Seattle’s Leaders Let Scientists Take the Lead. New York’s Did Not

Seattle’s Leaders Let Scientists Take the Lead. New York’s Did Not

The initial coronavirus outbreaks on the East and West Coasts emerged at roughly the same time. But the danger was communicated very differently.

By Charles Duhigg

April 26, 2020

The first diagnosis of the coronavirus in the United States occurred in mid-January, in a Seattle suburb not far from the hospital where Dr. Francis Riedo, an infectious-disease specialist, works. When he heard the patient’s details—a thirty-five-year-old man had walked into an urgent-care clinic with a cough and a slight fever, and told doctors that he’d just returned from Wuhan, China—Riedo said to himself, “It’s begun.”

For more than a week, Riedo had been e-mailing with a group of colleagues who included Seattle’s top doctor for public health and Washington State’s senior health officer, as well as hundreds of epidemiologists from around the country; many of them, like Riedo, had trained at the Centers for Disease Control and Prevention, in Atlanta, in a program known as the Epidemic Intelligence Service. Alumni of the E.I.S. are considered America’s shock troops in combatting disease outbreaks. The program has more than three thousand graduates, and many now work in state and local governments across the country. “It’s kind of like a secret society, but for saving people,” Riedo told me. “If you have a question, or need to understand the local politics somewhere, or need a hand during an outbreak—if you reach out to the E.I.S. network, they’ll drop everything to help.”

Riedo is the medical director for infectious disease at EvergreenHealth, a hospital in Kirkland, just east of Seattle. Upon learning of the first domestic diagnosis, he told his staff—from emergency-room nurses to receptionists—that, from then on, everything they said was just as important as what they did. One of the E.I.S.’s core principles is that a pandemic is a communications emergency as much as a medical crisis. Members of the public entering the hospital, Riedo told his staff, must be asked if they had travelled out of the country; if someone had respiratory trouble, staff needed to collect as much information as possible about the patient’s recent interactions with other people, including where they had taken place. You never know, Riedo explained, which chance encounter will shape a catastrophe. There are so many terrifying possibilities in a pandemic; information brings relief.

A national shortage of diagnostic kits for the new coronavirus meant that only people who had recently visited China were eligible for testing. Even as EvergreenHealth’s beds began filling with cases of flulike symptoms—including a patient from Life Care, a nursing home two miles away—the hospital’s doctors were unable to test them for the new disease, because none of the sufferers had been to China or been in contact with anyone who had. For nearly a month, as the hospital’s patients complained of aches, fevers, and breathing problems—and exhibited symptoms associated with covid-19, such as “glassy” patches in X-rays of their lungs—none of them were evaluated for the disease. Riedo wanted to start warning people that evidence of an outbreak was growing, but he had only suspicions, not facts.

At the end of February, the C.D.C. began allowing the testing of patients with unexplained respiratory-tract infections or “fever and/or symptoms of acute respiratory illness.” Riedo called a friend—an E.I.S. alum at the local department of health. If he sent her swabs from two patients who had needed ventilators but had tested negative for influenza and other common respiratory diseases, would she test them for covid-19? At that point, there had been only sixteen detections of the coronavirus in the U.S., and only the one in Washington State. “I can’t remember why we picked those two patients,” Riedo told me. “I was sure they’d be negative. But we thought it would be good to start collecting data, and it was a way to make sure the testing lab was working.” The health official told him to send the samples to her lab.

Riedo remembered that other local researchers had been conducting a project called the Seattle Flu Study. For months, they had collected nasal swabs from volunteers, to better understand how influenza spread through the community. During the previous few weeks, the researchers, in quiet violation of C.D.C. guidance, had jury-rigged a coronavirus test in their lab and had started using it on their samples. They had just found a positive hit: a high-school student in a suburb twenty-eight miles from Seattle, with no recent history of foreign travel and no known interactions with anyone from China. The boy wasn’t seriously ill; if the researchers hadn’t done the test, the infection probably never would have been detected. The genetic sequence of the boy’s virus was unnervingly similar to that of the man with the first known case, even though the researchers couldn’t find any connections between them. The frightening implication was that the coronavirus was already so widespread that contagion was passing invisibly among community members.

At seven-forty that evening, Riedo got a call from his friend at the public-health lab. Both of the samples he had sent were positive. Riedo sent over swabs from nine other EvergreenHealth patients. Eight were positive. Riedo grabbed the patients’ charts and saw that seven of them had come from the Life Care nursing home. It didn’t make any sense: nursing-home residents don’t travel, and interact mainly with just family members and staff.

Riedo sent in more samples. Most of the patients tested positive, including a woman who had been told that she had pneumonia, another woman who had complained of sweating and clammy hands, and a man in his fifties with serious respiratory problems. For three days, dozens of that man’s family members had sat at his bedside in the hospital, coming in and out of the building and going from home to work, visiting restaurants and shaking people’s hands, inadvertently exposing themselves and others to covid-19.

At that moment, there were no known U.S. coronavirus fatalities. Schools, restaurants, and workplaces were open. Stock markets were near all-time highs. But when Riedo stopped to calculate how many of his hospital employees had been exposed to the coronavirus he had to quit when his list surpassed two hundred people. “If we sent all of those workers home for two weeks, which is what the C.D.C. was recommending, we’d have to shut down the entire hospital,” he told me. He felt like a man who, having casually swatted at a buzzing insect, suddenly realized that he was beneath a beehive.

The next day, the man with all the family visitors died. It was America’s first known covid-19 death. Riedo called his wife. “I told her I didn’t know when I would be coming home,” he said to me. “And then I started e-mailing everyone I knew to say we were past containment. It had already escaped.”

Epidemiology is a science of possibilities and persuasion, not of certainties or hard proof. “Being approximately right most of the time is better than being precisely right occasionally,” the Scottish epidemiologist John Cowden wrote, in 2010. “You can only be sure when to act in retrospect.” Epidemiologists must persuade people to upend their lives—to forgo travel and socializing, to submit themselves to blood draws and immunization shots—even when there’s scant evidence that they’re directly at risk.

Epidemiologists also must learn how to maintain their persuasiveness even as their advice shifts. The recommendations that public-health professionals make at the beginning of an emergency—there’s no need to wear masks; children can’t become seriously ill—often change as hypotheses are disproved, new experiments occur, and a virus mutates. The C.D.C.’s Field Epidemiology Manual, which devotes an entire chapter to communication during a health emergency, indicates that there should be a lead spokesperson whom the public gets to know—familiarity breeds trust. The spokesperson should have a “Single Overriding Health Communication Objective, or sohco (pronounced sock-O),” which should be repeated at the beginning and the end of any communication with the public. After the opening sohco, the spokesperson should “acknowledge concerns and express understanding of how those affected by the illnesses or injuries are probably feeling.” Such a gesture of empathy establishes common ground with scared and dubious citizens—who, because of their mistrust, can be at the highest risk for transmission. The spokesperson should make special efforts to explain both what is known and what is unknown. Transparency is essential, the field manual says, and officials must “not over-reassure or overpromise.”

The lead spokesperson should be a scientist. Dr. Richard Besser, a former acting C.D.C. director and an E.I.S. alumnus, explained to me, “If you have a politician on the stage, there’s a very real risk that half the nation is going to do the opposite of what they say.” During the H1N1 outbreak of 2009—which caused some twelve thousand American deaths, infections in every state, and seven hundred school closings—Besser and his successor at the C.D.C., Dr. Tom Frieden, gave more than a hundred press briefings. President Barack Obama spoke publicly about the outbreak only a few times, and generally limited himself to telling people to heed scientific experts and promising not to let politics distort the government’s response. “The Bush Administration did a good job of creating the infrastructure so that we can respond,” Obama said at the start of the pandemic, and then echoed the sohco by urging families, “Wash your hands when you shake hands. Cover your mouth when you cough. I know it sounds trivial, but it makes a huge difference. ”At no time did Obama recommend particular medical treatments, nor did he forecast specifics about when the pandemic would end.

Whereas the C.D.C. protocol encourages politicians to practice restraint, it invites the lead scientific spokesperson to demonstrate his or her advice ostentatiously, and to be a living example of the importance of, say, wearing a mask or getting a shot. When polio inoculations began, in the nineteen-fifties, many people worried that they were unsafe, so New York City’s commissioner of health—who happened to be married to the E.I.S.’s founder—invited reporters to watch schoolchildren getting injections. She also enlisted Elvis to publicly get his shot.

E.I.S. personnel in the field have carried boxes of masks and gloves to distribute to pilots, flight attendants, journalists, and health workers—supplies that may not be needed by the recipients but emphasize how important universal compliance is. When Besser gave briefings during the H1N1 pandemic, he sometimes started by describing how he had recently soaped up his fingers, or pointedly waited until everyone was away from the microphone before taking the stage. At the time, there was almost no chance that Besser and his colleagues were at immediate risk of contracting H1N1. “To maintain trust, you have to be as honest as possible, and make damn sure that everyone walks the walk,” Besser told me. “If we order people to wear masks, then every C.D.C. official must wear a mask in public. If we order hand washing, then we let the cameras see us washing our hands. We’re trying to do something nearly impossible, which is get people to take an outbreak seriously when, for most Americans, they don’t know anyone who’s sick and, if the plan works, they’ll never meet anyone who’s sick.”

Public-health officials say that American culture poses special challenges. Our freedoms to assemble, to speak our minds, to ignore good advice, and to second-guess authority can facilitate the spread of a virus. “We’re not China—we can’t order people to stay inside,” Besser said. “Democracy is a great thing, but it means, for something like covid-19, we have to persuade people to coöperate if we want to save their lives.”

On February 28th, around the time that Riedo learned of the covid-19 cluster at the Life Care nursing home, the news was also relayed to another E.I.S. alum, Dr. Jeff Duchin, the top public-health physician for Seattle and surrounding King County. To Duchin, the cluster suggested that there was already an area-wide outbreak. He told Dow Constantine, the King County Executive, that it was time to start considering restrictions on public gatherings and telling residents to stay home. This advice struck Constantine as possibly crazy. There were only two dozen covid-19 diagnoses in the entire nation. Life looked normal. How could people be persuaded to stop going to bars, much less to work, just because a handful of old people were sick?

Constantine told me, “Jeff recognized what he was asking for was impractical. He said if we advised social distancing right away there would be zero acceptance. And so the question was: What can we say today so that people will be ready to hear what we need to say tomorrow?” In e-mails and phone calls, the men began playing a game: What was the most extreme advice they could give that people wouldn’t scoff at? Considering what would likely be happening four days from then, what would they regret not having said?

Even for public-health professionals, the trade-offs were painful to contemplate. At a meeting of public-health supervisors and E.I.S. officials in Seattle, an analyst became emotional when describing the likely consequences of shutting Seattle’s schools. Thousands of kids relied on schools for breakfast and lunch, or received medicine like insulin from school nurses. If schools closed, some of those students would likely go hungry; others might get sick, or even die. Everyone also knew that, if the city shut down, domestic-violence incidents would rise. And what about the medical providers who would have to stop working, because they had to stay home with young kids? “It was overwhelming,” one E.I.S. official told me. “Every single decision had a million ripples.”

Yet the burdens caused by closing the schools could make an enormous difference in curtailing the spread of the virus: all kinds of parents would have to stay home. In 2019, Seattle had closed schools for five days after a series of snowstorms. Afterward, the Seattle Flu Study discovered that traffic in some areas had nearly disappeared, public-transit use had tumbled, and the transmission of influenza had dropped.

Constantine thought that announcing school closings was a potent communication strategy for reaching even people who weren’t parents, because it forced the community to see the coronavirus crisis in a different light. “We’re accustomed to schools closing when something really serious happens,” Constantine told me. “It was a way to speed up people’s perceptions—to send a message they could understand.”

While the logistics of classroom closures were being worked out, Constantine contacted Brad Smith, the president of Microsoft—which is headquartered in Redmond, east of Seattle—and asked him to consider ordering employees to work from home. “Microsoft is a big deal here,” Constantine told me. “I thought if they told everyone to stay home it could shift how the state was thinking—make the pandemic real.” Microsoft, as a tech company, was poised to switch quickly to remote work, and could demonstrate to other businesses that the transition could occur smoothly. On March 4th, with only twelve known covid-19 fatalities across the nation and no diagnoses among Microsoft workers, the company told employees to stay home if they could. Smith told me, “King County has a strong reputation for excellent public-health experts, and the worst thing we could have done is substitute our judgment for the expertise of people who have devoted their lives to serving the public.” 

Amazon, which is also headquartered in the area, told many of its local employees to work from home as well. “That’s a hundred thousand people suddenly staying home,” one Seattle resident told me. “From commute traffic alone, you knew something big had happened.”

On February 29th, Constantine held a press conference. He had asked Riedo, Duchin, and Kathy Lofy—another E.I.S. alum and the state’s top health officer—to play prominent roles. Duchin spoke first, and it was as if he had prepared his remarks with the Field Epidemiology Manual in hand. “I want to just start by expressing our deep and sincere condolences to the family members and loved ones of the person who died,” he said. He explained what scientists knew and did not know about the coronavirus, and noted, “We’re in the beginning stages of our investigation, and new details and information will emerge over the next days and weeks.” He predicted that “telecommuting” was likely to become mandatory for many residents, and repeated several times an easy-to-remember sohco: “more hand washing, less face touching.” Duchin told me that his words had been chosen carefully: “You have to think about managing the public’s emotions, perceptions, trust. You have to bring them along the path with you.” Since then, Washington State politicians have largely ceded health communications to the scientists, making them unlikely celebrities. “Hey people!! Jeff Duchin is the real deal,” one fan tweeted. A newspaper hailed him as “a bespectacled, calming presence.”

Constantine told me that he understands why politicians “want to be front and center and take the credit.” And he noted that Seattle has many of “the same problems here you see in Congress, with the partisanship and toxicity.” But, he said, “everyone, Republicans and Democrats, came together behind one message and agreed to let the scientists take the lead.”

By the time Seattle’s schools were formally closed, on March 11th, students and teachers were already abandoning their classrooms. The messaging had worked: parents were voluntarily keeping their kids home. Cell-phone tracking data showed that, in the preceding week, the number of people going to work had dropped by a quarter. Within days, even before Washington’s governor, Jay Inslee, issued official work-from-home orders, almost half of Seattle’s workers were voluntarily staying away from their offices. When bars and restaurants were officially closed, on March 15th, many of them were already empty. Constantine himself had been working from home for a week. He was giving interviews all day, and always underscored to reporters that he was speaking from his bedroom, and that the noises in the background were coming from his children, who were home from school. After he heard that the county’s basketball courts were still being heavily used, he ordered them closed.

The county had bought a motel to house homeless residents who tested positive for the coronavirus. When one homeless man at the motel, who was asymptomatic, left to buy a beer, Constantine immediately went to court, so that police could arrest him the next time he went out. The man’s actions had posed little risk: he had gone to a gas station across the street, then returned. But, Constantine told me, “the fact is some people are not going to follow the rules—and we need to show everyone there are consequences.”

Today, Washington State has less than two per cent of coronavirus cases in the U.S. At EvergreenHealth, hospital administrators have stopped daily crisis meetings, because the rate of incoming patients has slowed. They have empty beds and extra ventilators. The administrators remain worried, but are cautiously optimistic. “It feels like we might have stopped the tsunami before it hit,” Riedo told me. “I don’t want to tempt fate, but it seems like it’s working. Which is what makes it so much harder when I look at places like New York.”

The Epidemic Intelligence Service was founded in 1951, when American troops in Korea began experiencing fevers, aches, vomiting, and fatal hemorrhages. Some three thousand soldiers fell ill, leading military leaders to conclude that Chinese-backed Communists had weaponized bacteria. “The planning of appropriate defensive measures must not be delayed,” an epidemiologist at a new federal agency, the Communicable Disease Center, declared. He proposed a new division, named to evoke the Central Intelligence Agency. But when the first class of E.I.S. officers landed in Korea they found that the fevers were not caused by a crafty enemy. Soldiers, it turned out, had been accidentally consuming rodent feces. In later conflicts, generals were instructed to use thicker food-storage bags and to set more rat traps.

E.I.S. officers became known as “disease detectives.” In 1952, one of them studied a group of children in a Chicago slum who had all developed similar symptoms—muscle weakness, spasms, joint pain—but had tested negative for likely diseases. When the E.I.S. officer visited one of the children’s homes, he noticed a toddler chewing on chips of paint that had flaked off a windowsill. The paint chips were soft because they contained lead, which is toxic. A year later, that E.I.S. officer helped found the country’s first poison-control program, which taught parents that the first principle of safety was communication. The program advised parents to tell their children not to put paint chips in their mouths, and to signal the dangers of bleach, insecticide, and cleaning chemicals by storing them on high shelves.

E.I.S. alumni went on to take powerful health-care jobs across the country. “Nearly ninety per cent of E.I.S. graduates embark on public-health careers at the local, state, federal or international level,” a 2001 study found. Four former C.D.C. directors are E.I.S. alumni; half a dozen graduates have served as the U.S. Surgeon General.

When the coronavirus pandemic started, E.I.S. alumni began working non-stop, with some setting up cots inside their offices. While the virus remained overseas, the C.D.C. led communications, scrupulously following E.I.S. protocols. But soon after the coronavirus landed on American shores the White House took over. E.I.S. officers were dismayed to see the communication principles that the C.D.C. had honed over the years being disregarded, and sometimes turned on their head. A Coronavirus Task Force, led by Vice-President Mike Pence, was formed, excluding everyone from the C.D.C. except its director, Dr. Robert Redfield. “The C.D.C. was ordered into lockdown,” a former senior official at the agency told me. “They can’t speak to the media. These are people who have trained their entire lives for epidemics—the finest public-health army in history—and they’ve been told to shut up!”

Since then, the primary spokesperson during the pandemic has been not a scientist but President Donald Trump—a politician notoriously hostile to science. Further complicating matters, Trump has highlighted a rotating cast of supporting characters, including Pence; Dr. Anthony Fauci, from the National Institutes of Health; Dr. Deborah Birx, from the State Department; and the President’s son-in-law, Jared Kushner. “When there are so many different figures, it can cause real confusion about whom to listen to, or who’s in charge of what,” Dr. Tom Inglesby, the director of the Center for Health Security, at Johns Hopkins, said. “And, if the response becomes political, it’s a disaster, because people won’t know if you are making recommendations based on science or politics, and so there’s the risk they’ll start to tune out.”

Already, it’s clear that some confusion has taken hold. Though the C.D.C. formally recommended, in mid-March, that Americans practice social distancing, governors in five states have refused to order residents to stay home. (One of those states, South Dakota, is now contending with a major outbreak.) Federal leaders have given shifting advice—initially, Americans were told that they did not need to wear masks in public, but on April 3rd, at a White House press briefing, masks were recommended—and this has risked undermining public confidence. Trump announced the change by saying, “You don’t have to do it. I’m choosing not to do it.” Had the C.D.C. been in charge of communicating about masks, the agency surely would have used the change in guidance as a teaching opportunity, explaining that scientists had come to understand that people infected with the coronavirus can be contagious but asymptomatic for longer than originally thought—which means that we need to be more careful when we cough, even if we feel healthy or just have seasonal allergies. 

Trump’s daily briefings, however, are chaotic and contradictory. Within the span of a few days, Trump threatened to quarantine New York City, then reversed himself; soon after declaring that he intended to “reopen” the U.S. economy within two weeks, he called for thirty additional days of social distancing. Such inconstancy from a leader is distracting in the best of times. It is dangerous in a pandemic. “Right now, everyone is so confused by all the conflicting messages that, each time the guidance evolves, fewer and fewer people might follow it,” Besser, the former C.D.C. director, said. “We’re going backward in our sophistication.”

Morale at the C.D.C. has plummeted. “For all the responses that I was involved in, there was always this feeling of camaraderie, that you were part of something bigger than yourself,” another former high-ranking C.D.C. official told me. “Now everyone I talk to is so dispirited. They’re working sixteen-hour days, but they feel ignored. I’ve never seen so many people so frustrated and upset and sad. We could have saved so many more lives. We have the best public-health agency in the world, and we know how to persuade people to do what they need to do. Instead, we’re ignoring everything we’ve learned over the last century.”

The initial coronavirus outbreaks in New York City emerged at roughly the same time as those in Seattle. But the cities’ experiences with the disease have markedly differed. By the second week of April, Washington State had roughly one recorded fatality per fourteen thousand residents. New York’s rate of death was nearly six times higher.

There are many explanations for this divergence. New York is denser than Seattle and relies more heavily on public transportation, which forces commuters into close contact. In Seattle, efforts at social distancing may have been aided by local attitudes—newcomers are warned of the Seattle Freeze, which one local columnist compared to the popular girl in high school who “always smiles and says hello” but “doesn’t know your name and doesn’t care to.” New Yorkers are in your face, whether you like it or not. (“Stand back at least six feet, playa,” a sign in the window of a Bronx bodega cautioned. “covid-19 is some real shit!”) New York also has more poverty and inequality than Seattle, and more international travellers. Moreover, as Mike Famulare, a senior research scientist at the Institute for Disease Modeling, put it to me, “There’s always some element of good luck and bad luck in a pandemic.”

It’s also true, however, that the cities’ leaders acted and communicated very differently in the early stages of the pandemic. Seattle’s leaders moved fast to persuade people to stay home and follow the scientists’ advice; New York’s leaders, despite having a highly esteemed public-health department, moved more slowly, offered more muddied messages, and let politicians’ voices dominate.

New York’s mayor, Bill de Blasio, has long had a fraught relationship with the city’s Department of Health and Mental Hygiene, which, though technically under his control, seeks to function independently and avoid political fights. “There’s always a bit of a split between the political appointees, whose jobs are to make a mayor look good, and public-health professionals, who sometimes have to make unpopular recommendations,” a former head of the Department of Health told me. “But, with the de Blasio people, that antagonism is ten times worse. They are so much more impossible to work with than other administrations.” 

In 2015, when Legionnaires’ disease sickened at least a hundred and thirty New Yorkers and killed at least twelve, tensions between de Blasio and the Health Department came to a head. After de Blasio ordered health officials to force their way into buildings in the Bronx to test cooling towers for contamination, even though the outbreak’s source had already been identified, the officials complained that the Mayor was wasting their time in order to brag to reporters that he’d done everything possible to stamp out the disease. When the deputy commissioner for environmental health, Daniel Kass, refused City Hall’s demands, one of the city’s deputy mayors urged the commissioner of health, Mary Bassett, to fire Kass. She ignored the suggestion, but Kass eventually resigned. He later told colleagues he felt that his rebellion had made cooperation with City Hall impossible.

“Dan Kass is one of the best environmental-health experts in the country,” Bassett, who now teaches at Harvard, said. “New York has one of the best health departments in the United States, possibly the world. We’d all be better off if we were listening really closely to them right now.”

In early March, as Dow Constantine was asking Microsoft to close its offices and putting scientists in front of news cameras, de Blasio and New York’s governor, Andrew Cuomo, were giving speeches that deemphasized the risks of the pandemic, even as the city was announcing its first official cases. De Blasio initially voiced caution, saying that “no one should take the coronavirus situation lightly,” but soon told residents to keep helping the city’s economy. “Go on with your lives + get out on the town despite Coronavirus,” he tweeted on March 2nd—one day after the first covid-19 diagnosis in New York. He urged people to see a movie at Lincoln Center. On the day that Seattle schools closed, de Blasio said at a press conference that “if you are not sick, if you are not in the vulnerable category, you should be going about your life.” Cuomo, meanwhile, had told reporters that “we should relax.” He said that most infected people would recover with few problems, adding, “We don’t even think it’s going to be as bad as it was in other countries.”

De Blasio’s and Cuomo’s instincts are understandable. A political leader’s job, in most situations, is to ease citizens’ fears and buoy the economy. During a pandemic, however, all those imperatives are reversed: a politician’s job is to inflame our paranoia, because waiting until we can see the danger means holding off until it’s too late. The city’s epidemiologists were horrified by the comforting messages that de Blasio and Cuomo kept giving. Jeffrey Shaman, a disease modeller at Columbia, said, “All you had to do was look at the West Coast, and you knew it was coming for us. That’s why Seattle and San Francisco and Portland were shutting things down.” But New York “dithered instead of telling people to stay home.”

By early March, the city’s Department of Health had sent the Mayor numerous proposals on fighting the virus’s spread. Since there weren’t enough diagnostic kits to conduct extensive testing, public-health officials proposed “sentinel surveillance”: asking local hospitals to provide the Department of Health with swabs collected from people who had flulike symptoms and had tested negative for influenza. By testing a selection of those swabs, the department could estimate how rapidly and widely the coronavirus was moving through the city. In previous outbreaks, such studies had been tremendously useful in guiding governmental responses—and this spring Los Angeles effectively deployed the strategy, as did Santa Clara County, in California, and the state of Hawaii.

In New York City, the Health Department began collecting swabs, but the initiative met swift resistance. Under federal health laws, such swabs have to be anonymized for patients who haven’t consented to a coronavirus test. This meant that, even if city officials learned that many people were infected, officials wouldn’t be able to identify, let alone warn, any of them. The Mayor’s office refused to authorize testing the swabs. “They didn’t want to have to say, ‘There are hundreds, maybe thousands, of you who are positive for coronavirus, but we don’t know who,’ ” a Department of Health official told me, adding, “It was a real opportunity to communicate to New Yorkers that this is serious—you have to stay home.” The effort was blocked over fears that it might create a panic, but such alarm might have proved useful. After all, the official told me, panic is pretty effective at getting people to change their behavior. Instead, the Mayor’s office informed the Health Department that the city would sponsor a job fair to find a few new “disease detectives.” That event was held on March 12th, in Long Island City. The Department of Health official said, “We’re in the middle of a catastrophe, and their solution is to make us waste time interviewing and onboarding people!” (The Mayor’s office eventually relented on the sentinel-surveillance samples, and testing began on March 23rd—almost a month after samples were first collected. By then, the outbreak was well under way.)

As New York City schools, bars, and restaurants remained open, relations between the Department of Health and City Hall devolved. Health supervisors were “very, very angry,” one official told me. In particular, health officials were furious that de Blasio kept telling New Yorkers to go out and get a test if they suspected they were infected. On March 4th, he tweeted, “If you feel flu-like symptoms (fever, cough and shortness of breath), and recently traveled to an area affected by coronavirus . . . go to your doctor.” This was the opposite of what city health supervisors were advising: people needed to stay inside and call their doctor if they felt sick. Making trips to doctors’ offices or emergency rooms only increased the odds that the virus would spread, and the city’s limited supply of tests needed to be saved for people with life-threatening conditions. De Blasio’s staff, however, had started micromanaging the department’s communications, including on Twitter. Finally, on March 15th, the Department of Health was allowed to post a thread: “If you are sick, stay home. If you do not feel better in 3 to 4 days, consult with your health care provider”; “Testing should only be used for people who need to be hospitalized”; “Everyone in NYC should act as if they have been exposed to coronavirus. . . . New Yorkers who are not sick should also stay home as much as possible.” One City Council member told me that health officials “had been trying to say that publicly for weeks, but this mayor refuses to trust the experts—it’s mind-boggling.”

As the city’s scientists offered plans for more aggressive action and provided data showing that time was running out, the Mayor’s staff responded that the health officials were politically naïve. At one point, Dr. Marcelle Layton, the city’s assistant commissioner of communicable diseases, and an E.I.S. alum who is revered by health officials across the nation for her inventiveness and dedication, was ordered to City Hall, in case she was needed to help the Mayor answer questions from the press. She sat on a bench in a hallway for three hours, away from her team, while politicians spoke to the media. (Layton declined interview requests.) At press conferences, Layton and other physicians played minimal roles while de Blasio and Cuomo, longtime rivals, each attempted to take center stage. The two men even began publicly feuding—arguing in the press, and through aides, about who had authority over schools and workplace closures.

Eventually, three of the top leaders of the city’s Department of Health met with de Blasio and demanded that he quickly instate social-distancing rules and begin sending clear messages to the public to stay indoors. Layton and a deputy health commissioner, Dr. Demetre Daskalakis, indicated to de Blasio’s staff that if the Mayor didn’t act promptly they would resign. (The next day, Layton’s staff greeted her with applause, and at least one employee offered to give her some money if she had to make good on the ultimatum.) De Blasio was in a corner: he had long positioned himself as a champion of the underclass, and closing schools would disproportionately hurt the poor and vulnerable. What’s more, unions representing health-care workers had threatened that nurses, orderlies, and others might stay home unless there was a plan to provide child care.

Nevertheless, de Blasio finally acceded to the health officials’ demands. On March 16th, after a compromise was reached with the health-care unions, city schools were closed, and Cuomo ordered all gyms and similar facilities to shut down. The messaging remained jumbled, however. Right before the gym closure was set to take effect, de Blasio asked his driver to take him to the Y.M.C.A. in Park Slope, near his old home, for a final workout. Even de Blasio’s allies were outraged. A former adviser tweeted, “The mayor’s actions today are inexcusable and reckless.” Another former consultant tweeted that the gym visit was “Pathetic. Self-involved. Inexcusable.”

De Blasio and Cuomo kept bickering. On March 17th, de Blasio told residents to “be prepared right now for the possibility of a shelter-in-place order.” The same day, Cuomo told a reporter, “There’s not going to be any ‘you must stay in your house’ rule.” Cuomo’s staff quietly told reporters that de Blasio was acting “psychotic.” Three days later, though, Cuomo announced an executive order putting the state on “pause”—which was essentially indistinguishable from stay-at-home orders issued by cities in Washington State, California, and elsewhere. (A spokesperson for de Blasio said that City Hall’s “messaging changed as the situation and the science changed” and that there was “no dithering.” A spokesperson for Cuomo said that “the Governor communicated clearly the seriousness of this pandemic” and that “the Governor has been laser focused on communicating his actions in a way that doesn’t scare people.”)

To a certain extent, de Blasio’s and Cuomo’s tortured delays make sense. Good politicians should worry about poor children missing school just as much as they worry about the threat of an emerging disease. “That’s why E.I.S. training is so important,” Sonja Rasmussen, a former C.D.C. official, told me. In a pandemic, “the old ways of thinking get flipped around.” She added, “You have to make the kinds of choices that, if you aren’t trained for them, are really hard to make. And there’s no time to learn from your mistakes.”

Today, New York City has the same social-distancing policies and business-closure rules as Seattle. But because New York’s recommendations came later than Seattle’s—and because communication was less consistent—it took longer to influence how people behaved. According to data collected by Google from cell phones, nearly a quarter of Seattleites were avoiding their workplaces by March 6th. In New York City, another week passed until an equivalent percentage did the same. Tom Frieden, the former C.D.C. director, has estimated that, if New York had started implementing stay-at-home orders ten days earlier than it did, it might have reduced covid-19 deaths by fifty to eighty per cent. Another former New York City health commissioner told me that “de Blasio was just horrible,” adding, “Maybe it was unintentional, maybe it was his arrogance. But, if you tell people to stay home and then you go to the gym, you can’t really be surprised when people keep going outside.”

More than fifteen thousand people in New York are believed to have died from covid-19. Last week in Washington State, the estimate was fewer than seven hundred people. New Yorkers now hear constant ambulance sirens, which remind them of the invisible viral threat; residents are currently staying home at even higher rates than in Seattle. And de Blasio and Cuomo—even as they continue to squabble over, say, who gets to reopen schools—have become more forceful in their warnings. Rasmussen said, “It seems silly, but all these rules and sohcos and telling people again and again to wash their hands—they make a huge difference. That’s why we study it and teach it.” She continued, “It’s really easy, with the best of intentions, to say the wrong thing or send the wrong message. And then more people die.”