Early on in the Covid-19 pandemic, the San Francisco Bay Area was one of the hot spots. The region moved quickly into containment mode: Businesses shut, travel was restricted, city officials issued orders to stay home. And by most measures, it seemed to work. An outbreak that would overwhelm local hospitals never materialized. But rather than stamp out the virus, after the first month or two the number of cases instead fell to a low, uneasy sizzle. The virus, despite extraordinary measures to contain it, was still circulating out in the community, somewhere.
Diane Havlir, chief of the HIV division at UC San Francisco, had a hunch about where it might be lurking. Around the country, communities were beginning to show wide disparities in who was being affected by Covid-19. Infections had been tied to essential workers—doctors and nurses, of course, but also people who work in meatpacking plants or as bus drivers, grocery cashiers, and other jobs that require a lot of contact with people. It was also becoming clear that the virus was disproportionately killing Black and Latinx people. In San Francisco itself, more than 80 percent of Covid-19 patients at the city’s public hospital were Latinx. So Havlir and a team of researchers decided to do a local investigation. They proposed an experiment: Pick a neighborhood, then test everyone, symptoms or not. “We wanted to know: Who are we missing?” she says.
They chose a census district in the Mission neighborhood—a largely Latinx community with a known high rate of Covid-19 infections—and partnered with community organizations there to set up testing. “At first, there was a lot of skepticism, because it’s a study,” recalls Susana Rojas, executive director of Calle 24, one of the partners. “How are you going to use my information? What are you going to do to our community?” So researchers, community workers, and volunteers went door to door, hoping to explain their intentions and build trust. And on a bright weekend in late April, about 4,000 local residents and workers showed up in neighborhood parks and soccer fields to receive a swab up the nose to test for the genetic traces of an active Covid-19 infection, plus a finger prick to look for antibodies that would reveal a prior one.
The initial results, released last month, pointed to a blunt conclusion: “It is essentially the low-wage essential workers that are bearing the burden of this disease,” Havlir says. The researchers found that about 2 percent of the people they tested had an active Covid-19 infection—there’s that low background sizzle. And within that group, the results were stark: 96 percent were Hispanic, even though they represented only 40 percent of the study participants, and 90 percent said they were unable to shelter in place at home. Many were service industry workers: delivery drivers, cashiers. Others worked in construction, which had recently restarted in San Francisco. Among white residents, there was one positive test. In a part of the city known for its rapid gentrification, the virus had made plain the divides between neighbors.
But that data offered just a snapshot—the state of infection on the weekend testing occurred. Now a full analysis of the data, posted as a preprint Wednesday, presents a clearer picture of how Covid-19 infections changed as the shelter-in-place period progressed. The data now includes people whose antibody tests revealed that they’d recovered from an infection prior to that weekend flurry of testing. Antibody tests were positive for 3.1 percent of the census tract residents and 7.7 percent of people who worked there. (Overall, the researchers estimate 6 percent of residents had been infected by late April.) That group, the researchers found, was more diverse.These people were more likely to be white and have higher incomes; about half said they had been able to shelter at home comfortably. There are gaps in that data—an antibody test can’t tell you precisely when those infections took hold, only that they occurred. So it’s unclear if a given infection happened before or after shelter-in-place rules began. But to the researchers, the absence of those more privileged people in the pool of people with more recent infections underlined their conclusion: Shelter-in-place was working—for those with the luxury to stay home.
Among hard-hit essential workers, the precise dynamics of how infections spread are hard to pin down, Havlir notes. Had transmission actually occurred at work? Or maybe on transit? Perhaps the disparity reflected differences at home. Essential workers, she notes, are more likely to live in crowded conditions or lack housing security or live with other essential workers subjected to potential exposure. “San Francisco is complicated. There’s a lot of couch surfing,” Havlir says.
To attempt to decipher the answer, the researchers turned to genetic techniques. In the lab of Joe DeRisi, co-president of the Chan-Zuckerberg Biohub, a research center affiliated with UCSF, researchers sequenced viral RNA from 49 of the positive samples. As the virus invades cells and replicates, it picks up mutations over time. Similar mutations in the same parts of the virus’s genetic code mean two infections are related more closely: It’s thus possible to infer if the virus spread locally, or if it came from a different introduction, such as a traveler coming from abroad. “You can drill down and try to block transmission chains,” DeRisi says.
They found that the sequences, based on their mutations, fell into five distinct groups—a diverse bunch, but nothing unexpected based on previous cases in San Francisco. That told the researchers that the workers hadn’t likely been infected from the same superspreading event—say, a single workplace they all shared. Instead, they saw that people with similar viral strains tended to be clustered together in households. That told them that these workers had likely picked up the virus somewhere in the community and brought the infection home, where it spread.
In addition, more than half of those who tested positive were asymptomatic at the time of testing, meaning they didn’t realize they were potentially infectious. That presents a challenge for curbing more infections, because people who don't realize they are sick are unlikely to be tested—and even less likely to pursue it if being asked to quarantine would cause them financial hardship. “What you would hear from the community is, ‘I can’t get tested, because if I’m positive, I have to work. This is the only way to feed my family,’” says Hillary Ronen, who represents the neighborhood on the San Francisco Board of Supervisors.
The testing team tried an approach Havlir terms “test and support”—a riff on “test and isolate,” which refers to quarantining positive cases. Contact tracers at the San Francisco Department of Health would follow up with those who tested positive, as usual, but then, when appropriate, they would refer individuals to a team of wellness workers who live in the neighborhood. The wellness workers would call to check in, offering help like cleaning supplies, gift cards for food, and even housing in a hotel room, for those who don’t have space to quarantine themselves within their household. This outreach didn’t require only Spanish language skills, Rojas says, but also trust that only neighbors could provide. “Most houses would be like, ‘No, I think we’re OK,’” she says. “But it was because they were shy, and they didn’t want to take advantage.” Getting some people to accept help took persistence, she says—a call every day.
In one case, Rojas recalls, an older man who had tested positive feared for the health of his wife and daughter, who is immunocompromised. He told a contact tracer from the health department that he had been able to isolate himself in the home of a friend who was temporarily living in London. “London?” The wellness team was skeptical, Rojas says. In follow-up calls, they eventually learned that he was squatting in an abandoned building near his family. They convinced him to accept a hotel room so he could recover safely. “You can’t take the first no for an answer,” Rojas says.
In response to the study, city officials plan to extend their financial support more broadly. Starting next week, San Francisco will begin distributing $1 million of a Right to Recover fund to replace wages for workers who need to quarantine, including for undocumented workers who have been shut out of government help. “The top priority is to make it possible for people to be able to quarantine and feel they have that option if they test positive,” Ronen says.
Even with many financial barriers removed, Rojas expects challenges will remain in getting support to the most at-risk people, like essential workers, and communicating to them how and why they should get tested even if they feel fine. “It’s about helping people understand the effect that it could have on the community and in their family,” Rojas says. “You may not have symptoms, but what if your mom gets sick? Your grandma?” With that perspective, she says, people generally become far more receptive to testing.
As the country continues to reopen, a broader part of the population will be exposed to the virus. “Going forward with that requires the maximum amount of data,” Havlir says. In late May, the researchers conducted tests for a similar study in a nearby neighborhood, the Bayview, an area with a large black and Asian population, and with higher-than-average rates of infection. Rojas says they hope to test again in the Mission in a few months.
Havlir acknowledges that a universal testing model may be difficult to export outside of San Francisco—a city with big academic labs with enough testing capacity to take on research. But the support strategies being piloted in her city are certainly portable. To stamp out the virus, it’s not just about finding out who has it. It’s about supporting people who do, so they can stay home until they recover. “People should not be penalized for following health guidance,” Havlir says.
Update 6-22-2020 1:32 PM EST: This story was updated to correct the estimated prevalence of Covid-19 infections among residents and workers in the Mission District of San Francisco.
- “Let’s save some lives”: A doctor’s journey into the pandemic
- Inside the early days of China’s coronavirus coverup
- An oral history of the day everything changed
- How is the coronavirus pandemic affecting climate change?
- FAQs and your guide to all things Covid-19
- Read all of our coronavirus coverage here
