On March 10th, the day before Tom Hanks announced that he had tested positive for COVID-19, I drove my three-year-old daughter to her day care in Berkeley, California. The drive took us up College Avenue, where white-haired professors huddled at sidewalk café tables; we passed the fraternity houses where students gathered on the lawns. Just a few blocks from my daughter’s school, there’s a coffee shop with a clientele split equally between students and senior citizens. I sat there until noon, working on my laptop, before moving to one of the food courts frequented by international students from Asia; a few weeks later, many of them would be returning from winter trips home.
Roughly a hundred and twenty thousand people live in Berkeley—about twelve thousand per square mile. The city is about three times as dense as Houston, and roughly as crowded as Washington, D.C., Philadelphia, or Chicago. About fourteen per cent of its population counts as elderly; those seniors interact to an unusual degree with college students and travellers from Asia and Europe. Given all this, it was reasonable to guess in the early spring that Berkeley would soon be hit badly by the coronavirus. The same was true for San Francisco, which is just across the water from Berkeley and is the population center of the Bay Area, a nine-county region stretching from the farms and vineyards of Napa and Sonoma, in the north, to tech-centric San Mateo and Santa Clara Counties, in the south. San Francisco is a major tourist destination; the universities in the city and surrounding it are home to tens of thousands of international students, who crowd into classrooms and dormitories, swapping droplets. Tech workers often travel to and from San Francisco and its neighbors—Palo Alto, San Mateo—using the region’s heavily trafficked public-transportation system; they are served by hundreds of thousands of immigrants who live in multigenerational homes or unofficial workers’ dormitories. The wealth gap in the Bay Area has created an acute housing crisis, with thousands of destitute people living in tents and R.V.s, in temporary villages near or beneath city freeways. A viral spike in San Francisco, Berkeley, East Palo Alto, Oakland, or any of the other dense communities at the heart of the Bay Area seemed nearly unavoidable, and would have had disastrous consequences for the whole region.
We now know that some of the earliest coronavirus infections in the United States happened in Santa Clara County, home to Stanford University: the virus killed at least two county residents in February, well before the first official American death was recorded. And yet the bomb didn’t detonate. By January 1st, San Francisco, a city of almost a million people, had seen just a hundred and eighty-nine deaths. In Berkeley, only twelve residents had died. Although the region as a whole has struggled with the virus—during the current winter surge, many of its hospitals have come close to, or reached, their regular I.C.U. capacity—it has still done better than many parts of California, and better than much of the nation. Some eight million people live in the Bay Area; just more than twenty-six hundred have died of COVID-19. Roughly the same number of people live in New York City, where, by the end of the year, there were more than twenty-five thousand deaths. In Los Angeles County, which is home to ten million people, four times as many have died as in the Bay Area. Even now, when the virus is less controlled and is spreading throughout urban, exurban, and rural communities, the number of cases in the densest Bay Area cities remains relatively low.
Credit for this outcome has tended to go to three groups of people: local officials, who acted with admirable synchrony and speed in issuing shelter-in-place orders; tech C.E.O.s, who were quick to move their businesses to all-remote operation; and the region’s citizens, who have, by and large, been willing and able to go along with restrictions. And yet their actions in themselves can’t wholly account for the relative success of the Bay Area’s response. “The narrative we like is that the citizens were all wonderful and informed and our leaders were all wonderful,” Bob Wachter, the chair of the Department of Medicine at the University of California, San Francisco, told me. “That might make us feel good, and part of that is absolutely true. But then you look anywhere else on the map, and you see places where they did something completely different and had a similar outcome—or you look somewhere that did pretty well and got overrun.” Boston, another college town with a highly educated population, shut down at roughly the same time as San Francisco. It also has a world-class health system, a smaller population, and less international traffic from Asia and Europe. Even so, it has experienced seven times as many deaths.
In trying to understand the pandemic, we gravitate toward general truths: if everyone wears masks and follows distancing protocols, viral spread will slow. That’s true; our practices matter. And yet the prevalence of the virus varies block by block, and some blocks are more vulnerable than others. To truly keep infections low, a pandemic response must find and help those people who are, for whatever reason, most likely to get infected. “You have to look for the virus,” Ranu Dhillon, a researcher and physician in the division of Global Health Equity at Brigham and Women’s Hospital, in Boston, told me. Dhillon has helped the governments of Guinea, Rwanda, Liberia, and Nigeria create health-care systems that address infectious diseases, including malaria and Ebola; he is based in the Bay Area and has been working in the emergency rooms of a handful of its hospitals since before the American pandemic began. In July, he gave me a sense of the communities in which the virus was spreading. “Every COVID-positive patient I’ve seen at the hospitals, with one exception, has been Latino or African-American,” he said. “The one white person was an older gentleman whose family member worked in a prison. . . . Everyone else, every single person, works at a food-processing plant in Vallejo, or they work at that Windsor nursing home that had the huge outbreak”—a facility, also in Vallejo, where sixteen residents died—“or they live with someone who works at the food-processing plant or at the nursing home.”
Containing the virus in such places isn’t just a matter of strategy. It requires considerable public-health capacity. Health officials must find and communicate with vulnerable, hard-to-reach people; those individuals, in turn, must be aided in taking steps to protect themselves. Today, we associate the San Francisco Bay Area with the disembodied ethos of Silicon Valley, libertarian and technocratic—an outlook that seems ill suited to meet the public-health challenge posed by the virus. And yet the region can also draw on a different, more communitarian legacy. Long a center of progressive government, it’s a place where, at least some of the time, officials are close to, rather than estranged from, poor and immigrant communities; during the H.I.V./AIDS crisis, it developed a robust public-health infrastructure capable of helping people who are at risk. If the Bay Area has weathered the pandemic with some success, that may be due to a unique and little-discussed factor: its past.
In 1866, the fledgling city of San Francisco decided to build a four-story poorhouse for unlucky gold rushers. For the site, planners chose a sixty-four-acre property that had once belonged to José de Noé, the last mayor to run the city as a Mexican territory; the plot sloped down to a deep lagoon—la laguna honda, in Spanish. Laguna Honda, as the poorhouse became known, was a place for city residents who were old, impoverished, mentally ill, and disabled. After the great earthquake of 1906, it expanded into a refuge for the homeless; by 1920, it had become a municipally run “relief home,” providing long-term care to the city’s elderly. Today, Laguna Honda is by most measures the biggest skilled-nursing facility, or S.N.F., in the United States. It usually houses seven hundred and twenty residents, who are cared for by seventeen hundred staff members—as many people as one might find in all the S.N.F.s in a midsize American city. In January, therefore, when San Francisco began preparing for the coronavirus, it did so with one remarkable advantage: an unusually large proportion of its nursing-home residents lived in a facility owned and operated by the San Francisco Department of Public Health.
Around the world, a high number of COVID-19 deaths have been associated with nursing homes. In the U.S., the figure is forty per cent. The danger the virus posed to elderly people, and especially to those in nursing homes, became undeniable in March, when thirty-nine residents of the Life Care Center, an S.N.F. in Kirkland, Washington, died from the virus. The center, which was part of a for-profit chain of two hundred facilities, was fined and sanctioned by state and federal authorities for its low standards of care. But its deficiencies—poor infection control, insufficient staffing, bureaucratic lethargy—are not unusual in the elder-care industry. (Life Care has appealed the sanctions.) “Every geriatrician knew what was coming,” Michael Wasserman, a prior president of the California Association of Long Term Care Medicine and the former C.E.O. of a large nursing-home chain, told me. “Every S.N.F. should have freaked out when they heard about Kirkland. They should have known two things: that they didn’t want to be another Kirkland, and maybe that this virus had a predilection for S.N.F.s.”
London Breed, the mayor of San Francisco, recalled how “everything we heard back in January was about China, and we have a big Chinese population in the city.” But by February, she said, “we were concerned about the nursing homes. I knew we had advantages with the department of health and the University of California, San Francisco, but we also had a lot of these vulnerable spots.” The city’s public-health officials watched the catastrophe in Kirkland and thought immediately of Laguna Honda. “It was at the top of our priority list,” Grant Colfax, the head of the city’s department of health, said. “Given the fragility of the residents there and the sheer size, we knew an outbreak there could be beyond tragic.”
Nursing homes vary widely in quality, and studies have found that a few factors combined can predict the level of care they provide. Size is important: facilities larger than a hundred beds tend to be harder to oversee and more prone to outbreaks. The percentage of patients on Medicaid is another indicator: because Medicaid reimburses care providers at lower rates than Medicare does, facilities that rely on it seek to cut costs. More than half of American nursing homes are owned by chains, and these tend to be worse than nonprofit, smaller for-profit, and government-run facilities. (Having been purchased by private-equity firms, many are under intense pressure to slash budgets.) But the most crucial determinant of quality is probably nurse staffing. Higher staffing levels, especially of registered nurses, or R.N.s, are consistently associated with significant improvements in care, while for lower levels the reverse is true. In 2001, the Centers for Medicare and Medicaid Services, the main regulatory body for S.N.F.s, suggested that facilities offer their residents a minimum of 0.75 “registered nurse hours per resident day”—that is, forty-five minutes of daily attention from a registered nurse. Studies have since shown that adequate care requires more time than this; meanwhile, in 2017, regulators found that seven in ten nursing homes had been overreporting their nurse-staffing levels. Industry-wide, understaffing is the norm—a reality that has had devastating consequences during the pandemic.
Residents are cared for not only by R.N.s but also by certified nursing assistants, or C.N.A.s. Minimally trained, C.N.A.s are sometimes called “butt-wipers” in the elder-care industry because of the unpleasant and unsanitary tasks they perform. Unlike R.N.s—who usually spend two to four years in school, can do everything from administrative management to infection control, and, in a typical California nursing home, earn about eighty-eight thousand dollars a year—the average Californian C.N.A. makes just over thirty thousand dollars a year; to make ends meet, many C.N.A.s have multiple jobs, often travelling between them over the course of a day. In Washington State, COVID-19 radiated out from Life Care Center through a network of low-paid C.N.A.s, many of whom worked at more than one facility; in Los Angeles, the virus also followed C.N.A.s from one poorly staffed S.N.F. to the next. Earlier this year, three economists used phone-tracking data to analyze the interconnections between nursing homes; they found that, even after visitor restrictions had been imposed, the average facility remained connected to fifteen others, probably through its staff. A facility’s degree of connection strongly predicted its number of COVID-19 cases: a nursing home that didn’t rely on shared staff could cut infections by forty-four per cent.
As the pandemic began, health officials in the Bay Area, like their peers elsewhere, confronted S.N.F.s that were too understaffed and poorly managed to mount a meaningful response. In early April, for example, officials in Contra Costa County—a mostly suburban part of San Francisco’s East Bay—started receiving e-mails from an S.N.F. there, the Orinda Care Center, about a dire staff shortage. Visiting to assess the situation, they found that the center’s staff didn’t know how to use P.P.E. correctly, and that one of its administrators, who said he was afraid that the supplies would be stolen, was stockpiling them in his garage at home. Orinda had chronic understaffing issues before the pandemic. Now, as staff started testing positive, the facility had to decide whether to send sick workers home, risking resident neglect, or let them come in despite their infections. Eventually, more than fifty staff members and residents became infected, and four people died. (The Orinda Care Center told The New Yorker that it had been in compliance with both P.P.E. and staffing regulations, and that those who had died had suffered from “complex and chronic underlying medical conditions.”)
Other parts of the Bay Area had better luck with S.N.F.s. Despite its large elderly population, Berkeley has only five small S.N.F.s and one long-term-care facility, all located in the historically poor and Black parts of town. Four of the S.N.F.s are for-profit. The fifth, Chaparral House, is a small nonprofit organization that charges upward of eleven thousand dollars a month for a semiprivate room, and so has the means to control infections and care for residents during the lockdown.
Berkeley is the only city in the Bay Area with its own public-health agency (it was established in 1880, to help fight off communicable diseases); the city’s chief public-health officer, Lisa Hernandez, was able to draw up early, customized plans in partnership with Berkeley’s handful of S.N.F.s. She and her team knew the facilities intimately, and the plans were tailored to account for their physical layouts and operations. Until May, none of the S.N.F.s in Berkeley had a single confirmed case of COVID-19—a remarkable result. (Since then, four of the facilities have reported outbreaks; it seems that even well-prepared S.N.F.s can hold on for only so long, especially in times of spiking community spread.) In accounting for the relative success of a city like Berkeley, it seems natural to look to the behavior of ordinary people, who might wear masks or congregate to varying degrees. But infection statistics are also shaped by the distribution and quality of S.N.F.s. If you want to predict how many infections a given area will have, you can start by asking a series of questions: How many nursing homes are there? How understaffed are they? Are they capable of working with local health officials to keep case counts low?
To answer such questions about San Francisco, you must start with Laguna Honda: a single facility, run by the city, housing enough vulnerable people to fill eight average for-profit S.N.F.s. Laguna Honda’s sheer size, in terms of both its patient and staff populations and its sprawling campus, should have been an added vulnerability, offering the virus hundreds of possible entryways. Managing its large staff had proved difficult even in ordinary times. In 2019, Laguna Honda’s C.E.O. stepped down after a horrific scandal involving C.N.A.s who physically and emotionally abused elderly and mentally disabled patients over a period of three years. Despite strict penalties from state and federal regulators, Laguna Honda failed a subsequent checkup and currently has a two-out-of-five-star rating from the official site for Medicare.
The campus itself evokes the history of the city: a pavilion containing a cafeteria, a rehab center, and three on-site morgues connects two skilled-nursing towers, which are of modern design, while dozens of outbuildings and a central administrative complex are in the Spanish Revival style, painted pink. Frescoes depict laborers building San Francisco; Art Deco light fixtures illuminate Mediterranean mosaics. Near the lagoon, there’s a petting zoo, with goats, ducks, and sheep, evoking a bygone California; it’s easy to imagine Bing Crosby performing at Laguna Honda, as he sometimes did in the nineteen-seventies. In June, however, when I first visited, all the outdoor spaces were empty. On the orders of the San Francisco health department, Laguna Honda had closed its doors to visitors and nonessential staff on March 6th—more than two weeks before the first COVID-19 death was recorded in the city.
Samantha Avecilla, a thirty-two-year-old registered nurse, had wanted to work at Laguna Honda for a long time before she was hired there. Like many R.N.s in the United States, Avecilla emigrated from the Philippines; eventually, she and her husband started a family in Martinez, a small city in Contra Costa County, northeast of San Francisco. Until last January, she split her hours between an S.N.F. in Oakland and home-care work, making forty dollars an hour in both positions. Staffing at her other jobs had been a revolving door—every week, there would be a new C.N.A. or R.N. on the floor—and she looked forward to working at a place where her colleagues rarely left, because there was nowhere better to go. When she was hired at Laguna Honda, she got a raise to sixty-five dollars an hour; as a city employee, she also started receiving a pension, a 401(k), and health benefits. She could now afford to work only one job.
In late March, Avecilla was working a shift in North 4—the fourth “neighborhood” in the facility’s north tower—when she heard that a staffer in another area, South 5, had tested positive for the coronavirus. Planning for this eventuality had started in January. Irin Blanco, one of Laguna Honda’s nursing directors, sent the staffer and her shift-mates home, and began a contact-tracing investigation; in total, twenty-two staffers and fifteen patients had been exposed. At the time, tests were difficult to come by, but Laguna Honda reached out to the city’s health department, which helped assemble enough test kits for the first circle of contacts. By March 26th, six staffers and one resident had tested positive. It was the beginning of a worrisome outbreak.
Blanco and Laguna Honda’s executive team quickly locked down the hospital’s side and back entrances, funnelling everyone through a single door, with a screening station. Meanwhile, in the San Francisco health department, a crisis team was formed specifically for Laguna Honda; Ayanna Bennett, the city’s director of health equity, visited the facility to review its protocols. Because the department had direct control, Bennett and her team were able to make changes that would have been fought by managers and families at many for-profit S.N.F.s. They halted admissions completely, locked doorways, and froze nurse rotations; they posted guards to prevent smokers from congregating in their usual spots. As Laguna Honda refined its protocols, Bennett and her colleagues began sharing them with other local nursing homes. “Some of the smaller places had a lot more resistance to doing things they felt they couldn’t do,” she told me. “They kept saying, ‘We can’t manage that; we can’t do that.’ ” Many struggled with understaffing. “If they’re already shorthanded all the time, that’s when you have people showing up sick.”
As infections spread, pandemic staffing became a system-wide problem. Staff members could be exposed or infected at any of the S.N.F.s at which they worked; all of those facilities then had to find a virus-free replacement. In a chronically understaffed industry, the labor pool was too small to handle emergencies. But Laguna Honda was adequately staffed, with well-paid nurses and C.N.A.s who didn’t need to work anywhere else, allowing management to effectively quarantine and cover shifts when the inevitable exposures took place. Avecilla was part of the team assigned to cover for those who’d been quarantined; she was transferred from North 4 to South 5, which now became Laguna Honda’s COVID-19 unit. She, in turn, had backup—nurses from the state and the Centers for Disease Control and Prevention arrived to help. The health department, meanwhile, focussed a healthy portion of its testing capacity on Laguna Honda, and by May the facility was testing two hundred people a day.
So far, only two Laguna Honda residents have died of COVID-19. “I did the math in my head,” Bennett said. A worst-case outbreak in a seven-hundred-and-twenty-bed facility could lead to “hundreds of people dying over a short period of time.” Such an outbreak at Laguna Honda could have turned San Francisco’s success on its head. The unusual existence of such a large city-run S.N.F. could be seen as a stroke of luck. Alternatively, it could be understood as the earned good fortune of a city that cared enough to build a refuge for indigent gold rushers in 1866, and then to continue to run and fund it for the next hundred and fifty-four years.
Diane Havlir, an infectious-disease specialist, came to the University of California, San Francisco, in 1984, at the height of the city’s H.I.V./AIDS crisis. In the years afterward, she became a global leader in the treatment of the disease, helping to secure care for millions of people in the United States, Africa, and other parts of the world. In March, as the American coronavirus pandemic ramped up in earnest, Havlir noticed something striking: about eighty per cent of the COVID-19 patients arriving at Zuckerberg San Francisco General Hospital were Latino, belying the association of the newly arrived virus with travel from China and Europe. “These weren’t people who were anxious about the virus and rushing to the hospital for peace of mind at the first sign of a symptom,” Carina Marquez, an assistant professor of medicine at U.C.S.F. and a physician at General, told me. “Almost everyone we saw during those early days was in really bad shape.” Havlir, Marquez, and other doctors at the hospital quickly grasped the implications: the virus was spreading rapidly in specific San Francisco communities.
Over the next few weeks, an effort coalesced, organized by Havlir, Marquez, and other doctors, community organizers, and translators from U.C.S.F. and a community group called the Latino Task Force. Their goal was to set up a series of testing sites in the Mission, a historically Latino neighborhood that has been gentrified by wealthy and mostly white tech workers; they hoped to convince a few thousand residents—many of them immigrants, some undocumented—to submit themselves to testing, tracing, and quarantine. As they prepared, they drew on a tightly interwoven network of public-health initiatives, including a city-run quarantine-hotel program and contact-tracing system, and a community-led drive for food and cleaning supplies. They partnered with community-outreach groups and, under the name Unidos en Salud (United in Health), reached out to Mission residents using bilingual canvassing, Spanish-language radio advertisements, and phone banking. Leaders from the Latino Task Force told friends and acquaintances that they planned to get tested themselves, and to submit to quarantine, if necessary.
By drawing on resources and expertise that predated the coronavirus pandemic, the U.C.S.F. team was able to act with remarkable speed. Over four days in April, the partnership performed some four thousand P.C.R. and antibody tests in the Mission. (The former detect current infections; the latter show if a person has been infected in the past.) Forty-one per cent of those tested identified as white, forty per cent identified as Latino, and nine per cent identified as Asian; there were smaller numbers of Black and mixed-race people. Over all, two per cent of the P.C.R. tests came back positive—a significant number. Ninety-five per cent of those who tested positive were Latino. Roughly nine in ten said that they couldn’t work from home, made less than fifty thousand dollars a year, and lived in a household of three or more people. “I was surprised when we got the results,” Marquez said. Though she had “suspected that the infection rates would be higher in the neighborhoods with the most essential workers,” she hadn’t anticipated racial disparities this stark. The effort proved what Ranu Dhillon, the community-health researcher, had seen anecdotally in the region’s hospitals: the virus was concentrated in the neighborhoods where nursing-home and hospital staffers, food-service and meat-processing employees, and other essential workers lived.
The Mission wasn’t the only at-risk neighborhood in San Francisco. In the course of the next two months, Havlir and her team set up testing sites in Hunters Point-Bayview, a traditionally Black neighborhood on the city’s south side, and Sunnydale, which is home to many of the city’s Pacific Islanders, in addition to having a large Black and Latino population. On June 1st, I drove to Sunnydale to see the site. A group of white tents was spread across a playground and baseball field, looking strikingly dystopian; my uneasiness was overpowered by a d.j. who had set up his speakers near a barbecue pit, where volunteers were serving food to those who’d come out to get tested.
Monique LeSarre, the doctor who directs the Rafiki Coalition for Health and Wellness, sat at a folding table at the entrance to the testing tent. In the nineteen-eighties and nineties, when Rafiki helped Black residents devastated by H.I.V. get care, the organization was known as the Black Coalition on AIDS (B.C.A.); as H.I.V. treatment improved, B.C.A. changed its name and broadened its services. “When I heard about the outbreaks in the Bronx and Detroit, I knew this was going to hit our Black neighborhoods,” LeSarre told me. Like Havlir, LeSarre lived through San Francisco’s H.I.V. epidemic and has connections and expertise from that period. To combat the stigma and general ignorance around H.I.V., community organizers like LeSarre, as well as politicians, health officials, and doctors of all kinds, built an infrastructure of trust and communication connecting the city’s government, universities, and marginalized communities. This network has been invaluable during the coronavirus pandemic. In places such as Texas’s Rio Grande Valley, where poor and minority communities are more alienated from their local governments, it has been nearly impossible for health officials to mount a humane and targeted intervention, and cases have surged.
In such interventions, the last mile is crucial. Kim Rhoads, a cancer researcher and associate professor of epidemiology and biostatistics at U.C.S.F., and Kevin Epps, a community-outreach coördinator and documentarian—he is known for his 2003 film “Straight Outta Hunters Point”—brought to Havlir’s team an intimate knowledge of the southern neighborhoods of San Francisco, as well as a wide range of contacts, built over decades, who could spread the word. “If you don’t have community buy-in, there’s always going to be a level of distrust that makes these things almost impossible to run,” Rhoads told me. Over the course of a week, Epps executed what he called a “guerrilla marketing campaign” in District 10, which includes Hunters Point, Bayview, and Sunnydale. “This was door-to-door, posters, putting it out on social media,” Epps said. “I even was out on the corner with a bullhorn.” Epps combatted rampant misinformation. “Early on, a lot of Black people had heard they couldn’t get the virus,” he said. “Even more didn’t want to sign up for testing because they heard Google was running it and thought all their information was going to get stolen.” (The testing sites run by Havlir and her colleagues used Chan-Zuckerberg BioHub/U.C.S.F. labs; Verily, a research organization that is operated by Google’s parent company, runs most of the testing in the Bay Area.) “They’ve all known me for years,” Epps concluded. “That got through some of the trust issues.”
By May, the testing-trace-and-isolate efforts mounted by U.C.S.F. and the Latino Task Force had tested 3,953 people, isolating many of them at a time when the first wave was at its most dangerous. The testing had side benefits: it included a world-class community-based communications effort, designed to inform non-English speakers, and spurred the creation of a new municipal program, Right to Recover, which offered wage replacement to workers who needed to stay home. During the same period, the city’s health department, working at Laguna Honda and other S.N.F.s, was able to take direct action to safeguard an unusually high proportion of the metro region’s most vulnerable elderly residents. All of this happened in a part of the country that had shut down early enough to prevent massive viral spread. But the shutdowns weren’t the end of the response; they were the beginning.
The testing effort drew a map of infection in San Francisco; that map illuminated not just the physical location of the virus but the socioeconomic dynamics of its dispersion. It showed that infections among people who could work from home were relatively rare. Instead, most transmission took place in poor neighborhoods, or in workplaces where people huddled together: nursing homes, restaurant kitchens, meatpacking plants. Differences in case numbers reflected the distribution of certain kinds of workers and workplaces. Berkeley had a low infection rate partly because it had done a good job with its S.N.F.s, but it also helped that there were so few of them. A city is more likely to beat the coronavirus if the indigent elderly who live in understaffed S.N.F.s and the poor essential workers who attend to them cannot afford to live there. Berkeley, in effect, had priced out the worst of the pandemic.
East Palo Alto, known locally as E.P.A., is a good example of a Bay Area city that has suffered during the pandemic without the shield of gentrification. Situated between the marshy south end of San Francisco Bay and the wealthy communities of Menlo Park and Palo Alto, it is home to about thirty thousand people; although it’s just a few miles from Stanford’s campus, no well-funded projects headed by superstar epidemiologists took place within its borders. Its history is deeply intertwined with the story of race in the Bay Area. In the nineteen-thirties and forties, Japanese-American farmers lived on its land; after they were placed in internment camps during the Second World War, Black domestic workers moved in, and for the next fifty years East Palo Alto was a Black city. In the sixties and seventies, Nairobi College, a radical training space for Black and Latino youth inspired by the Third World Liberation movement, was run out of private homes and meeting spaces there. East Palo Alto was devastated by the crack epidemic—in 1992, it was called “the murder capital” of America—and the Black population was eventually replaced almost entirely by working-class Latino immigrants. Latinos now make up more than sixty per cent of the city’s population. The majority are undocumented, and many are essential workers. E.P.A. residents live so close to Facebook’s headquarters, in Menlo Park, that some have taken to stealing the bikes that Facebook employees use to get around campus and tricking them out.
“We’re last in census response and last in voter turnout,” Walfred Solorzano, an East Palo Alto city clerk, told me. “And we’re first in coronavirus.” In early July, at a time when a surge in COVID cases had prompted Governor Gavin Newsom to shut large parts of the state back down, I went for a bike ride around East Palo Alto with Solorzano and a couple of his friends. We started off in Jack Farrell Park, where protesters had painted a mural of George Floyd on one of the concrete walls. The park was surrounded by single-story, single-family homes, most of them fenced off from the street and in disrepair. “Lots more people live in these houses than you’d think,” Solorzano said. “Families living in garages, stacked up inside the houses. So when one of them gets sick with corona, everyone’s gonna get it.”
We headed out toward the industrial flats, on the shores of the bay, passing the new office buildings that housed Amazon’s Web Services operation. On our right was the former site of Whiskey Gulch, once the city’s main street and night-life district; it was now a modern corporate fortress, with gated parking, a bottleneck entrance, and two large, beige buildings bookending a Four Seasons Hotel. Because of the lockdowns, the site was completely empty. “All of this is in E.P.A., but on their Web site they say it’s Silicon Valley. It’s like they’re ashamed to be associated with the city,” Solorzano said. “This space used to have the Nairobi Cultural Center here, and all these old businesses that meant a lot to the people who have been here a long time.”
As we biked, Solorzano summarized the state of testing in East Palo Alto. More Latinos had been getting tested, having been persuaded by word-of-mouth campaigns. Still, he said, many people, especially African-Americans, were hesitant. “I think they might feel this apathy,” he said. “They’ve just seen every promise get broken, and maybe they’re just giving up.” This hopelessness has extended into the winter. East Palo Alto’s test-positivity rates have hovered around fifteen to eighteen per cent, which is roughly five to six times higher than the surge that shut San Francisco back down this winter; at times, rates have reached thirty per cent.
After beginning its response with broad, early shutdowns, San Francisco and its surrounding cities and counties faced a challenge. How do you manage a pandemic in a city where some people can afford to tutor their kids in pods, have all their groceries delivered, and leave the city at a moment’s notice for houses in Tahoe, on the northern coast, or even in Hawaii, while others live in tent cities? Confronting the pandemic required the coöperation of young, relatively affluent residents who could work from home; it also demanded targeted interventions among low-income essential workers, immigrants, the poor, and people in nursing homes. San Francisco, perhaps more than any other city in the country, has particularly stark gaps between these groups.
“It’s not been easy,” Mayor Breed told me, when I asked her about the city’s response in Latino neighborhoods, where people fear that coöperating with health officials will ensnare them in an immigration dragnet. “There’s definitely a trust issue there. I can’t control ICE. I can only control what we’re able to do. We’ve made it clear to people, whether for the census or for testing, that we don’t share information with ICE and that we’re there to provide support, but that message takes a lot of work.” Kim Rhoads explained the other side of the problem: communicating the inequalities of the pandemic to white voters. “It’s so complicated,” she said. “We don’t want it to be called a Latinx disease here, so we have to be very careful about that.” She added, “You can’t just say that the Latinx community is contracting the virus without explaining exactly why. You have to say who is still going to work in food plants, who are the essential workers in kitchens and delivering groceries.” To describe the racial inequalities of the pandemic, she concluded, is also to risk “inciting the worst anti-Latinx sentiments.”
In the coronavirus’s winter wave, San Francisco and the Bay Area have not been spared. Infection rates, while still comparatively low, have spiked; for the first time since the start of the pandemic, I.C.U.s and emergency rooms are inching toward full capacity. But the approaches used in the spring and summer—hyper-localized targeting; an emphasis on the hardest-hit communities; the coördination of disparate resources in hospitals, clinics, academia, the community, and government—are just as relevant now. You still have to look for the virus. Having found it, you can suppress its spread; you can also, in the months to come, vaccinate at-risk populations.
After finishing up their studies in San Francisco, the U.C.S.F. team went across the bay to Alameda County. In an initiative led by Alicia Fernández, a physician and researcher, they set up testing sites in Deep East and West Oakland, both traditionally Black neighborhoods, and Fruitvale, a majority-Latino neighborhood that, despite being situated in the center of the relatively calm Bay Area, had some of the highest test-positivity rates in the country. In Fruitvale, the researchers determined that the epicenter of the outbreak wasn’t the Latino community in general but a group of Mayan workers from Guatemala who speak the indigenous Mam language. Rhoads and her colleagues believe that these sorts of specific insights, which are currently vital for testing, isolation, and education, will also be needed when it comes time to distribute vaccines. The U.C.S.F. team has learned to fine-tune its testing efforts; it has discovered, for instance, that, while scheduled-mass-testing sites work well within Latino and Asian neighborhoods, Black residents in Oakland are more likely to visit pop-up testing sites. “The same issues with testing are going to be present with the vaccine,” Rhoads said. “A lot of health-care institutions and even research universities have this vision that, if they provide it, people will come. That’s just never been true, and it’s certainly not going to start being true now. We need to meet people where they are.”
The architects of San Francisco’s coronavirus response may find a receptive ear in the Biden Administration: three U.C.S.F. faculty members, with a broad combined range of expertise, from biostatistics to emergency-room management, will serve on the President’s coronavirus advisory board. But it’s worth asking if any top-down effort can replicate the city’s success in a country that now averages more than two hundred thousand new confirmed cases per day. Even the best plans require basic infrastructure; in San Francisco, much of that infrastructure had been put in place long before the pandemic arrived. It is difficult, in a matter of months, to build bridges between minority populations and local governments, or to fix long-standing problems with understaffing in privately run nursing homes. Around the country, even well-intentioned test-and-trace programs have struggled to reach vulnerable people who, afraid of the government, do not want to share their names and contacts with a stranger on the phone. While Operation Warp Speed, the Trump Administration’s effort to produce and distribute a vaccine, has found success, no similar effort has been aimed at the protection of nursing homes, which, nine months into the pandemic, still account for forty per cent of COVID-19 deaths nationwide. (In individual states, the figure has sometimes risen to more than seventy per cent.)
Addressing these issues requires a redirection of our collective attention. Our society has a habit of ignoring essential workers and the indigent elderly; we readily scrutinize the behaviors of the able-bodied and the financially comfortable but look away from the settings where risk and suffering are placed on the most vulnerable. “The public has this blind spot about essential workers,” Rhoads said. “If you can shelter in place, and there’s no real threat to your family, you have no real impetus to look at the problem for what it is. Someone has to keep the essential work going so that other people can stay home and program algorithms for Google. And as long as you don’t acknowledge that the real problem is that some people get to stay home while others need to go in to dangerous jobs or basically starve, you don’t have to do anything to fix it. You don’t have to look the actual problem in the face, and the zero-sum game of capitalism can just keep rolling along.”
In November, Berkeley finally had a major outbreak. It didn’t start on campus, where students have been slowly returning, or at a restaurant or an underground party. It happened at Golden Gate Fields, a horse-racing track on the eastern edge of the San Francisco Bay. To get there from downtown, you need to drive all the way to I-80, through an underpass where hundreds of homeless people live in tents and R.V.s; as you wind around the track toward the parking lot, you can see dozens of small green shacks poking up above the fence. Roughly three hundred and seventy “horse people,” the vast majority of whom are Latino, live in rooms attached to these horse stalls. Since early November, more than three hundred workers at the site have tested positive for the coronavirus. They account for more than twenty per cent of all cases in the city of Berkeley since the start of the pandemic.
Over the past eleven months, as Berkeley’s infection rate has stayed remarkably low, many of the city’s residents have touted the power of personal responsibility: wearing masks, social distancing, ordering groceries via app. There’s no question that these measures make a difference. But as infection rates in the Bay Area climb during the winter surge, and I.C.U. beds finally start to fill up, the lesson to take from Berkeley, San Francisco, and many other successful cities isn’t that, by listening to science, an educated and careful population can stop the spread. It’s that, even within a city that has systematically priced out the most vulnerable populations, an outbreak can still happen. The virus will find vulnerable people where they live—even if it’s in barns on the outer edges of the city, even if almost nobody knows they exist.
More on the Coronavirus
- To protect American lives and revive the economy, Donald Trump and Jared Kushner should listen to Anthony Fauci rather than trash him.
- We should look to students to conceive of appropriate school-reopening plans. It is not too late to ask what they really want.
- A pregnant pediatrician on what children need during the crisis.
- Trump is helping tycoons who have donated to his reëlection campaign exploit the pandemic to maximize profits.
- Meet the high-finance mogul in charge of our economic recovery.
- The coronavirus is likely to reshape architecture. What kinds of space are we willing to live and work in now?
