sides share a baseline level of skepticism about contact tracing apps — just for different reasons.
“Republicans tend to be more wary of Big Government, and Democrats of Big Tech,” Kreps told CityLab. “There’s a lot of baggage on both sides of the aisle when it comes to surveillance.”
Independents, meanwhile, were generally wary of perceived liberties violations, and were less likely to support both traditional and technical contact tracing.
“The government doesn’t need to be tracking anyone,” wrote one Republican respondent. “They could not even handle things so far and they want to trample on people's rights even more. No thank you.” Another said, “There’s no need. The virus is a danger no matter what. If I'm going to get sick I’d rather be living life with personal freedom.”
A surveyed Democrat, meanwhile, said that “This would give way too much information to tech companies and the government.”
Some of the resistance was driven by confusion and misunderstanding about the underlying technology. One Democrat respondent said that they didn’t have a “blue tooth,” and would refuse to install one for contact tracing purposes; a Republican said that they’d leave their phone at home to avoid tracking. “Then I wouldn’t have one in case of an emergency,” they wrote. (Individuals would have to download tracing apps and opt-in to the feature before being tracked.)
Fears that contact tracing as a practice goes “too far” also reflected an information gap. “If people reasonably [think] they are sick, they should have easy access to testing, not have to say every single person they come into contact with,” wrote one Independent. “Some people would define in-person contact different than other people.” Contact tracers only need to know about your “close contacts,” who are generally defined as someone you’ve been with for more than 10 to 15 minutes at a time.
In part because health officials understand this reticence — and in part because of coordination and technical issues — the U.S. has yet to roll out widespread app-based testing, as many other nations in Europe and Asia have, and the Apple and Google API has yet to be adopted at scale. Even officials from Singapore and Iceland, where sophisticated contact tracing apps were part of an early coronavirus response, advocate a hybrid approach. Pairing targeted phone alerts with traditional human “disease detectives” who call on the phone can capture the widest swath of the population, they say.
“It’s not too late for public health authorities to communicate better about this tech, what it entails and why it’s important,” said Kreps.

But trust on the ground has to be earned, too, says Sandra Mullin, the senior vice president of policy, advocacy and communication at the global health organization Vital Strategies. Her team ran focus groups with African Americans, other people of African descent, and Latinx people in New York and Philadelphia to understand what messaging tactics might work best for eliciting responses. She found that people were more likely to trust contact tracers who look like them, come from their communities, and speak their language.
“It's really tough to build trust in a crisis,” she said. “Trust is best built before a crisis.” She encourages public health departments to lean on voices that people know already. Pastors could send out flyers encouraging their congregation to cooperate with contact tracers. City leaders, who are generally more trusted than national leaders, could run PSAs.
Meanwhile, to assuage privacy concerns, perhaps cities could coin a new term for contact tracing teams, like “the Covid Corps or the Virus-Prevention Corps” — de-emphasizing the “contact” question — Salsberg suggested. That’s what Baltimore did when it launched a city program of community health workers, dubbed the Baltimore Health Corps, to assist in Covid-19 tracing.
Sultana, the contact tracer for Florida’s Larkin Community Hospital, says that there’s a life-saving value in the tracing process, even if people aren’t willing to share their contacts. She makes sure to give them information they need to protect themselves and their family by self-isolating and helping them link up with resources. Many cities and states offer food and groceries for those who have to quarantine themselves after exposure. But for essential workers, especially those making hourly wages, missing up to two weeks of work can be devastating, and a care package isn’t enough to incentivize isolation. In San Francisco, where Latino residents are three times more likely to contract the virus than white residents, a Latino Task Force has developed a wage replacement program for people who are sick, so they don’t avoid testing, the L.A. Times reported.
“I think we need to be able to provide some support for people to stay home,” said Gurley. “Monetary support.”
Linking tracing with such direct benefits could help health experts revise their messaging and reverse the erosion of government trust. And as the pandemic rages on, public opinion may eventually come around to the idea that we already possess a powerful weapon against the disease — if enough people agree to use it. “We may do better by clarifying that contact tracing is really for the individuals’ benefit,” said Salsberg. “It really needs to be sold as something you’re doing to help yourself and your loved ones. This isn’t for the government.”